| CPC G09G 5/377 (2013.01) [G06T 5/50 (2013.01); G06T 5/92 (2024.01); G06T 5/94 (2024.01); G06V 10/60 (2022.01); H04N 23/741 (2023.01); H04N 23/743 (2023.01); G06T 2207/10144 (2013.01); G06T 2207/20012 (2013.01); G06T 2207/20021 (2013.01); G06T 2207/20208 (2013.01); H04N 25/589 (2023.01)] | 20 Claims |

|
1. A method comprising:
receiving, by a gain adjust component, a long exposure image having a long exposure time;
receiving, by the gain adjust component, a short exposure image having a short exposure time;
determining, by the gain adjust component, a gain adjustment for the long exposure time based on the short exposure time divided by the long exposure time;
outputting, by the gain adjust component, an adjusted long exposure image based on the gain adjustment for the long exposure time multiplied with each pixel in the long exposure image;
outputting, by the gain adjust component, an adjusted short exposure image;
determining, by a weight component, a merging weight associated with the long exposure image by comparing a single pixel value of the long exposure image with a saturation threshold, wherein the merging weight is not determined based on a block of pixels around the single pixel value;
determining, by the weight component, a motion adaptation weight of the long exposure image based on a difference of luminance of the single pixel value of the long exposure image and a corresponding pixel value of the short exposure image;
determining, by the weight component, a first weight based on the merging weight and further based on the motion adaption weight;
receiving, by a merge component, the first weight, the adjusted long exposure image, and the adjusted short exposure image; and
outputting, by the merge component, a merged high dynamic range image based on the first weight, the adjusted long exposure image, and the adjusted short exposure image.
|