Elastic relaxation algorithm-based vehicle type identification system and elastic relaxation algorithm-based vehicle type identification method

Abstract

The invention relates to an elastic relaxation algorithm-based vehicle type identification system and an elastic relaxation algorithm-based vehicle type identification method. The system comprises a video camera, an image acquisition card and image processing and display equipment which are connected through data wires, and has the advantages of simplicity, convenience, flexibility, low cost, high efficiency and strong practicability. The method comprises the following steps: acquiring vehicle video through the video camera and establishing a Gaussian mean-based background model through acquired images as a detection foreground area; and dividing and extracting the foreground area where a moving object is out of the background as a background image by a frame difference operating method, processing the background image to acquire the length and the height of a vehicle by an elastic relaxation algorithm and then judging a vehicle type according to national vehicle type identification standards. In the method, a background transformation problem in a vehicle area detection and dividing process is solved; and the simple and practicable elastic relaxation algorithm takes the place of a fussy dividing algorithm, which greatly reduces the time complexity of the system, improves the applicability and robustness of the system and can obtain good vehicle type identification effect.

Description

technical field [0001] The invention relates to the technical fields of image processing and pattern recognition, in particular to feature extraction and shape recognition of vehicle target images. Background technique [0002] Vehicle detection is an important part of ITS (Intelligent Transportation Systems). In the intelligent transportation system, the calculation and acquisition of vehicle information play a key role in guiding the establishment of automatic vehicle monitoring and automatic toll collection systems, while vehicle type detection and license plate recognition technology is the premise and condition for realizing highway traffic automation. Therefore, vehicle The research of image detection technology has great practical significance to improve the automation degree of highway traffic and promote the development of intelligent transportation system. [0003] As a basic part of ITS, the vehicle flow detection and identification system occupies a very importa...

AI Technical Summary

Problems solved by technology

[0005] In addition, the current video traffic flow information collection system mostly requires users to calibrate the field of view. This operation is too complicated for users, especially non-professional users, and will introduce new errors.

In addition, the processing algorithms of most products are implemented by hardware. Although this will increase the processing speed, the scale and price of the system will increase accordingly.

Images