1 @chapter Filtering Introduction
2 @c man begin FILTERING INTRODUCTION
4 Filtering in FFmpeg is enabled through the libavfilter library.
6 In libavfilter, a filter can have multiple inputs and multiple
8 To illustrate the sorts of things that are possible, we consider the
12 input --> split ---------------------> overlay --> output
15 +-----> crop --> vflip -------+
18 This filtergraph splits the input stream in two streams, sends one
19 stream through the crop filter and the vflip filter before merging it
20 back with the other stream by overlaying it on top. You can use the
21 following command to achieve this:
24 ffmpeg -i INPUT -vf "split [main][tmp]; [tmp] crop=iw:ih/2:0:0, vflip [flip]; [main][flip] overlay=0:H/2" OUTPUT
27 The result will be that in output the top half of the video is mirrored
30 Filters in the same linear chain are separated by commas, and distinct
31 linear chains of filters are separated by semicolons. In our example,
32 @var{crop,vflip} are in one linear chain, @var{split} and
33 @var{overlay} are separately in another. The points where the linear
34 chains join are labelled by names enclosed in square brackets. In the
35 example, the split filter generates two outputs that are associated to
36 the labels @var{[main]} and @var{[tmp]}.
38 The stream sent to the second output of @var{split}, labelled as
39 @var{[tmp]}, is processed through the @var{crop} filter, which crops
40 away the lower half part of the video, and then vertically flipped. The
41 @var{overlay} filter takes in input the first unchanged output of the
42 split filter (which was labelled as @var{[main]}), and overlay on its
43 lower half the output generated by the @var{crop,vflip} filterchain.
45 Some filters take in input a list of parameters: they are specified
46 after the filter name and an equal sign, and are separated from each other
49 There exist so-called @var{source filters} that do not have an
50 audio/video input, and @var{sink filters} that will not have audio/video
53 @c man end FILTERING INTRODUCTION
56 @c man begin GRAPH2DOT
58 The @file{graph2dot} program included in the FFmpeg @file{tools}
59 directory can be used to parse a filtergraph description and issue a
60 corresponding textual representation in the dot language.
67 to see how to use @file{graph2dot}.
69 You can then pass the dot description to the @file{dot} program (from
70 the graphviz suite of programs) and obtain a graphical representation
73 For example the sequence of commands:
75 echo @var{GRAPH_DESCRIPTION} | \
76 tools/graph2dot -o graph.tmp && \
77 dot -Tpng graph.tmp -o graph.png && \
81 can be used to create and display an image representing the graph
82 described by the @var{GRAPH_DESCRIPTION} string. Note that this string must be
83 a complete self-contained graph, with its inputs and outputs explicitly defined.
84 For example if your command line is of the form:
86 ffmpeg -i infile -vf scale=640:360 outfile
88 your @var{GRAPH_DESCRIPTION} string will need to be of the form:
90 nullsrc,scale=640:360,nullsink
92 you may also need to set the @var{nullsrc} parameters and add a @var{format}
93 filter in order to simulate a specific input file.
97 @chapter Filtergraph description
98 @c man begin FILTERGRAPH DESCRIPTION
100 A filtergraph is a directed graph of connected filters. It can contain
101 cycles, and there can be multiple links between a pair of
102 filters. Each link has one input pad on one side connecting it to one
103 filter from which it takes its input, and one output pad on the other
104 side connecting it to the one filter accepting its output.
106 Each filter in a filtergraph is an instance of a filter class
107 registered in the application, which defines the features and the
108 number of input and output pads of the filter.
110 A filter with no input pads is called a "source", a filter with no
111 output pads is called a "sink".
113 @anchor{Filtergraph syntax}
114 @section Filtergraph syntax
116 A filtergraph can be represented using a textual representation, which is
117 recognized by the @option{-filter}/@option{-vf} and @option{-filter_complex}
118 options in @command{ffmpeg} and @option{-vf} in @command{ffplay}, and by the
119 @code{avfilter_graph_parse()}/@code{avfilter_graph_parse2()} function defined in
120 @file{libavfilter/avfiltergraph.h}.
122 A filterchain consists of a sequence of connected filters, each one
123 connected to the previous one in the sequence. A filterchain is
124 represented by a list of ","-separated filter descriptions.
126 A filtergraph consists of a sequence of filterchains. A sequence of
127 filterchains is represented by a list of ";"-separated filterchain
130 A filter is represented by a string of the form:
131 [@var{in_link_1}]...[@var{in_link_N}]@var{filter_name}=@var{arguments}[@var{out_link_1}]...[@var{out_link_M}]
133 @var{filter_name} is the name of the filter class of which the
134 described filter is an instance of, and has to be the name of one of
135 the filter classes registered in the program.
136 The name of the filter class is optionally followed by a string
139 @var{arguments} is a string which contains the parameters used to
140 initialize the filter instance. It may have one of the two allowed forms:
144 A ':'-separated list of @var{key=value} pairs.
147 A ':'-separated list of @var{value}. In this case, the keys are assumed to be
148 the option names in the order they are declared. E.g. the @code{fade} filter
149 declares three options in this order -- @option{type}, @option{start_frame} and
150 @option{nb_frames}. Then the parameter list @var{in:0:30} means that the value
151 @var{in} is assigned to the option @option{type}, @var{0} to
152 @option{start_frame} and @var{30} to @option{nb_frames}.
156 If the option value itself is a list of items (e.g. the @code{format} filter
157 takes a list of pixel formats), the items in the list are usually separated by
160 The list of arguments can be quoted using the character "'" as initial
161 and ending mark, and the character '\' for escaping the characters
162 within the quoted text; otherwise the argument string is considered
163 terminated when the next special character (belonging to the set
164 "[]=;,") is encountered.
166 The name and arguments of the filter are optionally preceded and
167 followed by a list of link labels.
168 A link label allows to name a link and associate it to a filter output
169 or input pad. The preceding labels @var{in_link_1}
170 ... @var{in_link_N}, are associated to the filter input pads,
171 the following labels @var{out_link_1} ... @var{out_link_M}, are
172 associated to the output pads.
174 When two link labels with the same name are found in the
175 filtergraph, a link between the corresponding input and output pad is
178 If an output pad is not labelled, it is linked by default to the first
179 unlabelled input pad of the next filter in the filterchain.
180 For example in the filterchain:
182 nullsrc, split[L1], [L2]overlay, nullsink
184 the split filter instance has two output pads, and the overlay filter
185 instance two input pads. The first output pad of split is labelled
186 "L1", the first input pad of overlay is labelled "L2", and the second
187 output pad of split is linked to the second input pad of overlay,
188 which are both unlabelled.
190 In a complete filterchain all the unlabelled filter input and output
191 pads must be connected. A filtergraph is considered valid if all the
192 filter input and output pads of all the filterchains are connected.
194 Libavfilter will automatically insert scale filters where format
195 conversion is required. It is possible to specify swscale flags
196 for those automatically inserted scalers by prepending
197 @code{sws_flags=@var{flags};}
198 to the filtergraph description.
200 Follows a BNF description for the filtergraph syntax:
202 @var{NAME} ::= sequence of alphanumeric characters and '_'
203 @var{LINKLABEL} ::= "[" @var{NAME} "]"
204 @var{LINKLABELS} ::= @var{LINKLABEL} [@var{LINKLABELS}]
205 @var{FILTER_ARGUMENTS} ::= sequence of chars (eventually quoted)
206 @var{FILTER} ::= [@var{LINKLABELS}] @var{NAME} ["=" @var{FILTER_ARGUMENTS}] [@var{LINKLABELS}]
207 @var{FILTERCHAIN} ::= @var{FILTER} [,@var{FILTERCHAIN}]
208 @var{FILTERGRAPH} ::= [sws_flags=@var{flags};] @var{FILTERCHAIN} [;@var{FILTERGRAPH}]
211 @section Notes on filtergraph escaping
213 Some filter arguments require the use of special characters, typically
214 @code{:} to separate key=value pairs in a named options list. In this
215 case the user should perform a first level escaping when specifying
216 the filter arguments. For example, consider the following literal
217 string to be embedded in the @ref{drawtext} filter arguments:
219 this is a 'string': may contain one, or more, special characters
222 Since @code{:} is special for the filter arguments syntax, it needs to
223 be escaped, so you get:
225 text=this is a \'string\'\: may contain one, or more, special characters
228 A second level of escaping is required when embedding the filter
229 arguments in a filtergraph description, in order to escape all the
230 filtergraph special characters. Thus the example above becomes:
232 drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
235 Finally an additional level of escaping may be needed when writing the
236 filtergraph description in a shell command, which depends on the
237 escaping rules of the adopted shell. For example, assuming that
238 @code{\} is special and needs to be escaped with another @code{\}, the
239 previous string will finally result in:
241 -vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
244 Sometimes, it might be more convenient to employ quoting in place of
245 escaping. For example the string:
247 Caesar: tu quoque, Brute, fili mi
250 Can be quoted in the filter arguments as:
252 text='Caesar: tu quoque, Brute, fili mi'
255 And finally inserted in a filtergraph like:
257 drawtext=text=\'Caesar: tu quoque\, Brute\, fili mi\'
260 See the ``Quoting and escaping'' section in the ffmpeg-utils manual
261 for more information about the escaping and quoting rules adopted by
264 @c man end FILTERGRAPH DESCRIPTION
266 @chapter Audio Filters
267 @c man begin AUDIO FILTERS
269 When you configure your FFmpeg build, you can disable any of the
270 existing filters using @code{--disable-filters}.
271 The configure output will show the audio filters included in your
274 Below is a description of the currently available audio filters.
278 Convert the input audio format to the specified formats.
280 The filter accepts a string of the form:
281 "@var{sample_format}:@var{channel_layout}".
283 @var{sample_format} specifies the sample format, and can be a string or the
284 corresponding numeric value defined in @file{libavutil/samplefmt.h}. Use 'p'
285 suffix for a planar sample format.
287 @var{channel_layout} specifies the channel layout, and can be a string
288 or the corresponding number value defined in @file{libavutil/channel_layout.h}.
290 The special parameter "auto", signifies that the filter will
291 automatically select the output format depending on the output filter.
297 Convert input to float, planar, stereo:
303 Convert input to unsigned 8-bit, automatically select out channel layout:
311 Apply a two-pole all-pass filter with central frequency (in Hz)
312 @var{frequency}, and filter-width @var{width}.
313 An all-pass filter changes the audio's frequency to phase relationship
314 without changing its frequency to amplitude relationship.
316 The filter accepts parameters as a list of @var{key}=@var{value}
317 pairs, separated by ":".
319 A description of the accepted parameters follows.
326 Set method to specify band-width of filter.
339 Specify the band-width of a filter in width_type units.
344 Apply a high-pass filter with 3dB point frequency.
345 The filter can be either single-pole, or double-pole (the default).
346 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
348 The filter accepts parameters as a list of @var{key}=@var{value}
349 pairs, separated by ":".
351 A description of the accepted parameters follows.
355 Set frequency in Hz. Default is 3000.
358 Set number of poles. Default is 2.
361 Set method to specify band-width of filter.
374 Specify the band-width of a filter in width_type units.
375 Applies only to double-pole filter.
376 The default is 0.707q and gives a Butterworth response.
381 Apply a low-pass filter with 3dB point frequency.
382 The filter can be either single-pole or double-pole (the default).
383 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
385 The filter accepts parameters as a list of @var{key}=@var{value}
386 pairs, separated by ":".
