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 an electronic device to detect an overlay area within an image, the method comprising:
defining, by the electronic device, first and second areas within the image, the first and second areas non-overlapping;
calculating, by the electronic device, an alpha-blended value of a mean color value of the second area with an overlay color value; and
if a 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, then defining an overlay area comprising at least one pixel within the first area,
wherein the method is further performed separately on a first image, and on a second image, where the second image is a temporally averaged image, the method further comprising:
determining a common overlay area based on an intersection of the overlay area for the first image and the overlay area for the second image.
2. The method of claim 1 , wherein the overlay area is specified as a portion of a binary image, the method further comprising:
applying at least one of: a morphological closing and a morphological opening operation to at least a portion of the binary image to provide a second binary image.
3. A method for an electronic device to detect an overlay area within an image, the method comprising:
defining, by the electronic device, first and second areas within the image, the first and second areas non-overlapping;
calculating, by the electronic device, an alpha-blended value of a mean color value of the second area with an overlay color value; and
if a 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, then defining an overlay area comprising at least one pixel within the first area,
wherein the image is a time-averaged composition of a plurality of frames within a video based on a change-gated temporal averaging process.
4. The method of claim 3 , wherein the change-gated temporal averaging process includes a frame in the time-averaged composition when a predetermined difference exists between the frame and a previous frame already included in the time-averaged composition.
5. The method of claim 3 , wherein the change-gated temporal averaging process includes a frame in the time-averaged composition when a number of frames since a last update to the time-averaged composition exceeds a threshold.
6. A method for an electronic device to detect an overlay area within an image, the method comprising:
defining, by the electronic device, first and second areas within the image, the first and second areas non-overlapping;
calculating, by the electronic device, an alpha-blended value of a mean color value of the second area with an overlay color value;
if a 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, then defining an overlay area comprising at least one pixel within the first area,
successively defining other corresponding first and second areas within the image; and
repeating the calculating and if-then steps with the other corresponding first and second areas,
wherein each first area is located at a pre-defined offset from a location of its corresponding second area.
7. The method of claim 6 :
wherein the successive defining comprises defining a new corresponding first area for the second area, the new corresponding first area distinct from the first area;
wherein the new corresponding first area is located at a new offset from the location of the second area; and
wherein the new offset is greater than the first offset.
8. The method of claim 3 , wherein the successively defining is performed in a direction across the image selected from the group consisting of: horizontally, vertically, and diagonally.
9. The method of claim 6 , further comprising:
successively defining other first and second areas within the image until the entire image has been incorporated into first and second areas.
10. A method for an electronic device to detect an overlay area within an image, the method comprising:
defining, by the electronic device, first and second areas within the image, the first and second areas non-overlapping;
calculating, by the electronic device, an alpha-blended value of a mean color value of the second area with an overlay color value; and
if a 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, then defining an overlay area comprising at least one pixel within the first area,
wherein the image is a first frame of a series of frames within a video;
the method further comprising:
repeating the defining, comparing, and performing steps with a second frame within the video, the first and second frames distinct;
if overlay areas are found within a plurality of successive frames, then comparing overlay areas across a plurality of successive frames; and
if the successive overlay areas are consistent with one another, then including the overlay areas in an overlay mask.
11. A method for an electronic device to detect an overlay area within an image, the method comprising:
defining, by the electronic device, first and second areas within the image, the first and second areas non-overlapping;
calculating, by the electronic device, an alpha-blended value of a mean color value of the second area with an overlay color value;
if a 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, then defining an overlay area comprising at least one pixel within the first area; and
redefining the overlay area to exclude the at least one pixel within the first area if the mean color value of the first area is greater than the alpha-blended value plus a tolerance value, or is less than the alpha-blended value minus a tolerance value.
12. The method of claim 11 wherein the tolerance value depends on the mean image color value of the second area.
13. The method of claim 11 , wherein the tolerance value is based linearly on a luminance of the mean color value of the second area.
14. A method for an electronic device to detect an overlay area within an image, the method comprising:
defining, by the electronic device, first and second areas within the image, the first and second areas non-overlapping;
calculating, by the electronic device, a predicted mean color value of the first area based, at least in part, on a mean color value of the second area, wherein the predicted mean color value is greater than or equal to the mean color value of the second area; and
if a mean color value of the first area is closer to the predicted mean color value than it is to the mean color value of the second area, then:
defining an overlay area comprising at least one pixel within the first area; and
performing, by the electronic device, an action selected comprising at least one of: displaying a border around the overlay area, indicating that an overlay area has been found, providing a mask of the overlay area, influencing parameters to be used when encoding at least the overlay area, and causing an encoder to target a higher quality of encoded video output for the overlay area,
wherein a difference between the predicted mean color value of the first area and the mean color value of the second area decreases linearly as the mean color value of the second area increases.
15. A method for an electronic device to detect an overlay area within an image, the method comprising:
defining, by the electronic device, first and second areas within the image, the first and second areas non-overlapping;
calculating, by the electronic device, a predicted mean color value of the first area based, at least in part, on a mean color value of the second area, wherein the predicted mean color value is greater than or equal to the mean color value of the second area;
if a mean color value of the first area is closer to the predicted mean color value than it is to the mean color value of the second area, then:
defining an overlay area comprising at least one pixel within the first area; and
performing, by the electronic device, an action selected comprising at least one of: displaying a border around the overlay area, indicating that an overlay area has been found, providing a mask of the overlay area, influencing parameters to be used when encoding at least the overlay area, and causing an encoder to target a higher quality of encoded video output for the overlay area; and
redefining the overlay area to exclude the at least one pixel within the first area if:
the mean color value of the first area is greater than the predicted mean color value plus a tolerance value, or
the mean color value of the first area is less than the predicted mean color value minus a tolerance value.