Legibility enhancement for a logo, text or other region of interest in video
View Patent ↗A video processing system enhances quality of an overlay image, such as a logo, text, game scores, or other areas forming a region of interest (ROI) in a video stream. The system separately enhances the video quality of the ROI, particularly when screen size is reduced. The data enhancement can be accomplished at decoding with metadata provided with the video data for decoding so that the ROI that can be separately enhanced from the video. In improve legibility, the ROI enhancer can increase contrast, brightness, hue, saturation, and bit density of the ROI. The ROI enhancer can operate down to a pixel-by-pixel level. The ROI enhancer may use stored reference picture templates to enhance a current ROI based on a comparison. When the ROI includes text, a minimum reduction size for the ROI relative to the remaining video can be identified so that the ROI is not reduced below human perceptibility.
1. A method for video signal processing comprising:
identifying an overlay region of interest (ROI) in a frame of a video signal;
identifying a Snellen acuity factor value n in minutes of arc for gap sizes of text of the ROI;
setting a minimum ROI size limited to 1/n, while remaining portions of the frame of the video can be further reduced in size;
providing the minimum ROI size relative to a remaining portion of the frame of the video signal so that when a screen display size is below a predetermined amount, the ROI can be scaled relative to the remaining portion of the frame of the video signal; and
providing the minimum ROI size as metadata with the frame to a decoder for use in decoding the frame when a screen size for the decoded frame has a smaller form factor than an original form factor of the video signal.
2. The method of claim 1 , wherein the ROI comprises at least one of: a network logo and sports scores.
3. A method of video processing comprising:
receiving a video frame;
receiving ROI coordinates;
receiving a minimum ROI size value as metadata associated with the received video frame that is indicative of the Snellen acuity factor value n in minutes of arc for gap sizes of text of a ROI in the received frame;
downscaling the received video frame;
detecting when downscaling would reduce the size of the ROI more than indicated by minimum ROI size value; and
adjusting the size of the reduced ROI separate from the remaining portion of the frame of the video signal so that the reduced ROI size remains larger than indicated by the minimum ROI size value while remaining portions of the frame may be reduced further in size.