388 A description of the accepted parameters follows.
392 Set frequency in Hz. Default is 500.
395 Set number of poles. Default is 2.
398 Set method to specify band-width of filter.
411 Specify the band-width of a filter in width_type units.
412 Applies only to double-pole filter.
413 The default is 0.707q and gives a Butterworth response.
418 Boost or cut the bass (lower) frequencies of the audio using a two-pole
419 shelving filter with a response similar to that of a standard
420 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
422 The filter accepts parameters as a list of @var{key}=@var{value}
423 pairs, separated by ":".
425 A description of the accepted parameters follows.
429 Give the gain at 0 Hz. Its useful range is about -20
430 (for a large cut) to +20 (for a large boost).
431 Beware of clipping when using a positive gain.
434 Set the filter's central frequency and so can be used
435 to extend or reduce the frequency range to be boosted or cut.
436 The default value is @code{100} Hz.
439 Set method to specify band-width of filter.
452 Determine how steep is the filter's shelf transition.
457 Boost or cut treble (upper) frequencies of the audio using a two-pole
458 shelving filter with a response similar to that of a standard
459 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
461 The filter accepts parameters as a list of @var{key}=@var{value}
462 pairs, separated by ":".
464 A description of the accepted parameters follows.
468 Give the gain at whichever is the lower of ~22 kHz and the
469 Nyquist frequency. Its useful range is about -20 (for a large cut)
470 to +20 (for a large boost). Beware of clipping when using a positive gain.
473 Set the filter's central frequency and so can be used
474 to extend or reduce the frequency range to be boosted or cut.
475 The default value is @code{3000} Hz.
478 Set method to specify band-width of filter.
491 Determine how steep is the filter's shelf transition.
496 Apply a two-pole Butterworth band-pass filter with central
497 frequency @var{frequency}, and (3dB-point) band-width width.
498 The @var{csg} option selects a constant skirt gain (peak gain = Q)
499 instead of the default: constant 0dB peak gain.
500 The filter roll off at 6dB per octave (20dB per decade).
502 The filter accepts parameters as a list of @var{key}=@var{value}
503 pairs, separated by ":".
505 A description of the accepted parameters follows.
509 Set the filter's central frequency. Default is @code{3000}.
512 Constant skirt gain if set to 1. Defaults to 0.
515 Set method to specify band-width of filter.
528 Specify the band-width of a filter in width_type units.
533 Apply a two-pole Butterworth band-reject filter with central
534 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
535 The filter roll off at 6dB per octave (20dB per decade).
537 The filter accepts parameters as a list of @var{key}=@var{value}
538 pairs, separated by ":".
540 A description of the accepted parameters follows.
544 Set the filter's central frequency. Default is @code{3000}.
547 Set method to specify band-width of filter.
560 Specify the band-width of a filter in width_type units.
565 Apply a biquad IIR filter with the given coefficients.
566 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
567 are the numerator and denominator coefficients respectively.
571 Apply a two-pole peaking equalisation (EQ) filter. With this
572 filter, the signal-level at and around a selected frequency can
573 be increased or decreased, whilst (unlike bandpass and bandreject
574 filters) that at all other frequencies is unchanged.
576 In order to produce complex equalisation curves, this filter can
577 be given several times, each with a different central frequency.
579 The filter accepts parameters as a list of @var{key}=@var{value}
580 pairs, separated by ":".
582 A description of the accepted parameters follows.
586 Set the filter's central frequency in Hz.
589 Set method to specify band-width of filter.
602 Specify the band-width of a filter in width_type units.
605 Set the required gain or attenuation in dB.
606 Beware of clipping when using a positive gain.
611 Apply fade-in/out effect to input audio.
613 The filter accepts parameters as a list of @var{key}=@var{value}
614 pairs, separated by ":".
616 A description of the accepted parameters follows.
620 Specify the effect type, can be either @code{in} for fade-in, or
621 @code{out} for a fade-out effect. Default is @code{in}.
623 @item start_sample, ss
624 Specify the number of the start sample for starting to apply the fade
625 effect. Default is 0.
628 Specify the number of samples for which the fade effect has to last. At
629 the end of the fade-in effect the output audio will have the same
630 volume as the input audio, at the end of the fade-out transition
631 the output audio will be silence. Default is 44100.
634 Specify time in seconds for starting to apply the fade
635 effect. Default is 0.
636 If set this option is used instead of @var{start_sample} one.
639 Specify the number of seconds for which the fade effect has to last. At
640 the end of the fade-in effect the output audio will have the same
641 volume as the input audio, at the end of the fade-out transition
642 the output audio will be silence. Default is 0.
643 If set this option is used instead of @var{nb_samples} one.
646 Set curve for fade transition.
648 It accepts the following values:
651 select triangular, linear slope (default)
653 select quarter of sine wave
655 select half of sine wave
657 select exponential sine wave
661 select inverted parabola
677 Fade in first 15 seconds of audio:
683 Fade out last 25 seconds of a 900 seconds audio:
685 afade=t=out:ss=875:d=25
692 Set output format constraints for the input audio. The framework will
693 negotiate the most appropriate format to minimize conversions.
695 The filter accepts the following named parameters:
699 A '|'-separated list of requested sample formats.
702 A '|'-separated list of requested sample rates.
704 @item channel_layouts
705 A '|'-separated list of requested channel layouts.
709 If a parameter is omitted, all values are allowed.
711 For example to force the output to either unsigned 8-bit or signed 16-bit stereo:
713 aformat=sample_fmts=u8|s16:channel_layouts=stereo
718 Merge two or more audio streams into a single multi-channel stream.
720 The filter accepts the following named options:
725 Set the number of inputs. Default is 2.
729 If the channel layouts of the inputs are disjoint, and therefore compatible,
730 the channel layout of the output will be set accordingly and the channels
731 will be reordered as necessary. If the channel layouts of the inputs are not
732 disjoint, the output will have all the channels of the first input then all
733 the channels of the second input, in that order, and the channel layout of
734 the output will be the default value corresponding to the total number of
737 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
738 is FC+BL+BR, then the output will be in 5.1, with the channels in the
739 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
740 first input, b1 is the first channel of the second input).
742 On the other hand, if both input are in stereo, the output channels will be
743 in the default order: a1, a2, b1, b2, and the channel layout will be
744 arbitrarily set to 4.0, which may or may not be the expected value.
746 All inputs must have the same sample rate, and format.
748 If inputs do not have the same duration, the output will stop with the
755 Merge two mono files into a stereo stream:
757 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
764 amovie=input.mkv:si=0 [a0];
765 amovie=input.mkv:si=1 [a1];
766 amovie=input.mkv:si=2 [a2];
767 amovie=input.mkv:si=3 [a3];
768 amovie=input.mkv:si=4 [a4];
769 amovie=input.mkv:si=5 [a5];
770 [a0][a1][a2][a3][a4][a5] amerge=inputs=6" -c:a pcm_s16le output.mkv
776 Mixes multiple audio inputs into a single output.
780 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
782 will mix 3 input audio streams to a single output with the same duration as the
783 first input and a dropout transition time of 3 seconds.
785 The filter accepts the following named parameters:
789 Number of inputs. If unspecified, it defaults to 2.
792 How to determine the end-of-stream.
796 Duration of longest input. (default)
799 Duration of shortest input.
802 Duration of first input.
806 @item dropout_transition
807 Transition time, in seconds, for volume renormalization when an input
808 stream ends. The default value is 2 seconds.
814 Pass the audio source unchanged to the output.
818 Pad the end of a audio stream with silence, this can be used together with
819 -shortest to extend audio streams to the same length as the video stream.
824 Resample the input audio to the specified parameters, using the
825 libswresample library. If none are specified then the filter will
826 automatically convert between its input and output.
828 This filter is also able to stretch/squeeze the audio data to make it match
829 the timestamps or to inject silence / cut out audio to make it match the
830 timestamps, do a combination of both or do neither.
832 The filter accepts the syntax
833 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
834 expresses a sample rate and @var{resampler_options} is a list of
835 @var{key}=@var{value} pairs, separated by ":". See the
836 ffmpeg-resampler manual for the complete list of supported options.
842 Resample the input audio to 44100Hz:
848 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
849 samples per second compensation:
855 @section asetnsamples
857 Set the number of samples per each output audio frame.
859 The last output packet may contain a different number of samples, as
860 the filter will flush all the remaining samples when the input audio
863 The filter accepts parameters as a list of @var{key}=@var{value} pairs,
868 @item nb_out_samples, n
869 Set the number of frames per each output audio frame. The number is
870 intended as the number of samples @emph{per each channel}.
871 Default value is 1024.
874 If set to 1, the filter will pad the last audio frame with zeroes, so
875 that the last frame will contain the same number of samples as the
876 previous ones. Default value is 1.
879 For example, to set the number of per-frame samples to 1234 and
880 disable padding for the last frame, use:
882 asetnsamples=n=1234:p=0
887 Show a line containing various information for each input audio frame.
888 The input audio is not modified.
890 The shown line contains a sequence of key/value pairs of the form
891 @var{key}:@var{value}.
893 A description of each shown parameter follows:
897 sequential number of the input frame, starting from 0
900 Presentation timestamp of the input frame, in time base units; the time base
901 depends on the filter input pad, and is usually 1/@var{sample_rate}.
904 presentation timestamp of the input frame in seconds
907 position of the frame in the input stream, -1 if this information in
908 unavailable and/or meaningless (for example in case of synthetic audio)
917 sample rate for the audio frame
920 number of samples (per channel) in the frame
923 Adler-32 checksum (printed in hexadecimal) of the audio data. For planar audio
924 the data is treated as if all the planes were concatenated.
926 @item plane_checksums
927 A list of Adler-32 checksums for each data plane.
932 Split input audio into several identical outputs.
934 The filter accepts a single parameter which specifies the number of outputs. If
935 unspecified, it defaults to 2.
939 [in] asplit [out0][out1]
942 will create two separate outputs from the same input.
944 To create 3 or more outputs, you need to specify the number of
947 [in] asplit=3 [out0][out1][out2]
951 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
953 will create 5 copies of the input audio.
958 Forward two audio streams and control the order the buffers are forwarded.
960 The argument to the filter is an expression deciding which stream should be
961 forwarded next: if the result is negative, the first stream is forwarded; if
962 the result is positive or zero, the second stream is forwarded. It can use
963 the following variables:
967 number of buffers forwarded so far on each stream
969 number of samples forwarded so far on each stream
971 current timestamp of each stream
974 The default value is @code{t1-t2}, which means to always forward the stream
975 that has a smaller timestamp.
977 Example: stress-test @code{amerge} by randomly sending buffers on the wrong
978 input, while avoiding too much of a desynchronization:
980 amovie=file.ogg [a] ; amovie=file.mp3 [b] ;
981 [a] [b] astreamsync=(2*random(1))-1+tanh(5*(t1-t2)) [a2] [b2] ;
989 The filter accepts exactly one parameter, the audio tempo. If not
990 specified then the filter will assume nominal 1.0 tempo. Tempo must
991 be in the [0.5, 2.0] range.
997 Slow down audio to 80% tempo:
1003 To speed up audio to 125% tempo:
1011 Make audio easier to listen to on headphones.
1013 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
1014 so that when listened to on headphones the stereo image is moved from
1015 inside your head (standard for headphones) to outside and in front of
1016 the listener (standard for speakers).
1022 Mix channels with specific gain levels. The filter accepts the output
1023 channel layout followed by a set of channels definitions.
