97 #define OFFSET(x) offsetof(RemovelogoContext, x) 98 #define FLAGS AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM 117 #define apply_mask_fudge_factor(x) (((x) >> 2) + (x)) 134 int w,
int h,
int min_val,
141 int current_pass = 0;
144 for (y = 0; y <
h; y++)
145 for (x = 0; x <
w; x++)
146 data[y*linesize + x] = data[y*linesize + x] > min_val;
156 int has_anything_changed = 0;
157 uint8_t *current_pixel0 = data + 1 + linesize, *current_pixel;
160 for (y = 1; y < h-1; y++) {
161 current_pixel = current_pixel0;
162 for (x = 1; x < w-1; x++) {
174 if ( *current_pixel >= current_pass &&
175 *(current_pixel + 1) >= current_pass &&
176 *(current_pixel - 1) >= current_pass &&
177 *(current_pixel + linesize) >= current_pass &&
178 *(current_pixel - linesize) >= current_pass) {
183 has_anything_changed = 1;
187 current_pixel0 += linesize;
189 if (!has_anything_changed)
195 for (y = 1; y < h - 1; y++)
196 for (x = 1; x < w - 1; x++)
216 const char *
filename,
void *log_ctx)
220 uint8_t *src_data[4], *gray_data[4];
221 int src_linesize[4], gray_linesize[4];
224 if ((ret =
ff_load_image(src_data, src_linesize, w, h, &pix_fmt, filename, log_ctx)) < 0)
229 src_data, src_linesize, *w, *h, pix_fmt,
257 uint8_t *dst_data,
int dst_linesize,
258 int src_w,
int src_h,
265 for (y = 0; y < src_h/2; y++) {
266 for (x = 0; x < src_w/2; x++) {
269 dst_data[(y * dst_linesize) + x] =
270 src_data[((y << 1) * src_linesize) + (x << 1)] ||
271 src_data[((y << 1) * src_linesize) + (x << 1) + 1] ||
272 src_data[(((y << 1) + 1) * src_linesize) + (x << 1)] ||
273 src_data[(((y << 1) + 1) * src_linesize) + (x << 1) + 1];
274 dst_data[(y * dst_linesize) + x] =
FFMIN(1, dst_data[(y * dst_linesize) + x]);
279 src_w/2, src_h/2, 0, max_mask_size);
288 int full_max_mask_size, half_max_mask_size;
302 16, &full_max_mask_size);
309 w, h, &half_max_mask_size);
327 for (b = -a; b <=
a; b++) {
329 if (!mask[a][b + a]) {
333 for (c = -a; c <=
a; c++) {
334 if ((b * b) + (c * c) <= (a * a))
335 mask[
a][b +
a][c +
a] = 1;
337 mask[
a][b +
a][c +
a] = 0;
348 #define SHOW_LOGO_INFO(mask_type) \ 349 av_log(ctx, AV_LOG_VERBOSE, #mask_type " x1:%d x2:%d y1:%d y2:%d max_mask_size:%d\n", \ 350 s->mask_type##_mask_bbox.x1, s->mask_type##_mask_bbox.x2, \ 351 s->mask_type##_mask_bbox.y1, s->mask_type##_mask_bbox.y2, \ 352 mask_type##_max_mask_size); 366 "Mask image size %dx%d does not match with the input video size %dx%d\n",
389 const uint8_t *mask_data,
int mask_linesize,
390 uint8_t *image_data,
int image_linesize,
391 int w,
int h,
int x,
int y)
396 int start_posx, start_posy, end_posx, end_posy;
400 const uint8_t *image_read_position;
402 const uint8_t *mask_read_position;
405 mask_size = mask_data[y * mask_linesize + x];
406 start_posx =
FFMAX(0, x - mask_size);
407 start_posy =
FFMAX(0, y - mask_size);
408 end_posx =
FFMIN(w - 1, x + mask_size);
409 end_posy =
FFMIN(h - 1, y + mask_size);
411 image_read_position = image_data + image_linesize * start_posy + start_posx;
412 mask_read_position = mask_data + mask_linesize * start_posy + start_posx;
414 for (j = start_posy; j <= end_posy; j++) {
415 for (i = start_posx; i <= end_posx; i++) {
418 if (!(*mask_read_position) && mask[mask_size][i - start_posx][j - start_posy]) {
419 accumulator += *image_read_position;
423 image_read_position++;
424 mask_read_position++;
427 image_read_position += (image_linesize - ((end_posx + 1) - start_posx));
428 mask_read_position += (mask_linesize - ((end_posx + 1) - start_posx));
434 return divisor == 0 ? 255:
435 (accumulator + (divisor / 2)) / divisor;
462 const uint8_t *src_data,
int src_linesize,
463 uint8_t *dst_data,
int dst_linesize,
464 const uint8_t *mask_data,
int mask_linesize,
475 for (y = bbox->
y1; y <= bbox->y2; y++) {
476 src_line = src_data + src_linesize * y;
477 dst_line = dst_data + dst_linesize * y;
479 for (x = bbox->
x1; x <= bbox->x2; x++) {
480 if (mask_data[y * mask_linesize + x]) {
483 mask_data, mask_linesize,
484 dst_data, dst_linesize,
489 dst_line[x] = src_line[x];
504 outpicref = inpicref;
548 for (b = -a; b <=
a; b++) {
577 .
name =
"removelogo",
583 .
inputs = removelogo_inputs,
585 .priv_class = &removelogo_class,
static enum AVPixelFormat pix_fmt
This code implements a filter to remove annoying TV logos and other annoying images placed onto a vid...
