Research On The Automatic Collecting Technology Of The Maintenance Data Of The Road Marking

Nowadays, most of the information collecting tasks of the road marking are completed by manual method, which not only wastes much time and manpower but also is not so accurate for some man-made factors. The study on automatic collecting for traffic markings will have an important signification on improving the efficiency of the traffic marking information collecting and the maintaining management.The studies in this paper include automatic data collecting system of road markings information, real-time collecting of road markings distress image, image segmentation and road markings distress survey, integrated with techniques such as image processing, vehicle and sensor technology.In this paper, automatic data collection system based on machine vision is designed for road markings maintenance information. This system is made up of vehicle, computer, image collection subsystem, image information collection subsystem and locating subsystem. Image collection subsystem collects image by CCD camera. Location subsystem measures the distance of vehicle and then controls the frequency of collecting image. Geographical subsystem collects data such as longitude, latitude and so on. Due to the wide-angle lens with inclined light axis from road surface, the road markings image collected would be distorted and we accomplish the rectification.To segment the road markings is the first step to take the automatic measurement of the distress. Here, the optimal threshold method, maximum classes square error method and minimum error threshold method are used to segment the distressed road markings form some typical images and we draw a conclusion that the segmentation using minimum error threshold method is most perfect, but it cannot segment the details accurately. In order to achieve more perfect segmentation effect, the dynamic threshold method, based on the minimum error has been discussed in this paper. At last, we apply the dynamic threshold method integrated with global threshold, compared global threshold method with dynamic threshold method and developed both merits, to segment the road markings and achieve anperfect effect.After segment the road markings image correctly, we have studied how to carry out the automatic measurement of the markings distress. The experiments have shown that the algorithm given in this paper is correct and the distress rate of the markings can be accurately calculated with it.To manage the data effectively, we design a road markings maintenance information database and develop the software named road markings maintenance information system. At the same, the automatic data collecting experiments have been develeped.