关键词: 桥梁工程, 桥梁裂缝检测, 分数阶微分, sobel算子, 纹理细节

Abstract: Aiming at the shortcomings of time-wasting, high risk and cost in the traditional process of detecting the bridge crack, fractional differential theory and sobel operation were combined and applied into the image edge detection. Therefore, a novel bridge crack detection model based on fractional differential theory was obtained. The theoretical analysis and simulation results show that comparing with the current edge detecting methods, the proposed model not only can detect the edge features of the images better, but also can detect more texture detail information. Furthermore, the objective evaluation indexes, such as connectivity, false detection rate and missing rate, show that the proposed method is an effective edge detection method, which can better achieve the purpose of bridge crack detection.

Key words: bridge engineering;bridge crack detection, fractional differential, sobel operator, texture detail