The Design And Realization Of Image Processing In Vehicle-mounted Pavement Crack Detection System

As a common damage in the freeway, pavement crack is the key point of traffic maintenance. Since the traditional artificial detection method never fits for the demand of highway development, intelligent crack detection system becomes very important in the study. The complex road condition and the underdeveloped image acquisition software and hardware technology make the achieved pavement images with high noise, it brings difficult image processing and low crack detection. Therefore, it is significant to study a crack detection system processing high noisy images.This thesis focuses on image processing of vehicle-mounted pavement crack detection system, aiming to deal with high noisy images. Based on image features, this thesis divides crack into four types, and makes further algorithm's research for transverse and longitudinal crack detection.In this thesis, the system designs a particular research function of achieving the optimal parameters'threshold inspection. It well solves the problem of parameters in crack detection algorithm need to be set up different unified threshold to different background images. The system accomplishes the objective of the research through different modules of image preview, parameter setting, breakpoint processing, image viewer and image save. With the use of language C++ programming method, MFC technology and Opencv technology, this system achieves the function of intelligent detection from three perspectives of interface, algorithm and connector.Based on image features of high-noise transverse and longitudinal crack, this thesis proposes such image processing methods as image grid narrowing, template operation, length information of removing noise and regional extension. It is demonstrated efficient performance of algorithm.The system is based on high efficiency design and provides great reference value to the subsequent research of detection system.