Dese population estimation method and system based on multi-feature fusion

Abstract

The invention provides a dense population estimation method and a system based on multi-feature fusion. The method comprises the following steps: partitioning an image into N equal sub-blocks; performing hierarchical background modeling on the image by using a method based on a CSLBP (Center-Symmetric Local Binary Pattern) histogram texture model and mixture Gaussian background modeling, extracting the foreground area of each sub-block subjected to perspective correction, detecting the edge density of each sub-block in combination with an improved Sobel edge detection operator, and extracting four important texture feature vectors in different directions for describing image texture features in combination with CSLBP transform and a gray-level co-occurrence matrix; performing dimension reduction processing on the extracted population foreground partition feature vectors and texture feature vectors through main component analysis; inputting the dimension-reduced feature vectors into an input layer of a nerve network model, and acquiring the population estimation of each sub-block through an output layer; adding to obtain the total population. The dense population estimation method and system have high accuracy and high robustness, and a good effect is achieved in the population counting experiment of subway station monitoring videos.

Description

technical field [0001] The invention belongs to the field of video analysis and processing, and in particular relates to a method and system for estimating a dense number of people. Background technique [0002] With the continuous increase of the world's population and the rapid development of social economy, mass incidents are increasing day by day, so the issue of crowd safety has become a hot issue that people pay attention to. The rapid development of video surveillance technology is an important technical guarantee for the control of mass incidents, and crowd counting is an important basis for effective management of mass incidents. [0003] The traditional visual surveillance application is based on the analog system of closed circuit television system (Closed Circuit Television, CCTV). The video monitoring of this system is mainly based on manual monitoring, and the manual method requires a huge monitoring workload. As the monitoring time increases, the attention of...

AI Technical Summary

Problems solved by technology

Conde C realizes people counting in monitoring scenes like square halls by extracting gray level difference matrix (GLDM) features. However, in this method, there are many feature values ​​to be extracted, and the calculation steps are too complicated.

Moctezuma D and Martin D used gray level co-occurrence matrix (GLCM) features for population counting, and simplified the final feature index through principal component analysis (PCA), which reduced the amount of calculation without affecting the accuracy, but the The method fails to consider the perspective effect well, that is, the number of pixels occupied by people in the image decreases with the increase of the distance from the camera, and the actual detection effect for complex scenes is average.

Albiol A and Hajer F can count the number of people in the scene by extracting the corner information of the moving area in the scene, and can count the number of people in the scene without extracting the foreground of the scene. However, this algorithm can only detect pedestrians in a moving state, and cannot detect a stationary state. the pedestrian

Images