1025 This filter is also designed to remap efficiently the channels of an audio
1028 The filter accepts parameters of the form:
1029 "@var{l}:@var{outdef}:@var{outdef}:..."
1033 output channel layout or number of channels
1036 output channel specification, of the form:
1037 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
1040 output channel to define, either a channel name (FL, FR, etc.) or a channel
1041 number (c0, c1, etc.)
1044 multiplicative coefficient for the channel, 1 leaving the volume unchanged
1047 input channel to use, see out_name for details; it is not possible to mix
1048 named and numbered input channels
1051 If the `=' in a channel specification is replaced by `<', then the gains for
1052 that specification will be renormalized so that the total is 1, thus
1053 avoiding clipping noise.
1055 @subsection Mixing examples
1057 For example, if you want to down-mix from stereo to mono, but with a bigger
1058 factor for the left channel:
1060 pan=1:c0=0.9*c0+0.1*c1
1063 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
1064 7-channels surround:
1066 pan=stereo: FL < FL + 0.5*FC + 0.6*BL + 0.6*SL : FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
1069 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
1070 that should be preferred (see "-ac" option) unless you have very specific
1073 @subsection Remapping examples
1075 The channel remapping will be effective if, and only if:
1078 @item gain coefficients are zeroes or ones,
1079 @item only one input per channel output,
1082 If all these conditions are satisfied, the filter will notify the user ("Pure
1083 channel mapping detected"), and use an optimized and lossless method to do the
1086 For example, if you have a 5.1 source and want a stereo audio stream by
1087 dropping the extra channels:
1089 pan="stereo: c0=FL : c1=FR"
1092 Given the same source, you can also switch front left and front right channels
1093 and keep the input channel layout:
1095 pan="5.1: c0=c1 : c1=c0 : c2=c2 : c3=c3 : c4=c4 : c5=c5"
1098 If the input is a stereo audio stream, you can mute the front left channel (and
1099 still keep the stereo channel layout) with:
1104 Still with a stereo audio stream input, you can copy the right channel in both
1105 front left and right:
1107 pan="stereo: c0=FR : c1=FR"
1110 @section silencedetect
1112 Detect silence in an audio stream.
1114 This filter logs a message when it detects that the input audio volume is less
1115 or equal to a noise tolerance value for a duration greater or equal to the
1116 minimum detected noise duration.
1118 The printed times and duration are expressed in seconds.
1122 Set silence duration until notification (default is 2 seconds).
1125 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
1126 specified value) or amplitude ratio. Default is -60dB, or 0.001.
1129 @subsection Examples
1133 Detect 5 seconds of silence with -50dB noise tolerance:
1135 silencedetect=n=-50dB:d=5
1139 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
1140 tolerance in @file{silence.mp3}:
1142 ffmpeg -f lavfi -i amovie=silence.mp3,silencedetect=noise=0.0001 -f null -
1147 Synchronize audio data with timestamps by squeezing/stretching it and/or
1148 dropping samples/adding silence when needed.
1150 This filter is not built by default, please use @ref{aresample} to do squeezing/stretching.
1152 The filter accepts the following named parameters:
1156 Enable stretching/squeezing the data to make it match the timestamps. Disabled
1157 by default. When disabled, time gaps are covered with silence.
1160 Minimum difference between timestamps and audio data (in seconds) to trigger
1161 adding/dropping samples. Default value is 0.1. If you get non-perfect sync with
1162 this filter, try setting this parameter to 0.
1165 Maximum compensation in samples per second. Relevant only with compensate=1.
1169 Assume the first pts should be this value. The time base is 1 / sample rate.
1170 This allows for padding/trimming at the start of stream. By default, no
1171 assumption is made about the first frame's expected pts, so no padding or
1172 trimming is done. For example, this could be set to 0 to pad the beginning with
1173 silence if an audio stream starts after the video stream or to trim any samples
1174 with a negative pts due to encoder delay.
1178 @section channelsplit
1179 Split each channel in input audio stream into a separate output stream.
1181 This filter accepts the following named parameters:
1183 @item channel_layout
1184 Channel layout of the input stream. Default is "stereo".
1187 For example, assuming a stereo input MP3 file
1189 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1191 will create an output Matroska file with two audio streams, one containing only
1192 the left channel and the other the right channel.
1194 To split a 5.1 WAV file into per-channel files
1196 ffmpeg -i in.wav -filter_complex
1197 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1198 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1199 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1204 Remap input channels to new locations.
1206 This filter accepts the following named parameters:
1208 @item channel_layout
1209 Channel layout of the output stream.
1212 Map channels from input to output. The argument is a comma-separated list of
1213 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1214 @var{in_channel} form. @var{in_channel} can be either the name of the input
1215 channel (e.g. FL for front left) or its index in the input channel layout.
1216 @var{out_channel} is the name of the output channel or its index in the output
1217 channel layout. If @var{out_channel} is not given then it is implicitly an
1218 index, starting with zero and increasing by one for each mapping.
1221 If no mapping is present, the filter will implicitly map input channels to
1222 output channels preserving index.
1224 For example, assuming a 5.1+downmix input MOV file
1226 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL\,DR-FR' out.wav
1228 will create an output WAV file tagged as stereo from the downmix channels of
1231 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1233 ffmpeg -i in.wav -filter 'channelmap=1\,2\,0\,5\,3\,4:channel_layout=5.1' out.wav
1237 Join multiple input streams into one multi-channel stream.
1239 The filter accepts the following named parameters:
1243 Number of input streams. Defaults to 2.
1245 @item channel_layout
1246 Desired output channel layout. Defaults to stereo.
1249 Map channels from inputs to output. The argument is a comma-separated list of
1250 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
1251 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
1252 can be either the name of the input channel (e.g. FL for front left) or its
1253 index in the specified input stream. @var{out_channel} is the name of the output
1257 The filter will attempt to guess the mappings when those are not specified
1258 explicitly. It does so by first trying to find an unused matching input channel
1259 and if that fails it picks the first unused input channel.
1261 E.g. to join 3 inputs (with properly set channel layouts)
1263 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
1266 To build a 5.1 output from 6 single-channel streams:
1268 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
1269 'join=inputs=6:channel_layout=5.1:map=0.0-FL\,1.0-FR\,2.0-FC\,3.0-SL\,4.0-SR\,5.0-LFE'
1274 Convert the audio sample format, sample rate and channel layout. This filter is
1275 not meant to be used directly.
1279 Adjust the input audio volume.
1281 The filter accepts the following named parameters. If the key of the
1282 first options is omitted, the arguments are interpreted according to
1283 the following syntax:
1285 volume=@var{volume}:@var{precision}
1291 Expresses how the audio volume will be increased or decreased.
1293 Output values are clipped to the maximum value.
1295 The output audio volume is given by the relation:
1297 @var{output_volume} = @var{volume} * @var{input_volume}
1300 Default value for @var{volume} is 1.0.
1303 Set the mathematical precision.
1305 This determines which input sample formats will be allowed, which affects the
1306 precision of the volume scaling.
1310 8-bit fixed-point; limits input sample format to U8, S16, and S32.
1312 32-bit floating-point; limits input sample format to FLT. (default)
1314 64-bit floating-point; limits input sample format to DBL.
1318 @subsection Examples
1322 Halve the input audio volume:
1326 volume=volume=-6.0206dB
1329 In all the above example the named key for @option{volume} can be
1330 omitted, for example like in:
1336 Increase input audio power by 6 decibels using fixed-point precision:
1338 volume=volume=6dB:precision=fixed
1342 @section volumedetect
1344 Detect the volume of the input video.
1346 The filter has no parameters. The input is not modified. Statistics about
1347 the volume will be printed in the log when the input stream end is reached.
1349 In particular it will show the mean volume (root mean square), maximum
1350 volume (on a per-sample basis), and the beginning of an histogram of the
1351 registered volume values (from the maximum value to a cumulated 1/1000 of
1354 All volumes are in decibels relative to the maximum PCM value.
1356 @subsection Examples
1358 Here is an excerpt of the output:
1360 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
1361 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
1362 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
1363 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
1364 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
1365 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
1366 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
1367 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
1368 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
1374 The mean square energy is approximately -27 dB, or 10^-2.7.
1376 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
1378 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
1381 In other words, raising the volume by +4 dB does not cause any clipping,
1382 raising it by +5 dB causes clipping for 6 samples, etc.
1384 @c man end AUDIO FILTERS
1386 @chapter Audio Sources
1387 @c man begin AUDIO SOURCES
1389 Below is a description of the currently available audio sources.
1393 Buffer audio frames, and make them available to the filter chain.
1395 This source is mainly intended for a programmatic use, in particular
1396 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
1398 It accepts the following mandatory parameters:
1399 @var{sample_rate}:@var{sample_fmt}:@var{channel_layout}
1404 The sample rate of the incoming audio buffers.
1407 The sample format of the incoming audio buffers.
1408 Either a sample format name or its corresponging integer representation from
1409 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
1411 @item channel_layout
1412 The channel layout of the incoming audio buffers.
1413 Either a channel layout name from channel_layout_map in
1414 @file{libavutil/channel_layout.c} or its corresponding integer representation
1415 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
1418 The number of channels of the incoming audio buffers.
1419 If both @var{channels} and @var{channel_layout} are specified, then they
1424 @subsection Examples
1427 abuffer=44100:s16p:stereo
1430 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
1431 Since the sample format with name "s16p" corresponds to the number
1432 6 and the "stereo" channel layout corresponds to the value 0x3, this is
1440 Generate an audio signal specified by an expression.
1442 This source accepts in input one or more expressions (one for each
1443 channel), which are evaluated and used to generate a corresponding
1446 It accepts the syntax: @var{exprs}[::@var{options}].
1447 @var{exprs} is a list of expressions separated by ":", one for each
1448 separate channel. In case the @var{channel_layout} is not
1449 specified, the selected channel layout depends on the number of
1450 provided expressions.
1452 @var{options} is an optional sequence of @var{key}=@var{value} pairs,
1455 The description of the accepted options follows.
1459 @item channel_layout, c
1460 Set the channel layout. The number of channels in the specified layout
1461 must be equal to the number of specified expressions.
1464 Set the minimum duration of the sourced audio. See the function
1465 @code{av_parse_time()} for the accepted format.
1466 Note that the resulting duration may be greater than the specified
1467 duration, as the generated audio is always cut at the end of a
1470 If not specified, or the expressed duration is negative, the audio is
1471 supposed to be generated forever.
1474 Set the number of samples per channel per each output frame,
1477 @item sample_rate, s
1478 Specify the sample rate, default to 44100.
1481 Each expression in @var{exprs} can contain the following constants:
1485 number of the evaluated sample, starting from 0
1488 time of the evaluated sample expressed in seconds, starting from 0
1495 @subsection Examples
1505 Generate a sin signal with frequency of 440 Hz, set sample rate to
1508 aevalsrc="sin(440*2*PI*t)::s=8000"
1512 Generate a two channels signal, specify the channel layout (Front
1513 Center + Back Center) explicitly:
1515 aevalsrc="sin(420*2*PI*t):cos(430*2*PI*t)::c=FC|BC"
1519 Generate white noise:
1521 aevalsrc="-2+random(0)"
1525 Generate an amplitude modulated signal:
1527 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
1531 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
1533 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) : 0.1*sin(2*PI*(360+2.5/2)*t)"
1540 Null audio source, return unprocessed audio frames. It is mainly useful
1541 as a template and to be employed in analysis / debugging tools, or as
1542 the source for filters which ignore the input data (for example the sox
1545 It accepts an optional sequence of @var{key}=@var{value} pairs,
1548 The description of the accepted options follows.
1552 @item sample_rate, s
1553 Specify the sample rate, and defaults to 44100.
1555 @item channel_layout, cl
1557 Specify the channel layout, and can be either an integer or a string
1558 representing a channel layout. The default value of @var{channel_layout}
1561 Check the channel_layout_map definition in
1562 @file{libavutil/channel_layout.c} for the mapping between strings and
1563 channel layout values.
