Automatic Road Crack Image Preprocessing for Detection and Identification

• Published in: 2009 Second International Conference on Intelligent Networks and Intelligent Systems pp. 319 - 322
• Main Authors: Wei Wei, Boying Liu, Peng Bai
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2009
• Subjects: Automated highways; Counting circuits; Feature extraction; Histograms; Holistic Thresholding; Image processing; Image segmentation; Intelligent vehicles; Preprocessing; Road Crack; Roads; Statistics; Vehicle detection
• ISBN: 142445557X; 9781424455577
• DOI: 10.1109/ICINIS.2009.88

Conventional visual and manual road crack detection method is no fit to the development of the highway: it is time-consuming, non-precise, dangerous, and costly. Automatic pavement survey is required. This paper treats a problem arising in the design of intelligent vehicles also describes an automatic road crack detection method based on image processing. It enhances the road image by comparing two fast correct method in the first, then processes segment threshold and extracts linear feature. It means that the diseases image is processed by Ostu algorithm and threshold division (by contrasting the whole pixel counter and statistics of the histogram) also we could set some parameters by experience and experimentation. Result shows that the Ostu algorithm makes more effective when there are some transverse cracks; longitudinal cracks or irregular cracks but is of no effect when there are just some noises. Threshold division makes much better by contrasting the whole pixel counter and statistics of the histogram. The experimental result is satisfactory.