Crowd crowding detection method, device, system and electronic equipment

Abstract

The present invention provides a crowd crowd detection method, device, system and electronic equipment, which relate to the technical field of image recognition. The method includes: acquiring a first image to be detected, that is, one of the images collected in various application scenarios; Extract the first crowd crowd feature in the first image, i.e. the background area feature of the first image; input the first crowd crowd feature into the pre-trained crowd degree detection classifier to obtain the classification of the crowd crowd degree of the first image Results; The congestion detection classifier is obtained by training the training sample data collected in various application scenarios whose number reaches a certain threshold. According to different application scenarios, the method can detect the degree of congestion in the scene combined with the background information of the collected images, reduce the dependence on high-configuration hardware devices, realize fast and accurate recognition of congestion, and meet the needs of various application scenarios. The need for rapid judgment of crowd crowding.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a crowd crowd detection method, device, system and electronic equipment. Background technique [0002] At present, the mainstream crowd crowding algorithm generally adopts the pattern recognition method of counting heads, first removes the background of the flow of people, and then distinguishes the number of heads in the remaining images according to the shape of the heads of the people, because the images in the video are dynamic. Continuous change, so a more accurate number of personnel can be obtained after a series of processing. This method has relatively high requirements on the accuracy of video equipment, and because it needs to use more complex algorithms to process video streams, the amount of background calculations required by the algorithms is relatively large, so special hardware is generally required to obtain more accurate head processing at a faster sp...

AI Technical Summary

Problems solved by technology

[0004] In view of the above situation, the existing traditional crowd crowd detection method, through the processing and analysis of video data, determines the number of people in the crowd, and then obtains the crowding degree in the corresponding scene. This process has the disadvantages of slow recognition speed and required hardware equipment. high configuration problem

Images