Automation Recognition of Pavement Surface Distress Based on Support Vector Machine

• Published in: 2009 Second International Conference on Intelligent Networks and Intelligent Systems pp. 346 - 349
• Main Authors: Nana Li, Xiangdan Hou, Xinyu Yang, Yongfeng Dong
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2009
• Subjects: Automation; Classification algorithms; feature extraction; genetic algorithm; Genetic algorithms; Image classification; Neural networks; pavement surface distress; Static VAr compensators; Statistics; support vector machine; Support vector machine classification; Support vector machines; Surface cracks
• ISBN: 142445557X; 9781424455577
• DOI: 10.1109/ICINIS.2009.95

In this paper, classification of pavement surface distress and the statistics of the distress data are discussed. In order to improve the accuracy and efficiency to identify the pavement surface distress by the image information, a new algorithm based on SVM is discussed. In this study, support vector classification (SVC), which is a novel and effective classification algorithm, is applied to crack images classification. In order to build an effective SVC classifier, parameters must be selected carefully. This study pioneered on using genetic algorithm to optimize the parameters of SVC. The performances of the SVC and the back-propagation neural network whose parameters are obtained by trial-and-error procedure have been compared with crack images data set. Experimental results demonstrate that SVC works better than the BPNN.