Logo presence detection based on blending characteristics
View Patent ↗A video processing system detects an overlay image, such as a logo, in a picture of a video stream, the overlay for example being a broadcaster's logo. The detection is based on evaluation of blending characteristics of a picture frame. The method of detection of an overlay defines first and second areas within the image, the first and second areas being non-overlapping. Next an alpha-blended value is calculated for the mean color value of the second area with an overlay color value. Then, if the mean color value of the first area is closer to the alpha-blended value than it is to the mean color value of the second area, the overlay can be indicated as detected and defined within the picture. Detection of the overlay can be used to identify an owner of the video, or detect when a scene change such as a commercial occurs.
1. A method for detecting overlay graphic presence within a frame of image data comprising:
specifying a first location corresponding to a pixel location within the frame of image data;
estimating a background pixel value for the first location, thereby providing an estimated background pixel value, wherein the estimated background pixel value is responsive to one or more image pixel values corresponding respectively to one or more pixel locations that are spatially separate from the first location, and within the frame of image data;
estimating a blended pixel value for the first location, thereby providing an estimated blended pixel value;
receiving a first image pixel value for the first location;
evaluating a graphic presence criterion corresponding to the first location to determine presence of the graphic corresponding to the first location, responsive to:
the estimated background pixel value,
the estimated blended pixel value, and
the first image pixel value,
to provide a graphic presence indicator value corresponding to the first location; and
specifying one or more transition directions;
for each of the one or more transition directions, receiving a corresponding spatial offset value; and
for each of the one or more transition directions, instantiating a record of the graphic presence indicator value corresponding respectively to one or more pixel locations within the frame of image data;
wherein the graphic presence criterion is evaluated at the one or more pixel locations within the frame of image data, thereby providing a graphic presence indicator value corresponding respectively to the one or more pixel locations within the frame of image data,
wherein the estimated background pixel value is further responsive to the each of the one or more transition directions, and
wherein the estimated background pixel value is further responsive to the spatial offset value corresponding to the each of the one or more transition directions.
2. The method of claim 1 , further comprising receiving a blending parameter, wherein the evaluation is also responsive to the blending parameter.
3. The method of claim 1 , further comprising receiving a graphic pixel value, wherein the evaluation is also responsive to the graphic pixel value.
4. The method of claim 1 , further comprising the steps of:
receiving a mask, wherein the mask indicates a first specified region corresponding to pixel locations located inside the mask;
specifying a second specified region comprising pixel locations located outside the mask and located within a third specified region;
averaging image data values corresponding to pixel locations within the first specified region, thereby forming an average image pixel value; and,
averaging image data values corresponding to pixel locations within the second specified region, thereby forming an average background pixel value;
wherein the image pixel value is responsive to the average image pixel value; and
wherein the estimated background pixel value is responsive to the average background pixel value.
5. The method of claim 1 , further comprising the steps of:
receiving a mask, the mask comprising a mask geometry, and wherein the mask indicates a first specified region corresponding to pixel locations located inside the mask;
providing a second specified region corresponding to selected pixel locations located inside the mask and responsive to the mask geometry;
providing a third specified region corresponding to selected pixel locations located outside the mask and responsive to the mask geometry;
averaging image data values corresponding to pixel locations within the second specified region, thereby forming an average image pixel value; and
averaging image data values corresponding to pixel locations within the third specified region, thereby forming an average background pixel value;
wherein the first image pixel value is responsive to the average image pixel value, and
wherein the estimated background pixel value is responsive to the average background pixel value.
6. The method of claim 1 , further comprising the steps of:
filtering image data in a first specified neighborhood of the first location and in a direction essentially orthogonal to the each of the one or more transition directions, thereby providing a first filtered image pixel value; and
filtering image data in a second specified neighborhood and in the direction essentially orthogonal to the each of the one or more transition directions, thereby providing a filtered background pixel value;
wherein the first image pixel value is responsive to the first filtered image pixel value;
wherein the estimated background pixel value is further responsive to the filtered background pixel value, and
wherein the second specified neighborhood is spatially offset from the first location responsive to the each of the one or more transition directions and the spatial offset value corresponding to the each of the one or more transition directions.
7. The method of claim 6 , further comprising the step of:
for each of the one or more transition directions, combining the instantiated record of the graphic presence indicator value corresponding respectively to one or more pixel locations within the frame of image data, thereby forming a combined transition frame of data.
8. The method of claim 6 , further comprising the step of:
for each of the one or more transition directions, combining with a logical OR function the instantiated record of the graphic presence indicator value corresponding respectively to one or more pixel locations within the frame of image data, thereby forming a combined transition frame of data.