Title :
Automatic Crack Detection and Segmentation Using a Hybrid Algorithm for Road Distress Analysis
Author :
Jinshan Tang ; Yanliang Gu
Author_Institution :
Sch. of Technol., Michigan Technol. Univ., Houghton, MI, USA
Abstract :
In this paper, we investigate advanced image processing technologies to detect cracks for road distress analysis. An algorithm which can detect and segment cracks effectively is proposed. The proposed algorithm is a hybrid crack detection and segmentation algorithm. In the proposed detection and segmentation algorithm, we first use histogram based thresholding method to get the rough locations of the cracks and then mathematics morphology technologies and B-spline based snake model based technology are used to refine the locations of the cracks. We conducted experiments on 10 images with different types of cracks and experimental results show that the proposed technologies can be used for find the cracks effectively.
Keywords :
crack detection; image segmentation; mechanical engineering computing; roads; B-spline based snake model based technology; automatic crack detection; automatic crack segmentation; crack location refinement; histogram based thresholding method; hybrid algorithm; image processing technologies; mathematics morphology technologies; road distress analysis; Anisotropic magnetoresistance; Image edge detection; Image segmentation; Morphology; Noise reduction; Roads; active contour model; cracks; road distress analysis; thresholding;
Conference_Titel :
Systems, Man, and Cybernetics (SMC), 2013 IEEE International Conference on
Conference_Location :
Manchester
DOI :
10.1109/SMC.2013.516