1566 Set the number of samples per requested frames.
1570 @subsection Examples
1574 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
1576 anullsrc=r=48000:cl=4
1580 Do the same operation with a more obvious syntax:
1582 anullsrc=r=48000:cl=mono
1587 Buffer audio frames, and make them available to the filter chain.
1589 This source is not intended to be part of user-supplied graph descriptions but
1590 for insertion by calling programs through the interface defined in
1591 @file{libavfilter/buffersrc.h}.
1593 It accepts the following named parameters:
1597 Timebase which will be used for timestamps of submitted frames. It must be
1598 either a floating-point number or in @var{numerator}/@var{denominator} form.
1604 Name of the sample format, as returned by @code{av_get_sample_fmt_name()}.
1606 @item channel_layout
1607 Channel layout of the audio data, in the form that can be accepted by
1608 @code{av_get_channel_layout()}.
1611 All the parameters need to be explicitly defined.
1615 Synthesize a voice utterance using the libflite library.
1617 To enable compilation of this filter you need to configure FFmpeg with
1618 @code{--enable-libflite}.
1620 Note that the flite library is not thread-safe.
1622 The source accepts parameters as a list of @var{key}=@var{value} pairs,
1625 The description of the accepted parameters follows.
1630 If set to 1, list the names of the available voices and exit
1631 immediately. Default value is 0.
1634 Set the maximum number of samples per frame. Default value is 512.
1637 Set the filename containing the text to speak.
1640 Set the text to speak.
1643 Set the voice to use for the speech synthesis. Default value is
1644 @code{kal}. See also the @var{list_voices} option.
1647 @subsection Examples
1651 Read from file @file{speech.txt}, and synthetize the text using the
1652 standard flite voice:
1654 flite=textfile=speech.txt
1658 Read the specified text selecting the @code{slt} voice:
1660 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
1664 Input text to ffmpeg:
1666 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
1670 Make @file{ffplay} speak the specified text, using @code{flite} and
1671 the @code{lavfi} device:
1673 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
1677 For more information about libflite, check:
1678 @url{http://www.speech.cs.cmu.edu/flite/}
1682 Generate an audio signal made of a sine wave with amplitude 1/8.
1684 The audio signal is bit-exact.
1686 It accepts a list of options in the form of @var{key}=@var{value} pairs
1687 separated by ":". If the option name is omitted, the first option is the
1688 frequency and the second option is the beep factor.
1690 The supported options are:
1695 Set the carrier frequency. Default is 440 Hz.
1697 @item beep_factor, b
1698 Enable a periodic beep every second with frequency @var{beep_factor} times
1699 the carrier frequency. Default is 0, meaning the beep is disabled.
1701 @item sample_rate, s
1702 Specify the sample rate, default is 44100.
1705 Specify the duration of the generated audio stream.
1707 @item samples_per_frame
1708 Set the number of samples per output frame, default is 1024.
1711 @subsection Examples
1716 Generate a simple 440 Hz sine wave:
1722 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
1726 sine=frequency=220:beep_factor=4:duration=5
1731 @c man end AUDIO SOURCES
1733 @chapter Audio Sinks
1734 @c man begin AUDIO SINKS
1736 Below is a description of the currently available audio sinks.
1738 @section abuffersink
1740 Buffer audio frames, and make them available to the end of filter chain.
1742 This sink is mainly intended for programmatic use, in particular
1743 through the interface defined in @file{libavfilter/buffersink.h}.
1745 It requires a pointer to an AVABufferSinkContext structure, which
1746 defines the incoming buffers' formats, to be passed as the opaque
1747 parameter to @code{avfilter_init_filter} for initialization.
1751 Null audio sink, do absolutely nothing with the input audio. It is
1752 mainly useful as a template and to be employed in analysis / debugging
1755 @section abuffersink
1756 This sink is intended for programmatic use. Frames that arrive on this sink can
1757 be retrieved by the calling program using the interface defined in
1758 @file{libavfilter/buffersink.h}.
1760 This filter accepts no parameters.
1762 @c man end AUDIO SINKS
1764 @chapter Video Filters
1765 @c man begin VIDEO FILTERS
1767 When you configure your FFmpeg build, you can disable any of the
1768 existing filters using @code{--disable-filters}.
1769 The configure output will show the video filters included in your
1772 Below is a description of the currently available video filters.
1774 @section alphaextract
1776 Extract the alpha component from the input as a grayscale video. This
1777 is especially useful with the @var{alphamerge} filter.
1781 Add or replace the alpha component of the primary input with the
1782 grayscale value of a second input. This is intended for use with
1783 @var{alphaextract} to allow the transmission or storage of frame
1784 sequences that have alpha in a format that doesn't support an alpha
1787 For example, to reconstruct full frames from a normal YUV-encoded video
1788 and a separate video created with @var{alphaextract}, you might use:
1790 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
1793 Since this filter is designed for reconstruction, it operates on frame
1794 sequences without considering timestamps, and terminates when either
1795 input reaches end of stream. This will cause problems if your encoding
1796 pipeline drops frames. If you're trying to apply an image as an
1797 overlay to a video stream, consider the @var{overlay} filter instead.
1801 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
1802 and libavformat to work. On the other hand, it is limited to ASS (Advanced
1803 Substation Alpha) subtitles files.
1807 Compute the bounding box for the non-black pixels in the input frame
1810 This filter computes the bounding box containing all the pixels with a
1811 luminance value greater than the minimum allowed value.
1812 The parameters describing the bounding box are printed on the filter
1815 @section blackdetect
1817 Detect video intervals that are (almost) completely black. Can be
1818 useful to detect chapter transitions, commercials, or invalid
1819 recordings. Output lines contains the time for the start, end and
1820 duration of the detected black interval expressed in seconds.
1822 In order to display the output lines, you need to set the loglevel at
1823 least to the AV_LOG_INFO value.
1825 This filter accepts a list of options in the form of
1826 @var{key}=@var{value} pairs separated by ":". A description of the
1827 accepted options follows.
1830 @item black_min_duration, d
1831 Set the minimum detected black duration expressed in seconds. It must
1832 be a non-negative floating point number.
1834 Default value is 2.0.
1836 @item picture_black_ratio_th, pic_th
1837 Set the threshold for considering a picture "black".
1838 Express the minimum value for the ratio:
1840 @var{nb_black_pixels} / @var{nb_pixels}
1843 for which a picture is considered black.
1844 Default value is 0.98.
1846 @item pixel_black_th, pix_th
1847 Set the threshold for considering a pixel "black".
1849 The threshold expresses the maximum pixel luminance value for which a
1850 pixel is considered "black". The provided value is scaled according to
1851 the following equation:
1853 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
1856 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
1857 the input video format, the range is [0-255] for YUV full-range
1858 formats and [16-235] for YUV non full-range formats.
1860 Default value is 0.10.
1863 The following example sets the maximum pixel threshold to the minimum
1864 value, and detects only black intervals of 2 or more seconds:
1866 blackdetect=d=2:pix_th=0.00
1871 Detect frames that are (almost) completely black. Can be useful to
1872 detect chapter transitions or commercials. Output lines consist of
1873 the frame number of the detected frame, the percentage of blackness,
1874 the position in the file if known or -1 and the timestamp in seconds.
1876 In order to display the output lines, you need to set the loglevel at
1877 least to the AV_LOG_INFO value.
1879 The filter accepts parameters as a list of @var{key}=@var{value}
1880 pairs, separated by ":". If the key of the first options is omitted,
1881 the arguments are interpreted according to the syntax
1882 blackframe[=@var{amount}[:@var{threshold}]].
1884 The filter accepts the following options:
1889 The percentage of the pixels that have to be below the threshold, defaults to
1893 Threshold below which a pixel value is considered black, defaults to 32.
1899 Blend two video frames into each other.
1901 It takes two input streams and outputs one stream, the first input is the
1902 "top" layer and second input is "bottom" layer.
1903 Output terminates when shortest input terminates.
1905 This filter accepts a list of options in the form of @var{key}=@var{value}
1906 pairs separated by ":". A description of the accepted options follows.
1914 Set blend mode for specific pixel component or all pixel components in case
1915 of @var{all_mode}. Default value is @code{normal}.
1917 Available values for component modes are:
1950 Set blend opacity for specific pixel component or all pixel components in case
1951 of @var{all_opacity}. Only used in combination with pixel component blend modes.
1958 Set blend expression for specific pixel component or all pixel components in case
1959 of @var{all_expr}. Note that related mode options will be ignored if those are set.
1961 The expressions can use the following variables:
1965 The sequential number of the filtered frame, starting from @code{0}.
1969 the coordinates of the current sample
1973 the width and height of currently filtered plane
1977 Width and height scale depending on the currently filtered plane. It is the
1978 ratio between the corresponding luma plane number of pixels and the current
1979 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
1980 @code{0.5,0.5} for chroma planes.
1983 Time of the current frame, expressed in seconds.
1986 Value of pixel component at current location for first video frame (top layer).
1989 Value of pixel component at current location for second video frame (bottom layer).
1993 @subsection Examples
1997 Apply transition from bottom layer to top layer in first 10 seconds:
1999 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
2003 Apply 1x1 checkerboard effect:
2005 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
2011 Apply boxblur algorithm to the input video.
2013 The filter accepts parameters as a list of @var{key}=@var{value}
2014 pairs, separated by ":". If the key of the first options is omitted,
2015 the arguments are interpreted according to the syntax
2016 @option{luma_radius}:@option{luma_power}:@option{chroma_radius}:@option{chroma_power}:@option{alpha_radius}:@option{alpha_power}.
2018 This filter accepts the following options:
2031 A description of the accepted options follows.
2034 @item luma_radius, lr
2035 @item chroma_radius, cr
2036 @item alpha_radius, ar
2037 Set an expression for the box radius in pixels used for blurring the
2038 corresponding input plane.
2040 The radius value must be a non-negative number, and must not be
2041 greater than the value of the expression @code{min(w,h)/2} for the
2042 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
2045 Default value for @option{luma_radius} is "2". If not specified,
2046 @option{chroma_radius} and @option{alpha_radius} default to the
2047 corresponding value set for @option{luma_radius}.
2049 The expressions can contain the following constants:
2052 the input width and height in pixels
2055 the input chroma image width and height in pixels
2058 horizontal and vertical chroma subsample values. For example for the
2059 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
2062 @item luma_power, lp
2063 @item chroma_power, cp
2064 @item alpha_power, ap
2065 Specify how many times the boxblur filter is applied to the
2066 corresponding plane.
2068 Default value for @option{luma_power} is 2. If not specified,
2069 @option{chroma_power} and @option{alpha_power} default to the
2070 corresponding value set for @option{luma_power}.
2072 A value of 0 will disable the effect.
2075 @subsection Examples
2079 Apply a boxblur filter with luma, chroma, and alpha radius
2082 boxblur=luma_radius=2:luma_power=1
2087 Set luma radius to 2, alpha and chroma radius to 0:
2089 boxblur=2:1:cr=0:ar=0
2093 Set luma and chroma radius to a fraction of the video dimension:
2095 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
2099 @section colormatrix
2101 Convert color matrix.
2103 The filter accepts parameters as a list of @var{key}=@var{value}
2104 pairs, separated by ":". If the key of the first options is omitted,
2105 the arguments are interpreted according to the syntax
2106 @var{src}:@var{dst}.
