Video flicker degree detection method and device

Abstract

The embodiment of the invention provides a video flicker degree detection method and device, and the method comprises the steps: extracting a to-be-detected video frame from a to-be-detected video, carrying out the statistics of a brightness mean value of the to-be-detected video frame, and calculating the difference between the brightness mean values of a front frame and a rear frame of the to-be-detected video; obtaining a brightness sensing threshold value of the human eyes, comparing the brightness mean value difference with the brightness sensing threshold value to obtain a logic value vector of the to-be-detected video frame, and obtaining the flicker intensity of the to-be-detected video according to the logic value vector; detecting a brightness abrupt change frame according to the brightness mean value of the front and rear frames, determining a brightness abrupt change group according to the brightness abrupt change frame, and determining the flicker volatility of the to-be-detected video according to the group number of the brightness abrupt change group; and determining the flicker degree grade of the to-be-detected video according to the flicker intensity and the flicker volatility of the to-be-detected video. According to the method, the flicker degree of the video can be comprehensively judged by detecting the flicker intensity and the flicker volatility of the video, manpower resources are saved, and the judgment accuracy of the flicker degree of the video is improved.

Description

technical field [0001] The present invention relates to the technical field of video analysis, in particular to a method and device for detecting video flickering degree. Background technique [0002] With the increasing popularity of camera devices in public and private places, camera devices may be used at various times and in various places, so the performance requirements of camera devices are also getting higher and higher, but in actual camera devices During the use of the camera, the captured camera video may flicker abnormally due to environmental exposure problems, equipment problems, etc., which affects the viewing of the camera video, so it is necessary to detect the flickering degree of the captured video. [0003] Existing video detection methods are divided into two types: manual on-duty and software detection. The method of manual guarding is that the watchman observes the video subjectively, or the watchman checks the feedback information of other video cons...

AI Technical Summary

Problems solved by technology

The method of manual guarding is that the watchman observes the video subjectively, or the watchman checks the feedback information of other video consumption users, and judges whether there is flickering in the video by human eyes. Due to the limited working time and attention of people, the video detection of a large number of users High human resource costs are required, and the intensity of video flicker is also affected by subjective perception

The existing software detection method is mainly to deal with the difference between the front and rear video frames, without considering the impact of human eyes on the intensity and volatility of video flicker on video perception, and the detection is not comprehensive enough, which increases the false detection rate of severe flicker

Images