Research on crack extraction based on the improved tensor voting algorithm
- PDF / 4,547,743 Bytes
- 16 Pages / 595.276 x 790.866 pts Page_size
- 0 Downloads / 207 Views
ORIGINAL PAPER
Research on crack extraction based on the improved tensor voting algorithm Cheng Xing 1,2 & Jingjing Huang 3 & Yaming Xu 1,2 & Jinfang Shu 4 & Chenyan Zhao 5 Received: 13 February 2018 / Accepted: 12 June 2018 / Published online: 27 June 2018 # Saudi Society for Geosciences 2018
Abstract The crack is an important index to evaluate the strength of buildings. However, for the tiny cracks with low signal-to-noise ratio, traditional methods cannot obtain good detection results. This paper proposes a new algorithm for crack extraction based on improved tensor voting. On the crack images after preprocessing, firstly, a contour dilation and filtration is proposed for denoising. Then, the tensor voting algorithm is used to obtain the probability map of cracks. Finally, based on the probability maps, the real cracks are extracted successively through sampling, refining, center line tracking, and projected positioning. The experimental results show that the proposed method is robust to noise and has good results on crack extraction. It can effectively extract linear cracks with tiny size, low contrast and poor continuity. Keywords Crack extraction . Tensor voting . Dilation and filtration
Introduction Cracks are widespread in concrete buildings. They not only reduce the impermeability of the building but can also cause the corrosion of steel bars and the carbonization of concrete, which harms the carrying capacity of buildings. In recent years, the methods of crack detection have gradually changed from ways of manual to digital, automatic, and intelligent. The researches of crack detection theory and method based on digital image processing are getting mature gradually (Kang, 2005). Since 1990s, a variety of crack extraction algorithms have been put forward. As the most direct method, threshold-based crack extraction has been extensively studied. Among them, the Otsu method (Otsu 1979), histogram method (Sezan 1990), and
* Jingjing Huang [email protected]
iterative cropping method (Oh et al. 1997) were the most commonly used. There are also many improved algorithms. For example, Wang (2000) proposed a multi-threshold segmentation algorithm combined with genetic algorithm, fuzzy C-means algorithm, and gray scale. Tsai et al. (2010) analyzed and compared several methods, including statistical threshold method, iterative method, and multi-scale wavelet transform method. Moreover, a low SNR (signal-to-noise ratio) segmentation method based on dynamic optimization was proposed for pavement crack detection. The method of edge detection is another topic that has been studied extensively. Among them, Canny operator (Canny 1986), Sobel operator (Sobel 1970), fast Fourier transform (Brigham 1988), and wavelet transform (Mallat 1989) are better edge detection operators. Li et al. (1991) combined Sobel
1
School of Geodesy and Geomatics, Wuhan University, Wuhan, China
2
Key Laboratory of Precise Engineering and Industry Surveying, National Administration of Surveying, Mapping and Geoinformation, Wuhan, China
Yamin
Data Loading...
