A pedestrian detection method and system

Abstract

A pedestrian detection method and system. The present invention provides a pedestrian detection method. Head and shoulder detection, face detection and whole person detection are performed on the collected images, and the characteristics of front head and shoulders, back head and shoulders or side head and shoulders can be obtained. , front full-person feature information, back full-person feature information, left full-person feature information, and right full-person feature information, so that for target pedestrians with face occlusion or camouflage, front, Back and side head and shoulder features, as well as full-person features from multiple angles, will be of great help to real-time handling of illegal acts and post-event analysis, which can greatly reduce the follow-up workload of relevant personnel. It overcomes the defects of the personnel bayonet system based on face detection in the prior art. It is not only suitable for frontal face detection, but also detects pedestrians on the back and side. A person's face snapshot plus a full-person snapshot basically includes a All the information of the person is convenient for some in-depth analysis in the later stage.

Description

technical field [0001] The invention relates to the technical field of image detection, in particular to a pedestrian detection method and system. Background technique [0002] With the rapid development of social economy, the application of video surveillance system in various industries is gradually maturing, and the number of cameras built and put into use is increasing rapidly. A large amount of pedestrian information can be obtained by using the camera set in the center of the traffic hub or at the checkpoint, which provides data support for the public security to chase criminals. How to effectively use these cameras to monitor personal behavior in public places, better maintain social stability and harmony, and eliminate all kinds of illegal and extreme pedestrians in the bud has become an urgent problem for the law enforcement departments of the whole society. [0003] In the prior art, the personnel bayonet system is based on face detection. The disadvantages of thi...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, the personnel bayonet system is based on face detection. The disadvantages of this system are: first, the face is easily camouflaged, such as wearing a mask or wearing sunglasses, the detection will fail, and the face of the pedestrian cannot be obtained Second, face detection is easily affected by the angle of the face, and non-frontal faces are prone to detection failures resulting in false positives and re-reports; thirdly, face detection has high requirements for camera setup, and the closer you are to people, the better the effect, but the closer you are to people, the better. It is tight and can easily cause people's vigilance; in addition, because face detection only extracts face information, the obtained information only contains a single frontal face photo, and the amount of information is relatively small, which is not conducive to subsequent applications

However, in this solution, face detection, human body detection, and head and shoulders detection are performed independently, and the amount of data processing is very large. In addition, what is obtained through similarity analysis in this solution is the counting information of people, which cannot comprehensively reflect the number of pedestrians. Information

Images