Method and device for image blurred screen detection

Abstract

The invention discloses a method and device for detecting image blurring, which includes taking each pixel of an image in a designated area as a pixel to be detected, and performing the following steps on the pixel to be detected: calculating the pixels to be detected respectively The difference between the RGB value of the point and the RGB value of the comparison pixel at the adjacent specified position; judge whether the absolute value of the difference is less than a certain preset distance threshold, if yes, add 1 to the accumulator, if not, skip this Points are not counted; divide the value of the accumulator by the total number of pixels in the area to obtain the ratio value; judge whether the ratio value is smaller than another preset artifact threshold, if so, the current image in the region is abnormal artifacts, if If not, it is not a Huaping abnormality. Beneficial effects of the present invention: It makes up for the defect in the prior art that the image acquisition of the encoder equipment fails, but the audio acquisition and encoding are normal, and the common alarm detection algorithm and the abnormality of the blurred screen that cannot be recognized by the equipment are generated. .

Description

technical field [0001] The invention relates to the technical field of image recognition, and in particular, to a method and device for detecting blurred images. Background technique [0002] Before the current TV program is broadcast, in order to avoid quality problems affecting the viewing effect during the broadcast of the program, a technical review of the program content is often carried out before the broadcast. All black picture), still frame (the screen stops at a certain picture for a period of time), silent (no sound can be heard when there should be sound) and other broadcast abnormalities, the judgment algorithm for these abnormalities is currently The technology is relatively mature, but for some complex abnormal judgments, there are few or even no good judgment algorithms. [0003] In the process of program recording, if there is a problem with the image capture device of the encoder, but the audio capture device is normal, and the encoding module is normal, i...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the problems existing in the prior art, the technical problem to be solved by the present invention is to combine the principle of locality to identify whether the current image has a blurred screen abnormality, and then solve the problem that such alarms cannot be identified in the current conventional signal anomaly monitoring algorithms. to provide a detection method and device that can accurately identify the blurring of the screen image at that time

Images