The Research And Implementation Of System For Passenger Recognition Based On Video Image Processing In Railway Passenger Station

Passenger recognition technology based on video image processing has great value of application and broad development prospects in transportation, shopping malls and other areas of public security. In video monitoring system, digital image processing for passenger recognition is a hot issue for the domestic research. It refers to many of the core problems in wide ranges, including the real-time acquisition of the video image, moving target detection, target tracking and image processing and others. Under the background of research on the security management of railway passenger station and the image recognition technology related, the thesis was completed by the work of parameter recognition which contained the number, density and velocity of the passengers in the railway station, and the work of the development and integration of the passenger recognition system.In this thesis, at first, the basic theory of digital image processing is described. And then, based on the background of existing knowledge, passenger number recognition model is constructed. Based on the background modeling, it uses the method of the background update automatically in approaching. There are better effects than the traditional update ways, especially in the complex situations.In the model of the density and velocity recognition, the innovative method that is used for measuring and calculating the target area of the video surveillance is put forward. The matrix of homography between the three-dimensional and two-dimensional image is calculated by the measurement technique, and that will be used for the calculation in the effective area and walking distance of the passenger flow region, as the basis of calculating the density and velocity of passenger traffic.Finally, the railway passenger recognition system that is based on the video image processing is designed and it is programmed by C#and MATLAB mixing to realize. Through designing of each functional module of the system and its integration, the system makes it possible that realizes the recognitions of passenger number, density and velocity in the station and judges the conditions of the security level for an objective analysis for the passenger safety warning and organization. According to the results of system testing, it is proved to operate well and be great valuable.