Complex Context Of Bus Passenger Traffic Statistics Research

With the acceleration of urbanization process, the population in cities is increasing, and the burden of the urban public traffic system is higher and higher. Therefore, more attention has been paid to the vision-based intelligent transportation system. Research on the technology for the passenger counting based on image processing is just the hot of the field. Getting related passenger data, and making business operations based on the analysed data would greatly close to the Market demand for passenger bus lines.This paper introduces the background of the application of automatic passenger counting system, summarizes the current research status. Then studies the basics of image processing related to the project, including the gray image, filtering, sharpening and other image preprocessing method, gray level threshold, histogram threshold, the maximum entropy threshold value of several commonly used methods of image segmentation, and Sobel, Prewitt, Log, Canny operator for edge detection, and the actual detection results were compared. complex background edge background subtraction detection is proposed in the passenger detection with complex background, based on this method Hough transform is used to extract the head feature. To improve the detection accuracy, detected head contour is classified from the perspective of visual perception in this article, and a large number of non-target head contour is deleted. Images are taken with the experimental device, which simulates the actual bus in the laboratory.on and off passenger information is got in the camera installed on the top of the device. Algorithm is programmed in the WINDOWS operating system platform with development tools MATLAB2008.Experimental results show that the algorithm can extract the passengers head profile, and count accurately.