2108 A description of the accepted options follows:
2112 Specify the source and destination color matrix. Both values must be
2115 The accepted values are:
2131 For example to convert from BT.601 to SMPTE-240M, use the command:
2133 colormatrix=bt601:smpte240m
2138 Copy the input source unchanged to the output. Mainly useful for
2143 Crop the input video to given dimensions.
2145 This filter accepts a list of @var{key}=@var{value} pairs as argument,
2146 separated by ':'. If the key of the first options is omitted, the
2147 arguments are interpreted according to the syntax
2148 @var{out_w}:@var{out_h}:@var{x}:@var{y}:@var{keep_aspect}.
2150 A description of the accepted options follows:
2153 Width of the output video. It defaults to @code{iw}.
2154 This expression is evaluated only once during the filter
2158 Height of the output video. It defaults to @code{ih}.
2159 This expression is evaluated only once during the filter
2163 Horizontal position, in the input video, of the left edge of the output video.
2164 It defaults to @code{(in_w-out_w)/2}.
2165 This expression is evaluated per-frame.
2168 Vertical position, in the input video, of the top edge of the output video.
2169 It defaults to @code{(in_h-out_h)/2}.
2170 This expression is evaluated per-frame.
2173 If set to 1 will force the output display aspect ratio
2174 to be the same of the input, by changing the output sample aspect
2175 ratio. It defaults to 0.
2178 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
2179 expressions containing the following constants:
2183 the computed values for @var{x} and @var{y}. They are evaluated for
2187 the input width and height
2190 same as @var{in_w} and @var{in_h}
2193 the output (cropped) width and height
2196 same as @var{out_w} and @var{out_h}
2199 same as @var{iw} / @var{ih}
2202 input sample aspect ratio
2205 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
2208 horizontal and vertical chroma subsample values. For example for the
2209 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
2212 the number of input frame, starting from 0
2215 timestamp expressed in seconds, NAN if the input timestamp is unknown
2219 The expression for @var{out_w} may depend on the value of @var{out_h},
2220 and the expression for @var{out_h} may depend on @var{out_w}, but they
2221 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
2222 evaluated after @var{out_w} and @var{out_h}.
2224 The @var{x} and @var{y} parameters specify the expressions for the
2225 position of the top-left corner of the output (non-cropped) area. They
2226 are evaluated for each frame. If the evaluated value is not valid, it
2227 is approximated to the nearest valid value.
2229 The expression for @var{x} may depend on @var{y}, and the expression
2230 for @var{y} may depend on @var{x}.
2232 @subsection Examples
2236 Crop area with size 100x100 at position (12,34).
2241 Using named options, the example above becomes:
2243 crop=w=100:h=100:x=12:y=34
2247 Crop the central input area with size 100x100:
2253 Crop the central input area with size 2/3 of the input video:
2255 crop=2/3*in_w:2/3*in_h
2259 Crop the input video central square:
2266 Delimit the rectangle with the top-left corner placed at position
2267 100:100 and the right-bottom corner corresponding to the right-bottom
2268 corner of the input image:
2270 crop=in_w-100:in_h-100:100:100
2274 Crop 10 pixels from the left and right borders, and 20 pixels from
2275 the top and bottom borders
2277 crop=in_w-2*10:in_h-2*20
2281 Keep only the bottom right quarter of the input image:
2283 crop=in_w/2:in_h/2:in_w/2:in_h/2
2287 Crop height for getting Greek harmony:
2289 crop=in_w:1/PHI*in_w
2293 Appply trembling effect:
2295 crop=in_w/2:in_h/2:(in_w-out_w)/2+((in_w-out_w)/2)*sin(n/10):(in_h-out_h)/2 +((in_h-out_h)/2)*sin(n/7)
2299 Apply erratic camera effect depending on timestamp:
2301 crop=in_w/2:in_h/2:(in_w-out_w)/2+((in_w-out_w)/2)*sin(t*10):(in_h-out_h)/2 +((in_h-out_h)/2)*sin(t*13)"
2305 Set x depending on the value of y:
2307 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
2313 Auto-detect crop size.
2315 Calculate necessary cropping parameters and prints the recommended
2316 parameters through the logging system. The detected dimensions
2317 correspond to the non-black area of the input video.
2319 The filter accepts parameters as a list of @var{key}=@var{value}
2320 pairs, separated by ":". If the key of the first options is omitted,
2321 the arguments are interpreted according to the syntax
2322 [@option{limit}[:@option{round}[:@option{reset}]]].
2324 A description of the accepted options follows.
2329 Set higher black value threshold, which can be optionally specified
2330 from nothing (0) to everything (255). An intensity value greater
2331 to the set value is considered non-black. Default value is 24.
2334 Set the value for which the width/height should be divisible by. The
2335 offset is automatically adjusted to center the video. Use 2 to get
2336 only even dimensions (needed for 4:2:2 video). 16 is best when
2337 encoding to most video codecs. Default value is 16.
2340 Set the counter that determines after how many frames cropdetect will
2341 reset the previously detected largest video area and start over to
2342 detect the current optimal crop area. Default value is 0.
2344 This can be useful when channel logos distort the video area. 0
2345 indicates never reset and return the largest area encountered during
2351 Apply color adjustments using curves.
2353 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
2354 component (red, green and blue) has its values defined by @var{N} key points
2355 tied from each other using a smooth curve. The x-axis represents the pixel
2356 values from the input frame, and the y-axis the new pixel values to be set for
2359 By default, a component curve is defined by the two points @var{(0;0)} and
2360 @var{(1;1)}. This creates a straight line where each original pixel value is
2361 "adjusted" to its own value, which means no change to the image.
2363 The filter allows you to redefine these two points and add some more. A new
2364 curve (using a natural cubic spline interpolation) will be define to pass
2365 smoothly through all these new coordinates. The new defined points needs to be
2366 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
2367 be in the @var{[0;1]} interval. If the computed curves happened to go outside
2368 the vector spaces, the values will be clipped accordingly.
2370 If there is no key point defined in @code{x=0}, the filter will automatically
2371 insert a @var{(0;0)} point. In the same way, if there is no key point defined
2372 in @code{x=1}, the filter will automatically insert a @var{(1;1)} point.
2374 The filter accepts parameters as a list of @var{key}=@var{value}
2375 pairs, separated by ":". If the key of the first options is omitted,
2376 the arguments are interpreted according to the syntax
2377 curves[=@var{preset}].
2379 A description of the accepted parameters follows.
2383 Set the key points for the red component.
2385 Set the key points for the green component.
2387 Set the key points for the blue component.
2389 Set the key points for all components.
2390 Can be used in addition to the other key points component
2391 options. In this case, the unset component(s) will fallback on this
2392 @option{all} setting.
2394 Select one of the available color presets. This option can not be used in
2395 addition to the @option{r}, @option{g}, @option{b} parameters.
2396 Available presets are:
2398 @item color_negative
2401 @item increase_contrast
2403 @item linear_contrast
2404 @item medium_contrast
2406 @item strong_contrast
2412 To avoid some filtergraph syntax conflicts, each key points list need to be
2413 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
2415 @subsection Examples
2419 Increase slightly the middle level of blue:
2421 curves=blue='0.5/0.58'
2427 curves=r='0/0.11 .42/.51 1/0.95':g='0.50/0.48':b='0/0.22 .49/.44 1/0.8'
2429 Here we obtain the following coordinates for each components:
2432 @code{(0;0.11) (0.42;0.51) (1;0.95)}
2434 @code{(0;0) (0.50;0.48) (1;1)}
2436 @code{(0;0.22) (0.49;0.44) (1;0.80)}
2440 The previous example can also be achieved with the associated built-in preset:
2442 curves=preset=vintage
2454 Drop frames that do not differ greatly from the previous frame in
2455 order to reduce frame rate.
2457 The main use of this filter is for very-low-bitrate encoding
2458 (e.g. streaming over dialup modem), but it could in theory be used for
2459 fixing movies that were inverse-telecined incorrectly.
2461 The filter accepts parameters as a list of @var{key}=@var{value}
2462 pairs, separated by ":". If the key of the first options is omitted,
2463 the arguments are interpreted according to the syntax:
2464 @option{max}:@option{hi}:@option{lo}:@option{frac}.
2466 A description of the accepted options follows.
2470 Set the maximum number of consecutive frames which can be dropped (if
2471 positive), or the minimum interval between dropped frames (if
2472 negative). If the value is 0, the frame is dropped unregarding the
2473 number of previous sequentially dropped frames.
2480 Set the dropping threshold values.
2482 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
2483 represent actual pixel value differences, so a threshold of 64
2484 corresponds to 1 unit of difference for each pixel, or the same spread
2485 out differently over the block.
2487 A frame is a candidate for dropping if no 8x8 blocks differ by more
2488 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
2489 meaning the whole image) differ by more than a threshold of @option{lo}.
2491 Default value for @option{hi} is 64*12, default value for @option{lo} is
2492 64*5, and default value for @option{frac} is 0.33.
2497 Suppress a TV station logo by a simple interpolation of the surrounding
2498 pixels. Just set a rectangle covering the logo and watch it disappear
2499 (and sometimes something even uglier appear - your mileage may vary).
2501 This filter accepts the following options:
2505 Specify the top left corner coordinates of the logo. They must be
2509 Specify the width and height of the logo to clear. They must be
2513 Specify the thickness of the fuzzy edge of the rectangle (added to
2514 @var{w} and @var{h}). The default value is 4.
2517 When set to 1, a green rectangle is drawn on the screen to simplify
2518 finding the right @var{x}, @var{y}, @var{w}, @var{h} parameters, and
2519 @var{band} is set to 4. The default value is 0.
2523 @subsection Examples
2527 Set a rectangle covering the area with top left corner coordinates 0,0
2528 and size 100x77, setting a band of size 10:
2530 delogo=x=0:y=0:w=100:h=77:band=10
2537 Attempt to fix small changes in horizontal and/or vertical shift. This
2538 filter helps remove camera shake from hand-holding a camera, bumping a
2539 tripod, moving on a vehicle, etc.
2541 The filter accepts parameters as a list of @var{key}=@var{value}
2542 pairs, separated by ":". If the key of the first options is omitted,
2543 the arguments are interpreted according to the syntax
2544 @var{x}:@var{y}:@var{w}:@var{h}:@var{rx}:@var{ry}:@var{edge}:@var{blocksize}:@var{contrast}:@var{search}:@var{filename}:@var{opencl}.
2546 A description of the accepted parameters follows.
2551 Specify a rectangular area where to limit the search for motion
2553 If desired the search for motion vectors can be limited to a
2554 rectangular area of the frame defined by its top left corner, width
2555 and height. These parameters have the same meaning as the drawbox
2556 filter which can be used to visualise the position of the bounding
2559 This is useful when simultaneous movement of subjects within the frame
2560 might be confused for camera motion by the motion vector search.
2562 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
2563 then the full frame is used. This allows later options to be set
2564 without specifying the bounding box for the motion vector search.
2566 Default - search the whole frame.
2569 Specify the maximum extent of movement in x and y directions in the
2570 range 0-64 pixels. Default 16.
