Method for processing high-dynamic-range image videos on basis of comprehensive gains

The invention discloses a method for processing high-dynamic-range image videos on the basis of comprehensive gains, and belongs to the field of processing technology for image videos. The method includes estimating comprehensive gain values according to a camera model, computing a global HDR (high-dynamic-range) image and generating an HDR video. The method includes particular steps of 1), selecting key video frames from an input video and aligning all the video frames; 2), grouping the key video frames and respectively computing the comprehensive gain values in groups; 3), automatically selecting feature points from the key video frames and computing the global HDR image according to the comprehensive gain values; and 4), inversely transforming the global HDR image according to aligned data of the video frames to obtain the final HDR video. The method has the advantages of independence from special hardware equipment and wide application range. Besides, the requirement on converting common shot videos into HDR videos can be met.