Artificial intelligence Technique for Pavement Diseases Identification

With the rapid development of road traffic construction, road maintenance and management tasks also follow. In particular, a number of high-grade highways built in the early stage have entered the period of intermediate or major repair, and the relevant road maintenance and management departments have paid more and more attention to the monitoring of road surface diseases and the collection of disease data. In order to improve the detection and recognition efficiency and accuracy of pavement surface diseases based on image processing, the Convolutional Neural Network (CNN) algorithm in target detection is introduced to quickly identify the type, location, and area for the extracted disease area with borders, the CNN based on crack contour network (CCN) method used to locate and extract the crack shape. CCN algorithm introduces the accuracy rate (P%), recall rate (R%), and F-score (F%) index to evaluate the algorithm in the problem during disease type identification, and determines the corresponding contour area of the disease frame according to the maximum F-score. A highway pavement image was carried out by using an inspection car with high definition camera. The results show the recognition efficiency and accuracy of the proposed method. After the optimal value of the degree threshold, the accuracy rate, recall rate, and F-score are recorded , and ,respectively.