2573 Specify how to generate pixels to fill blanks at the edge of the
2574 frame. Available values are:
2577 Fill zeroes at blank locations
2579 Original image at blank locations
2581 Extruded edge value at blank locations
2583 Mirrored edge at blank locations
2585 Default value is @samp{mirror}.
2588 Specify the blocksize to use for motion search. Range 4-128 pixels,
2592 Specify the contrast threshold for blocks. Only blocks with more than
2593 the specified contrast (difference between darkest and lightest
2594 pixels) will be considered. Range 1-255, default 125.
2597 Specify the search strategy. Available values are:
2600 Set exhaustive search
2602 Set less exhaustive search.
2604 Default value is @samp{exhaustive}.
2607 If set then a detailed log of the motion search is written to the
2611 If set to 1, specify using OpenCL capabilities, only available if
2612 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
2618 Draw a colored box on the input image.
2620 This filter accepts the following options:
2624 Specify the top left corner coordinates of the box. Default to 0.
2628 Specify the width and height of the box, if 0 they are interpreted as
2629 the input width and height. Default to 0.
2632 Specify the color of the box to write, it can be the name of a color
2633 (case insensitive match) or a 0xRRGGBB[AA] sequence. If the special
2634 value @code{invert} is used, the box edge color is the same as the
2635 video with inverted luma.
2638 Set the thickness of the box edge. Default value is @code{4}.
2641 @subsection Examples
2645 Draw a black box around the edge of the input image:
2651 Draw a box with color red and an opacity of 50%:
2653 drawbox=10:20:200:60:red@@0.5
2656 The previous example can be specified as:
2658 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
2662 Fill the box with pink color:
2664 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
2671 Draw text string or text from specified file on top of video using the
2672 libfreetype library.
2674 To enable compilation of this filter you need to configure FFmpeg with
2675 @code{--enable-libfreetype}.
2679 The description of the accepted parameters follows.
2684 Used to draw a box around text using background color.
2685 Value should be either 1 (enable) or 0 (disable).
2686 The default value of @var{box} is 0.
2689 The color to be used for drawing box around text.
2690 Either a string (e.g. "yellow") or in 0xRRGGBB[AA] format
2691 (e.g. "0xff00ff"), possibly followed by an alpha specifier.
2692 The default value of @var{boxcolor} is "white".
2695 Set an expression which specifies if the text should be drawn. If the
2696 expression evaluates to 0, the text is not drawn. This is useful for
2697 specifying that the text should be drawn only when specific conditions
2700 Default value is "1".
2702 See below for the list of accepted constants and functions.
2705 Select how the @var{text} is expanded. Can be either @code{none},
2706 @code{strftime} (deprecated) or
2707 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
2711 If true, check and fix text coords to avoid clipping.
2714 The color to be used for drawing fonts.
2715 Either a string (e.g. "red") or in 0xRRGGBB[AA] format
2716 (e.g. "0xff000033"), possibly followed by an alpha specifier.
2717 The default value of @var{fontcolor} is "black".
2720 The font file to be used for drawing text. Path must be included.
2721 This parameter is mandatory.
2724 The font size to be used for drawing text.
2725 The default value of @var{fontsize} is 16.
2728 Flags to be used for loading the fonts.
2730 The flags map the corresponding flags supported by libfreetype, and are
2731 a combination of the following values:
2738 @item vertical_layout
2739 @item force_autohint
2742 @item ignore_global_advance_width
2744 @item ignore_transform
2751 Default value is "render".
2753 For more information consult the documentation for the FT_LOAD_*
2757 The color to be used for drawing a shadow behind the drawn text. It
2758 can be a color name (e.g. "yellow") or a string in the 0xRRGGBB[AA]
2759 form (e.g. "0xff00ff"), possibly followed by an alpha specifier.
2760 The default value of @var{shadowcolor} is "black".
2762 @item shadowx, shadowy
2763 The x and y offsets for the text shadow position with respect to the
2764 position of the text. They can be either positive or negative
2765 values. Default value for both is "0".
2768 The size in number of spaces to use for rendering the tab.
2772 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
2773 format. It can be used with or without text parameter. @var{timecode_rate}
2774 option must be specified.
2776 @item timecode_rate, rate, r
2777 Set the timecode frame rate (timecode only).
2780 The text string to be drawn. The text must be a sequence of UTF-8
2782 This parameter is mandatory if no file is specified with the parameter
2786 A text file containing text to be drawn. The text must be a sequence
2787 of UTF-8 encoded characters.
2789 This parameter is mandatory if no text string is specified with the
2790 parameter @var{text}.
2792 If both @var{text} and @var{textfile} are specified, an error is thrown.
2795 If set to 1, the @var{textfile} will be reloaded before each frame.
2796 Be sure to update it atomically, or it may be read partially, or even fail.
2799 The expressions which specify the offsets where text will be drawn
2800 within the video frame. They are relative to the top/left border of the
2803 The default value of @var{x} and @var{y} is "0".
2805 See below for the list of accepted constants and functions.
2808 The parameters for @var{x} and @var{y} are expressions containing the
2809 following constants and functions:
2813 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
2816 horizontal and vertical chroma subsample values. For example for the
2817 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
2820 the height of each text line
2828 @item max_glyph_a, ascent
2829 the maximum distance from the baseline to the highest/upper grid
2830 coordinate used to place a glyph outline point, for all the rendered
2832 It is a positive value, due to the grid's orientation with the Y axis
2835 @item max_glyph_d, descent
2836 the maximum distance from the baseline to the lowest grid coordinate
2837 used to place a glyph outline point, for all the rendered glyphs.
2838 This is a negative value, due to the grid's orientation, with the Y axis
2842 maximum glyph height, that is the maximum height for all the glyphs
2843 contained in the rendered text, it is equivalent to @var{ascent} -
2847 maximum glyph width, that is the maximum width for all the glyphs
2848 contained in the rendered text
2851 the number of input frame, starting from 0
2853 @item rand(min, max)
2854 return a random number included between @var{min} and @var{max}
2857 input sample aspect ratio
2860 timestamp expressed in seconds, NAN if the input timestamp is unknown
2863 the height of the rendered text
2866 the width of the rendered text
2869 the x and y offset coordinates where the text is drawn.
2871 These parameters allow the @var{x} and @var{y} expressions to refer
2872 each other, so you can for example specify @code{y=x/dar}.
2875 If libavfilter was built with @code{--enable-fontconfig}, then
2876 @option{fontfile} can be a fontconfig pattern or omitted.
2878 @anchor{drawtext_expansion}
2879 @subsection Text expansion
2881 If @option{expansion} is set to @code{strftime},
2882 the filter recognizes strftime() sequences in the provided text and
2883 expands them accordingly. Check the documentation of strftime(). This
2884 feature is deprecated.
2886 If @option{expansion} is set to @code{none}, the text is printed verbatim.
2888 If @option{expansion} is set to @code{normal} (which is the default),
2889 the following expansion mechanism is used.
2891 The backslash character '\', followed by any character, always expands to
2892 the second character.
2894 Sequence of the form @code{%@{...@}} are expanded. The text between the
2895 braces is a function name, possibly followed by arguments separated by ':'.
2896 If the arguments contain special characters or delimiters (':' or '@}'),
2897 they should be escaped.
2899 Note that they probably must also be escaped as the value for the
2900 @option{text} option in the filter argument string and as the filter
2901 argument in the filtergraph description, and possibly also for the shell,
2902 that makes up to four levels of escaping; using a text file avoids these
2905 The following functions are available:
2910 The expression evaluation result.
2912 It must take one argument specifying the expression to be evaluated,
2913 which accepts the same constants and functions as the @var{x} and
2914 @var{y} values. Note that not all constants should be used, for
2915 example the text size is not known when evaluating the expression, so
2916 the constants @var{text_w} and @var{text_h} will have an undefined
2920 The time at which the filter is running, expressed in UTC.
2921 It can accept an argument: a strftime() format string.
2924 The time at which the filter is running, expressed in the local time zone.
2925 It can accept an argument: a strftime() format string.
2928 The frame number, starting from 0.
2931 The timestamp of the current frame, in seconds, with microsecond accuracy.
2935 @subsection Examples
2939 Draw "Test Text" with font FreeSerif, using the default values for the
2940 optional parameters.
2943 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
2947 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
2948 and y=50 (counting from the top-left corner of the screen), text is
2949 yellow with a red box around it. Both the text and the box have an
2953 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
2954 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
2957 Note that the double quotes are not necessary if spaces are not used
2958 within the parameter list.
2961 Show the text at the center of the video frame:
2963 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h-line_h)/2"
2967 Show a text line sliding from right to left in the last row of the video
2968 frame. The file @file{LONG_LINE} is assumed to contain a single line
2971 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
2975 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
2977 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
2981 Draw a single green letter "g", at the center of the input video.
2982 The glyph baseline is placed at half screen height.
2984 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
2988 Show text for 1 second every 3 seconds:
2990 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:draw=lt(mod(t\,3)\,1):text='blink'"
2994 Use fontconfig to set the font. Note that the colons need to be escaped.
2996 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
3000 Print the date of a real-time encoding (see strftime(3)):
3002 drawtext='fontfile=FreeSans.ttf:text=%@{localtime:%a %b %d %Y@}'
3007 For more information about libfreetype, check:
3008 @url{http://www.freetype.org/}.
3010 For more information about fontconfig, check:
3011 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
3015 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
3017 This filter accepts the following optional named parameters:
3021 Set low and high threshold values used by the Canny thresholding
3024 The high threshold selects the "strong" edge pixels, which are then
3025 connected through 8-connectivity with the "weak" edge pixels selected
3026 by the low threshold.
3028 @var{low} and @var{high} threshold values must be choosen in the range
3029 [0,1], and @var{low} should be lesser or equal to @var{high}.
3031 Default value for @var{low} is @code{20/255}, and default value for @var{high}
3037 edgedetect=low=0.1:high=0.4
3042 Apply fade-in/out effect to input video.
3044 This filter accepts the following options:
3048 The effect type -- can be either "in" for fade-in, or "out" for a fade-out
3050 Default is @code{in}.
3052 @item start_frame, s
3053 Specify the number of the start frame for starting to apply the fade
3054 effect. Default is 0.
3057 The number of frames for which the fade effect has to last. At the end of the
3058 fade-in effect the output video will have the same intensity as the input video,
3059 at the end of the fade-out transition the output video will be completely black.
3063 If set to 1, fade only alpha channel, if one exists on the input.
3067 @subsection Examples
3071 Fade in first 30 frames of video:
3076 The command above is equivalent to:
3082 Fade out last 45 frames of a 200-frame video:
3085 fade=type=out:start_frame=155:nb_frames=45
3089 Fade in first 25 frames and fade out last 25 frames of a 1000-frame video:
3091 fade=in:0:25, fade=out:975:25
3095 Make first 5 frames black, then fade in from frame 5-24:
3101 Fade in alpha over first 25 frames of video:
3103 fade=in:0:25:alpha=1
3109 Extract a single field from an interlaced image using stride
3110 arithmetic to avoid wasting CPU time. The output frames are marked as
3113 This filter accepts the following named options:
3116 Specify whether to extract the top (if the value is @code{0} or
3117 @code{top}) or the bottom field (if the value is @code{1} or
3121 If the option key is not specified, the first value sets the @var{type}
3122 option. For example:
3134 Transform the field order of the input video.
3136 This filter accepts the following options:
3141 Output field order. Valid values are @var{tff} for top field first or @var{bff}
3142 for bottom field first.
3145 Default value is @samp{tff}.
3147 Transformation is achieved by shifting the picture content up or down
3148 by one line, and filling the remaining line with appropriate picture content.
