Monitoring-video-oriented masked face detection method

Abstract

The invention relates to a monitoring-video-oriented masked face detection method. The method comprises the steps of: carrying out video frame skipping processing strategies on the basis of simple background difference value operation; realizing moving head and shoulder region detection on the basis of an improved background difference method; and detecting a masked face on the basis of a two-stage Haar classifier. By carrying out the video frame skipping processing strategies on the basis of the simple background difference value operation, the processing range is effectively reduced; the system overhead is reduced; and the processing efficiency is improved. A region with a head and shoulder image is judged to be an effective moving region; the condition that the background is detected to be a moving object by mistake is effectively reduced; by combining the frame skipping strategies, the background is only updated only when the existence of the head and shoulder moving region is detected; and the processing speed can be further accelerated, and the moving target extraction quality of the background difference method can be further improved. According to the Haar linear features of the common masked face, the training and the detection of a masked target are carried out on the basis of a binary image; the gray level distribution rules of training samples and a target to be detected can be intensified and emphasized; and an effect better than that of a pure gray level image is obtained.

Description

technical field [0001] The invention belongs to the field of video image processing for public security early warning, and in particular relates to a masked face detection method. Background technique [0002] Video surveillance is widely used in the field of public security, providing strong data and technical support for early warning and verification in public security management business. However, the level of intelligent analysis for surveillance video is still relatively low, and there is no effective technical means for this application in terms of masked face detection for surveillance video. Masked face detection is mainly used to quickly screen out suspicious targets whose faces are intentionally blocked from massive surveillance videos, which can enhance the supervision efficiency of specific targets by the public security department, and is useful for preventing and cracking down on crimes, tracking down suspects, and maintaining social security and stability. e...

AI Technical Summary

Problems solved by technology

[0003] For the various links in masked face detection, in terms of moving target detection, existing methods such as patent 201410110812.1 use the ViBe algorithm to establish a background model for the video image frame, and segment the foreground area by combining the ViBe algorithm of the frame difference method. The background update speed of the method is slow; patent 201110253323.8 uses edge detection and frame difference method for motion detection, and patent 201310586151.5 combines adjacent frame difference method and mixed Gaussian model to realize moving targets. The disadvantage of the above method is that it is prone to hollow areas; in human body detection On the one hand, patent 201010218630.8 uses fuzzy templates to detect multi-pose human bodies, which is slow; patent 201310415544.X uses color and depth information for human body detection methods, and the features obtained by joint feature extraction are used for human body detection, which is greatly affected by light; Patent 201110026465.0 detects people based on depth images, which is not suitable for conventional surveillance video images; in terms of masked face detection, patent 201210052716.7 detects masked heads for the entire frame of images, and the processing speed is relatively slow

Images