Intelligent video compression method based on image analysis

The invention belongs to the technical field of video compression, and particularly discloses an intelligent video compression method based on image analysis, and the method comprises the steps: carrying out the compression ratio determination of a to-be-compressed video from the three dimensions of an application scene, image features and a network state; the limitation and one-sidedness caused by determining the video compression ratio only according to the application scene in the prior art are greatly made up, so that the transmission quality of the video and the user experience are improved to the maximum extent; meanwhile, after the appropriate compression ratio is selected, the image frames with low application value are screened out based on the subject conformity, visual quality and content highlighting analysis results of all the image frames in the video to carry out contrast adjustment, targeted adjustment of the contrast of the image frames before video compression is achieved, the adjustment eff