Self-intended crowd density estimation method for camera capable of straddling

Abstract

The invention discloses a self-intended crowd density estimation method for a camera capable of straddling. The steps of the self-intended crowd density estimation method for the camera capable of straddling comprise capturing a video monitoring signal, obtaining a video monitoring population image and conducting space mapping processing toward the video monitoring population image, picking up a foreground image of population motion according to a geographical reference and conducting operation such as edge detection and morphological processing towards the foreground image. If a foreground edge pixel number is smaller than a threshold value being set, a crowd density is calculated through a low density population estimation model and is classified according to a crowd density degree standard. If the foreground edge pixel number is larger than the threshold value being set, a textual feature of the foreground image is picked up; the crowd density degree is estimated by using a camera capable of straddling support vector machine (SVM) crowd density classifier. The self-intended crowd density estimation method for the camera capable of straddling adopts a video data space mapping method to unify video data to the geographical reference so that the problem of various population image sizes of different monitoring devices is solved, scene dependency of a model is overcome and establishing efficiency of a crowd density estimation model is greatly improved.

Description

technical field [0001] The invention relates to a method for constructing a crowd density estimation model that can cross cameras, specifically, a method for constructing a crowd density estimation model that overcomes the dependence of video monitoring scenes. Background technique [0002] With the rapid development of social economy, large-scale crowd gathering activities such as entertainment activities, exhibition activities, and sports events appear frequently, and casualties caused by excessive crowd density are not uncommon. Therefore, crowd density is an important indicator for risk assessment and safety management of public gathering places, and real-time monitoring of crowd density in public gathering places has important practical significance. Video itself has the characteristics of time and space, intuitive expression, rich information, and dynamic real-time. In recent years, video-based intelligent crowd density monitoring technology has become a research hotsp...

AI Technical Summary

Problems solved by technology

In order to realize the cross-scenario use of the crowd density estimation model, such methods need to use methods such as camera calibration, standardization processing, and scale correlation description to calculate the scale conversion factors between different cameras, but such methods are also relatively cumbersome, such as multi-scale features Block generation, feature template establishment and matching, camera calibration, etc.

[0007] In general, at present, most of the crowd density estimation algorithms are studied for specific monitoring scenarios, which are highly dependent on the scene and have high requirements for the stability of the monitoring equipment. A slight change in the posture of the monitoring probe will affect the accuracy of the model.

For monitoring areas with a large number of monitoring probes, the workload of model training is heavy, wasting a lot of manpower and time

The reason is that the scene dependence of the model is caused by the scale diversity of the surveillance images.

Due to the different installation heights, attitudes, and internal parameters of the surveillance probes, the same target in the surveillance image has different performance characteristics, making the crowd analysis model for a specific scene not universal.

Images