3149 This method is consistent with most broadcast field order converters.
3151 If the input video is not flagged as being interlaced, or it is already
3152 flagged as being of the required output field order then this filter does
3153 not alter the incoming video.
3155 This filter is very useful when converting to or from PAL DV material,
3156 which is bottom field first.
3160 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
3165 Buffer input images and send them when they are requested.
3167 This filter is mainly useful when auto-inserted by the libavfilter
3170 The filter does not take parameters.
3175 Convert the input video to one of the specified pixel formats.
3176 Libavfilter will try to pick one that is supported for the input to
3179 This filter accepts the following parameters:
3183 A '|'-separated list of pixel format names, for example
3184 "pix_fmts=yuv420p|monow|rgb24".
3188 @subsection Examples
3192 Convert the input video to the format @var{yuv420p}
3194 format=pix_fmts=yuv420p
3197 Convert the input video to any of the formats in the list
3199 format=pix_fmts=yuv420p|yuv444p|yuv410p
3205 Convert the video to specified constant frame rate by duplicating or dropping
3206 frames as necessary.
3208 This filter accepts the following named parameters:
3212 Desired output frame rate. The default is @code{25}.
3217 Possible values are:
3220 zero round towards 0
3224 round towards -infinity
3226 round towards +infinity
3230 The default is @code{near}.
3234 Alternatively, the options can be specified as a flat string:
3235 @var{fps}[:@var{round}].
3237 See also the @ref{setpts} filter.
3241 Select one frame every N.
3243 This filter accepts in input a string representing a positive
3244 integer. Default argument is @code{1}.
3249 Apply a frei0r effect to the input video.
3251 To enable compilation of this filter you need to install the frei0r
3252 header and configure FFmpeg with @code{--enable-frei0r}.
3254 This filter accepts the following options:
3259 The name to the frei0r effect to load. If the environment variable
3260 @env{FREI0R_PATH} is defined, the frei0r effect is searched in each one of the
3261 directories specified by the colon separated list in @env{FREIOR_PATH},
3262 otherwise in the standard frei0r paths, which are in this order:
3263 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
3264 @file{/usr/lib/frei0r-1/}.
3267 A '|'-separated list of parameters to pass to the frei0r effect.
3271 A frei0r effect parameter can be a boolean (whose values are specified
3272 with "y" and "n"), a double, a color (specified by the syntax
3273 @var{R}/@var{G}/@var{B}, @var{R}, @var{G}, and @var{B} being float
3274 numbers from 0.0 to 1.0) or by an @code{av_parse_color()} color
3275 description), a position (specified by the syntax @var{X}/@var{Y},
3276 @var{X} and @var{Y} being float numbers) and a string.
3278 The number and kind of parameters depend on the loaded effect. If an
3279 effect parameter is not specified the default value is set.
3281 @subsection Examples
3285 Apply the distort0r effect, set the first two double parameters:
3287 frei0r=filter_name=distort0r:filter_params=0.5|0.01
3291 Apply the colordistance effect, take a color as first parameter:
3293 frei0r=colordistance:0.2/0.3/0.4
3294 frei0r=colordistance:violet
3295 frei0r=colordistance:0x112233
3299 Apply the perspective effect, specify the top left and top right image
3302 frei0r=perspective:0.2/0.2|0.8/0.2
3306 For more information see:
3307 @url{http://frei0r.dyne.org}
3311 The filter takes one, two, three or four equations as parameter, separated by ':'.
3312 The first equation is mandatory and applies to the luma plane. The two
3313 following are respectively for chroma blue and chroma red planes.
3315 The filter syntax allows named parameters:
3319 the luminance expression
3321 the chrominance blue expression
3323 the chrominance red expression
3325 the alpha expression
3328 If one of the chrominance expression is not defined, it falls back on the other
3329 one. If no alpha expression is specified it will evaluate to opaque value.
3330 If none of chrominance expressions are
3331 specified, they will evaluate the luminance expression.
3333 The expressions can use the following variables and functions:
3337 The sequential number of the filtered frame, starting from @code{0}.
3341 The coordinates of the current sample.
3345 The width and height of the image.
3349 Width and height scale depending on the currently filtered plane. It is the
3350 ratio between the corresponding luma plane number of pixels and the current
3351 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
3352 @code{0.5,0.5} for chroma planes.
3355 Time of the current frame, expressed in seconds.
3358 Return the value of the pixel at location (@var{x},@var{y}) of the current
3362 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
3366 Return the value of the pixel at location (@var{x},@var{y}) of the
3367 blue-difference chroma plane. Returns 0 if there is no such plane.
3370 Return the value of the pixel at location (@var{x},@var{y}) of the
3371 red-difference chroma plane. Returns 0 if there is no such plane.
3374 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
3375 plane. Returns 0 if there is no such plane.
3378 For functions, if @var{x} and @var{y} are outside the area, the value will be
3379 automatically clipped to the closer edge.
3381 @subsection Examples
3385 Flip the image horizontally:
3391 Generate a bidimensional sine wave, with angle @code{PI/3} and a
3392 wavelength of 100 pixels:
3394 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
3398 Generate a fancy enigmatic moving light:
3400 nullsrc=s=256x256,geq=random(1)/hypot(X-cos(N*0.07)*W/2-W/2\,Y-sin(N*0.09)*H/2-H/2)^2*1000000*sin(N*0.02):128:128
3406 Fix the banding artifacts that are sometimes introduced into nearly flat
3407 regions by truncation to 8bit color depth.
3408 Interpolate the gradients that should go where the bands are, and
3411 This filter is designed for playback only. Do not use it prior to
3412 lossy compression, because compression tends to lose the dither and
3413 bring back the bands.
3415 This filter accepts the following options:
3420 The maximum amount by which the filter will change any one pixel. Also the
3421 threshold for detecting nearly flat regions. Acceptable values range from .51 to
3422 64, default value is 1.2, out-of-range values will be clipped to the valid
3426 The neighborhood to fit the gradient to. A larger radius makes for smoother
3427 gradients, but also prevents the filter from modifying the pixels near detailed
3428 regions. Acceptable values are 8-32, default value is 16, out-of-range values
3429 will be clipped to the valid range.
3433 Alternatively, the options can be specified as a flat string:
3434 @var{strength}[:@var{radius}]
3436 @subsection Examples
3440 Apply the filter with a @code{3.5} strength and radius of @code{8}:
3446 Specify radius, omitting the strength (which will fall-back to the default
3456 Flip the input video horizontally.
3458 For example to horizontally flip the input video with @command{ffmpeg}:
3460 ffmpeg -i in.avi -vf "hflip" out.avi
3464 This filter applies a global color histogram equalization on a
3467 It can be used to correct video that has a compressed range of pixel
3468 intensities. The filter redistributes the pixel intensities to
3469 equalize their distribution across the intensity range. It may be
3470 viewed as an "automatically adjusting contrast filter". This filter is
3471 useful only for correcting degraded or poorly captured source
3474 The filter accepts parameters as a list of @var{key}=@var{value}
3475 pairs, separated by ":". If the key of the first options is omitted,
3476 the arguments are interpreted according to syntax
3477 @var{strength}:@var{intensity}:@var{antibanding}.
3479 This filter accepts the following named options:
3483 Determine the amount of equalization to be applied. As the strength
3484 is reduced, the distribution of pixel intensities more-and-more
3485 approaches that of the input frame. The value must be a float number
3486 in the range [0,1] and defaults to 0.200.
3489 Set the maximum intensity that can generated and scale the output
3490 values appropriately. The strength should be set as desired and then
3491 the intensity can be limited if needed to avoid washing-out. The value
3492 must be a float number in the range [0,1] and defaults to 0.210.
3495 Set the antibanding level. If enabled the filter will randomly vary
3496 the luminance of output pixels by a small amount to avoid banding of
3497 the histogram. Possible values are @code{none}, @code{weak} or
3498 @code{strong}. It defaults to @code{none}.
3503 Compute and draw a color distribution histogram for the input video.
3505 The computed histogram is a representation of distribution of color components
3508 The filter accepts the following named parameters:
3514 It accepts the following values:
3517 standard histogram that display color components distribution in an image.
3518 Displays color graph for each color component. Shows distribution
3519 of the Y, U, V, A or G, B, R components, depending on input format,
3520 in current frame. Bellow each graph is color component scale meter.
3523 chroma values in vectorscope, if brighter more such chroma values are
3524 distributed in an image.
3525 Displays chroma values (U/V color placement) in two dimensional graph
3526 (which is called a vectorscope). It can be used to read of the hue and
3527 saturation of the current frame. At a same time it is a histogram.
3528 The whiter a pixel in the vectorscope, the more pixels of the input frame
3529 correspond to that pixel (that is the more pixels have this chroma value).
3530 The V component is displayed on the horizontal (X) axis, with the leftmost
3531 side being V = 0 and the rightmost side being V = 255.
3532 The U component is displayed on the vertical (Y) axis, with the top
3533 representing U = 0 and the bottom representing U = 255.
3535 The position of a white pixel in the graph corresponds to the chroma value
3536 of a pixel of the input clip. So the graph can be used to read of the
3537 hue (color flavor) and the saturation (the dominance of the hue in the color).
3538 As the hue of a color changes, it moves around the square. At the center of
3539 the square, the saturation is zero, which means that the corresponding pixel
3540 has no color. If you increase the amount of a specific color, while leaving
3541 the other colors unchanged, the saturation increases, and you move towards
3542 the edge of the square.
3545 chroma values in vectorscope, similar as @code{color} but actual chroma values
3549 per row/column color component graph. In row mode graph in the left side represents
3550 color component value 0 and right side represents value = 255. In column mode top
3551 side represents color component value = 0 and bottom side represents value = 255.
3553 Default value is @code{levels}.
3556 Set height of level in @code{levels}. Default value is @code{200}.
3557 Allowed range is [50, 2048].
3560 Set height of color scale in @code{levels}. Default value is @code{12}.
3561 Allowed range is [0, 40].
3564 Set step for @code{waveform} mode. Smaller values are useful to find out how much
3565 of same luminance values across input rows/columns are distributed.
3566 Default value is @code{10}. Allowed range is [1, 255].
3569 Set mode for @code{waveform}. Can be either @code{row}, or @code{column}.
3570 Default is @code{row}.
3573 Set display mode for @code{waveform} and @code{levels}.
3574 It accepts the following values:
3577 Display separate graph for the color components side by side in
3578 @code{row} waveform mode or one below other in @code{column} waveform mode
3579 for @code{waveform} histogram mode. For @code{levels} histogram mode
3580 per color component graphs are placed one bellow other.
3582 This display mode in @code{waveform} histogram mode makes it easy to spot
3583 color casts in the highlights and shadows of an image, by comparing the
3584 contours of the top and the bottom of each waveform.
3585 Since whites, grays, and blacks are characterized by
3586 exactly equal amounts of red, green, and blue, neutral areas of the
3587 picture should display three waveforms of roughly equal width/height.
3588 If not, the correction is easy to make by making adjustments to level the
3592 Presents information that's identical to that in the @code{parade}, except
3593 that the graphs representing color components are superimposed directly
3596 This display mode in @code{waveform} histogram mode can make it easier to spot
3597 the relative differences or similarities in overlapping areas of the color
3598 components that are supposed to be identical, such as neutral whites, grays,
3601 Default is @code{parade}.
3604 @subsection Examples
3609 Calculate and draw histogram:
3611 ffplay -i input -vf histogram
3618 High precision/quality 3d denoise filter. This filter aims to reduce
3619 image noise producing smooth images and making still images really
3620 still. It should enhance compressibility.