This structure describes decoded (raw) audio or video data.
ptrdiff_t const GLvoid * data
AVFILTER_DEFINE_CLASS(removelogo)
Main libavfilter public API header.
int h
agreed upon image height
Miscellaneous utilities which make use of the libswscale library.
The reader does not expect b to be semantically here and if the code is changed by maybe adding a a division or other the signedness will almost certainly be mistaken To avoid this confusion a new type was SUINT is the C unsigned type but it holds a signed int to use the same example SUINT a
static void convert_mask_to_strength_mask(uint8_t *data, int linesize, int w, int h, int min_val, int *max_mask_size)
Pre-process an image to give distance information.
AVFilter ff_vf_removelogo
void * av_mallocz(size_t size)
Allocate a memory block with alignment suitable for all memory accesses (including vectors if availab...
AVFrame * ff_get_video_buffer(AVFilterLink *link, int w, int h)
Request a picture buffer with a specific set of permissions.
static int filter_frame(AVFilterLink *inlink, AVFrame *inpicref)
static unsigned int blur_pixel(int ***mask, const uint8_t *mask_data, int mask_linesize, uint8_t *image_data, int image_linesize, int w, int h, int x, int y)
Blur image.
#define SHOW_LOGO_INFO(mask_type)
#define AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC
Some filters support a generic "enable" expression option that can be used to enable or disable a fil...
const char * name
Pad name.
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
static av_cold void uninit(AVFilterContext *ctx)
Undefined Behavior In the C some operations are like signed integer dereferencing freed accessing outside allocated Undefined Behavior must not occur in a C it is not safe even if the output of undefined operations is unused The unsafety may seem nit picking but Optimizing compilers have in fact optimized code on the assumption that no undefined Behavior occurs Optimizing code based on wrong assumptions can and has in some cases lead to effects beyond the output of computations The signed integer overflow problem in speed critical code Code which is highly optimized and works with signed integers sometimes has the problem that often the output of the computation does not c
int ff_scale_image(uint8_t *dst_data[4], int dst_linesize[4], int dst_w, int dst_h, enum AVPixelFormat dst_pix_fmt, uint8_t *const src_data[4], int src_linesize[4], int src_w, int src_h, enum AVPixelFormat src_pix_fmt, void *log_ctx)
Scale image using libswscale.
FFBoundingBox full_mask_bbox
A filter pad used for either input or output.
A link between two filters.
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
void * priv
private data for use by the filter
#define apply_mask_fudge_factor(x)
Choose a slightly larger mask size to improve performance.
static void direct(const float *in, const FFTComplex *ir, int len, float *out)
static const AVFilterPad removelogo_outputs[]
static av_cold int init(AVFilterContext *ctx)
int w
agreed upon image width
static const AVOption removelogo_options[]
static void generate_half_size_image(const uint8_t *src_data, int src_linesize, uint8_t *dst_data, int dst_linesize, int src_w, int src_h, int *max_mask_size)
Generate a scaled down image with half width, height, and intensity.
int ff_load_image(uint8_t *data[4], int linesize[4], int *w, int *h, enum AVPixelFormat *pix_fmt, const char *filename, void *log_ctx)
Load image from filename and put the resulting image in data.
static void blur_image(int ***mask, const uint8_t *src_data, int src_linesize, uint8_t *dst_data, int dst_linesize, const uint8_t *mask_data, int mask_linesize, int w, int h, int direct, FFBoundingBox *bbox)
Blur image plane using a mask.
static const AVFilterPad outputs[]
#define AV_LOG_INFO
Standard information.
int av_frame_is_writable(AVFrame *frame)
Check if the frame data is writable.
static const AVFilterPad removelogo_inputs[]
int linesize[AV_NUM_DATA_POINTERS]
For video, size in bytes of each picture line.
these buffered frames must be flushed immediately if a new input produces new the filter must not call request_frame to get more It must just process the frame or queue it The task of requesting more frames is left to the filter s request_frame method or the application If a filter has several inputs
Describe the class of an AVClass context structure.
const char * name
Filter name.
int ff_calculate_bounding_box(FFBoundingBox *bbox, const uint8_t *data, int linesize, int w, int h, int min_val, int depth)
Calculate the smallest rectangle that will encompass the region with values > min_val.
static int config_props_input(AVFilterLink *inlink)
AVFilterLink ** outputs
array of pointers to output links
static enum AVPixelFormat pix_fmts[]
#define flags(name, subs,...)
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
The exact code depends on how similar the blocks are and how related they are to the and needs to apply these operations to the correct inlink or outlink if there are several Macros are available to factor that when no extra processing is inlink
planar YUV 4:2:0, 12bpp, (1 Cr & Cb sample per 2x2 Y samples)
static int load_mask(uint8_t **mask, int *w, int *h, const char *filename, void *log_ctx)
AVFilterContext * dst
dest filter
static uint8_t half(int a, int b)
static int query_formats(AVFilterContext *ctx)
#define av_malloc_array(a, b)
void av_image_copy_plane(uint8_t *dst, int dst_linesize, const uint8_t *src, int src_linesize, int bytewidth, int height)
Copy image plane from src to dst.
FFBoundingBox half_mask_bbox
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later.That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another.Frame references ownership and permissions
AVPixelFormat
Pixel format.
int av_frame_copy_props(AVFrame *dst, const AVFrame *src)
Copy only "metadata" fields from src to dst.
Miscellaneous utilities which make use of the libavformat library.