Research Of Pavement Pit Slot Detection Method Based On On-board Binocular Vision

In recent years, cars come into the ordinary family gradually, municipal roads and traffic situation has become a very important part in people’s daily life. In order to improve the level of road and traffic management, intelligent transportation system has received widespread attention. Normal real-time, accurate and efficient integrated transport management system is applied more and more widely in intelligent transportation system. However, in the traffic monitoring intelligent transportation system most attention focuses on the traffic flow, very little on city road pavement pothole detection and related test. It was only through the installation of special monitoring equipment in large scale road, or in the vehicle mounted camera and ground penetrating radar and other expensive equipment.In order to detect pavement pit efficiency, the vehicle mounted pavement detection system has been proposed in the paper, which is composed of a vibration sensor, GPS, CCD camera and so on. After describing the framework and the basic principle of Vehicle monitoring system, we put forward the measure to the area of pavement pit slot fast edge tracking algorithm which is based on region growing. For the detection of pavement pit depth, this paper takes the passive detection method based on binocular visual recognition technology, after comparing the advantages and disadvantages of active and passive, we put forward a method for image matching which use the correlation function of maximum measurement. Experimental results show that the system is able to measure the position, area, depth information of pit road accurately, most of all, low cost and high efficiency. This system can collect vibration information, GPS signal and video image information of driving cars in real time by car-collection-terminal, which is installed in the car. These data can not only be used on signs of electronic maps, but also provide a data for traffic management,by the mean time they also can provide reference for urban traffic maintenance as guide role in the driver’s driving.In the end of the paper, we test the system in real road. As a result, it shows that this system can identify pavement pit slot through using three-axis accelerometer. The road response accuracy can reach more than 90%, and the architecture of the system is quite clear, low cost. In a word, this system not only has an extremely strong feasibility in applying in industrial production, but also very important value in the practical application of road test.