Crowd density estimation method, device and monitoring system

Abstract

The invention provides a crowd density estimation method, a device and a monitoring system. The method comprises the following steps: acquiring an image to be estimated which is shot by an image shooting device; acquiring the textural feature of the image to be estimated; classifying the image to be estimated according to the textural feature of the image to be estimated by a classifier, determining the level of crowd density of the image to be estimated; and obtaining the classifier according to the corresponding relationship between the textural feature of a training image and the level of known crowd density of the training image. The invention does not need to count the number of people, thereby eliminating the problem that the estimation result which is obtained by directly counting the number of people in a monitoring scene is not very accurate, and being more suitable for video monitoring application in situations with high people density such as squares, railway stations and the like.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a crowd density estimation method, device and monitoring system. Background technique [0002] Intelligent video surveillance is a monitoring method based on computer vision technology to analyze video images in the surveillance scene, extract key information in the scene, and form corresponding events and alarms. It has extremely wide applications in public safety protection, traffic management, etc. [0003] The main function of crowd density estimation is to estimate the density of the crowd in the monitoring scene. According to the number of people, the crowd density in the monitoring scene is divided into several levels, such as low, medium, and high. It is suitable for crowded places such as squares, subway platforms, and supermarket entrances. The monitoring of occasions is an important application in the field of video surveillance. [0004] One of the most common appl...

AI Technical Summary

Problems solved by technology

[0007] For example, in a subway station, due to the existence of a large number of support columns, for a camera at a fixed position, the people behind the support columns cannot be photographed, so the human body detection using the captured images can also detect these people. The number of support columns in the station is extremely large, so there are also many human bodies that do not appear in the image, so the crowd density estimation result is very inaccurate

Images