3622 It accepts the following optional parameters:
3626 a non-negative float number which specifies spatial luma strength,
3629 @item chroma_spatial
3630 a non-negative float number which specifies spatial chroma strength,
3631 defaults to 3.0*@var{luma_spatial}/4.0
3634 a float number which specifies luma temporal strength, defaults to
3635 6.0*@var{luma_spatial}/4.0
3638 a float number which specifies chroma temporal strength, defaults to
3639 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}
3644 Modify the hue and/or the saturation of the input.
3646 This filter accepts the following optional named options:
3650 Specify the hue angle as a number of degrees. It accepts a float
3651 number or an expression, and defaults to 0.0.
3654 Specify the hue angle as a number of radians. It accepts a float
3655 number or an expression, and defaults to 0.0.
3658 Specify the saturation in the [-10,10] range. It accepts a float number and
3662 The @var{h}, @var{H} and @var{s} parameters are expressions containing the
3663 following constants:
3667 frame count of the input frame starting from 0
3670 presentation timestamp of the input frame expressed in time base units
3673 frame rate of the input video, NAN if the input frame rate is unknown
3676 timestamp expressed in seconds, NAN if the input timestamp is unknown
3679 time base of the input video
3682 The options can also be set using the syntax: @var{hue}:@var{saturation}
3684 In this case @var{hue} is expressed in degrees.
3686 @subsection Examples
3690 Set the hue to 90 degrees and the saturation to 1.0:
3696 Same command but expressing the hue in radians:
3702 Same command without named options, hue must be expressed in degrees:
3708 Note that "h:s" syntax does not support expressions for the values of
3709 h and s, so the following example will issue an error:
3715 Rotate hue and make the saturation swing between 0
3716 and 2 over a period of 1 second:
3718 hue="H=2*PI*t: s=sin(2*PI*t)+1"
3722 Apply a 3 seconds saturation fade-in effect starting at 0:
3727 The general fade-in expression can be written as:
3729 hue="s=min(0\, max((t-START)/DURATION\, 1))"
3733 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
3735 hue="s=max(0\, min(1\, (8-t)/3))"
3738 The general fade-out expression can be written as:
3740 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
3745 @subsection Commands
3747 This filter supports the following command:
3750 Modify the hue and/or the saturation of the input video.
3751 The command accepts the same named options and syntax than when calling the
3752 filter from the command-line.
3754 If a parameter is omitted, it is kept at its current value.
3759 Detect video interlacing type.
3761 This filter tries to detect if the input is interlaced or progressive,
3762 top or bottom field first.
3766 Deinterleave or interleave fields.
3768 This filter allows to process interlaced images fields without
3769 deinterlacing them. Deinterleaving splits the input frame into 2
3770 fields (so called half pictures). Odd lines are moved to the top
3771 half of the output image, even lines to the bottom half.
3772 You can process (filter) them independently and then re-interleave them.
3774 It accepts a list of options in the form of @var{key}=@var{value} pairs
3775 separated by ":". A description of the accepted options follows.
3779 @item chroma_mode, s
3781 Available values for @var{luma_mode}, @var{chroma_mode} and
3782 @var{alpha_mode} are:
3788 @item deinterleave, d
3789 Deinterleave fields, placing one above the other.
3792 Interleave fields. Reverse the effect of deinterleaving.
3794 Default value is @code{none}.
3797 @item chroma_swap, cs
3798 @item alpha_swap, as
3799 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
3804 Deinterlace input video by applying Donald Graft's adaptive kernel
3805 deinterling. Work on interlaced parts of a video to produce
3808 This filter accepts parameters as a list of @var{key}=@var{value}
3809 pairs, separated by ":". If the key of the first options is omitted,
3810 the arguments are interpreted according to the following syntax:
3811 @var{thresh}:@var{map}:@var{order}:@var{sharp}:@var{twoway}.
3813 The description of the accepted parameters follows.
3817 Set the threshold which affects the filter's tolerance when
3818 determining if a pixel line must be processed. It must be an integer
3819 in the range [0,255] and defaults to 10. A value of 0 will result in
3820 applying the process on every pixels.
3823 Paint pixels exceeding the threshold value to white if set to 1.
3827 Set the fields order. Swap fields if set to 1, leave fields alone if
3831 Enable additional sharpening if set to 1. Default is 0.
3834 Enable twoway sharpening if set to 1. Default is 0.
3837 @subsection Examples
3841 Apply default values:
3843 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
3847 Enable additional sharpening:
3853 Paint processed pixels in white:
3859 @section lut, lutrgb, lutyuv
3861 Compute a look-up table for binding each pixel component input value
3862 to an output value, and apply it to input video.
3864 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
3865 to an RGB input video.
3867 These filters accept in input a ":"-separated list of options, which
3868 specify the expressions used for computing the lookup table for the
3869 corresponding pixel component values.
3871 The @var{lut} filter requires either YUV or RGB pixel formats in
3872 input, and accepts the options:
3875 set first pixel component expression
3877 set second pixel component expression
3879 set third pixel component expression
3881 set fourth pixel component expression, corresponds to the alpha component
3884 The exact component associated to each option depends on the format in
3887 The @var{lutrgb} filter requires RGB pixel formats in input, and
3888 accepts the options:
3891 set red component expression
3893 set green component expression
3895 set blue component expression
3897 alpha component expression
3900 The @var{lutyuv} filter requires YUV pixel formats in input, and
3901 accepts the options:
3904 set Y/luminance component expression
3906 set U/Cb component expression
3908 set V/Cr component expression
3910 set alpha component expression
3913 The expressions can contain the following constants and functions:
3917 the input width and height
3920 input value for the pixel component
3923 the input value clipped in the @var{minval}-@var{maxval} range
3926 maximum value for the pixel component
3929 minimum value for the pixel component
3932 the negated value for the pixel component value clipped in the
3933 @var{minval}-@var{maxval} range , it corresponds to the expression
3934 "maxval-clipval+minval"
3937 the computed value in @var{val} clipped in the
3938 @var{minval}-@var{maxval} range
3940 @item gammaval(gamma)
3941 the computed gamma correction value of the pixel component value
3942 clipped in the @var{minval}-@var{maxval} range, corresponds to the
3944 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
3948 All expressions default to "val".
3950 @subsection Examples
3956 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
3957 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
3960 The above is the same as:
3962 lutrgb="r=negval:g=negval:b=negval"
3963 lutyuv="y=negval:u=negval:v=negval"
3973 Remove chroma components, turns the video into a graytone image:
3975 lutyuv="u=128:v=128"
3979 Apply a luma burning effect:
3985 Remove green and blue components:
3991 Set a constant alpha channel value on input:
3993 format=rgba,lutrgb=a="maxval-minval/2"
3997 Correct luminance gamma by a 0.5 factor:
3999 lutyuv=y=gammaval(0.5)
4003 Discard least significant bits of luma:
4005 lutyuv=y='bitand(val, 128+64+32)'
4011 Apply an MPlayer filter to the input video.
4013 This filter provides a wrapper around most of the filters of
4016 This wrapper is considered experimental. Some of the wrapped filters
4017 may not work properly and we may drop support for them, as they will
4018 be implemented natively into FFmpeg. Thus you should avoid
4019 depending on them when writing portable scripts.
4021 The filters accepts the parameters:
4022 @var{filter_name}[:=]@var{filter_params}
4024 @var{filter_name} is the name of a supported MPlayer filter,
4025 @var{filter_params} is a string containing the parameters accepted by
4028 The list of the currently supported filters follows:
4055 The parameter syntax and behavior for the listed filters are the same
4056 of the corresponding MPlayer filters. For detailed instructions check
4057 the "VIDEO FILTERS" section in the MPlayer manual.
4059 @subsection Examples
4063 Adjust gamma, brightness, contrast:
4069 See also mplayer(1), @url{http://www.mplayerhq.hu/}.
4075 This filter accepts an integer in input, if non-zero it negates the
4076 alpha component (if available). The default value in input is 0.
4080 Force libavfilter not to use any of the specified pixel formats for the
4081 input to the next filter.
4083 This filter accepts the following parameters:
4087 A '|'-separated list of pixel format names, for example
4088 "pix_fmts=yuv420p|monow|rgb24".
4092 @subsection Examples
4096 Force libavfilter to use a format different from @var{yuv420p} for the
4097 input to the vflip filter:
4099 noformat=pix_fmts=yuv420p,vflip
4103 Convert the input video to any of the formats not contained in the list:
4105 noformat=yuv420p|yuv444p|yuv410p
4111 Add noise on video input frame.
4113 This filter accepts a list of options in the form of @var{key}=@var{value}
4114 pairs separated by ":". A description of the accepted options follows.
4122 Set noise seed for specific pixel component or all pixel components in case
4123 of @var{all_seed}. Default value is @code{123457}.
4125 @item all_strength, alls
4126 @item c0_strength, c0s
4127 @item c1_strength, c1s
4128 @item c2_strength, c2s
4129 @item c3_strength, c3s
4130 Set noise strength for specific pixel component or all pixel components in case
4131 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
4133 @item all_flags, allf
4138 Set pixel component flags or set flags for all components if @var{all_flags}.
4139 Available values for component flags are:
4142 averaged temporal noise (smoother)
4144 mix random noise with a (semi)regular pattern
4146 higher quality (slightly better looking, slightly slower)
4148 temporal noise (noise pattern changes between frames)
4150 uniform noise (gaussian otherwise)
4154 @subsection Examples
4156 Add temporal and uniform noise to input video:
4158 noise=alls=20:allf=t+u
4163 Pass the video source unchanged to the output.
4167 Apply video transform using libopencv.
4169 To enable this filter install libopencv library and headers and
4170 configure FFmpeg with @code{--enable-libopencv}.
4172 This filter accepts the following parameters:
4177 The name of the libopencv filter to apply.
4180 The parameters to pass to the libopencv filter. If not specified the default
4185 Refer to the official libopencv documentation for more precise
4187 @url{http://opencv.willowgarage.com/documentation/c/image_filtering.html}
4189 Follows the list of supported libopencv filters.
4194 Dilate an image by using a specific structuring element.
4195 This filter corresponds to the libopencv function @code{cvDilate}.
4197 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
4199 @var{struct_el} represents a structuring element, and has the syntax:
4200 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
4202 @var{cols} and @var{rows} represent the number of columns and rows of
4203 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
4204 point, and @var{shape} the shape for the structuring element, and
4205 can be one of the values "rect", "cross", "ellipse", "custom".
4207 If the value for @var{shape} is "custom", it must be followed by a
4208 string of the form "=@var{filename}". The file with name
4209 @var{filename} is assumed to represent a binary image, with each
4210 printable character corresponding to a bright pixel. When a custom
4211 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
4212 or columns and rows of the read file are assumed instead.
4214 The default value for @var{struct_el} is "3x3+0x0/rect".
4216 @var{nb_iterations} specifies the number of times the transform is
4217 applied to the image, and defaults to 1.
4219 Follow some example:
4221 # use the default values
4224 # dilate using a structuring element with a 5x5 cross, iterate two times
4225 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
4227 # read the shape from the file diamond.shape, iterate two times
4228 # the file diamond.shape may contain a pattern of characters like this:
4234 # the specified cols and rows are ignored (but not the anchor point coordinates)
4235 ocv=dilate:0x0+2x2/custom=diamond.shape|2