一种改进的边界扫描的路面裂缝检测算法

An Algorithm of Pavement Crack Detection Based on Edge Detection

针对路面图像噪声较多、目标裂缝跟踪难等问题,分析对比了几种传统的经典边界扫描方法,如Sobel、Canny等算法,并根据路面裂缝图像的特点,提出了基于绝对梯度值的Sobel改进方法,使得边缘信息得到加强、减少了噪声以及伪边缘。经过后续图像的处理,能够较好地跟踪识别路面图像的裂缝信息。

The road is easy to be influenced by traffic load and natural factors. Pavement is easy to produce all kinds of breakage. Crack is a common form of most pavement diseases. In view of the difficulties such as pavement image noise and target tracking cracks, analyzing and comparing the several classical boundary scan methods, such as Sobel, Canny and other algorithms, according to the characteristics of pavement crack image, an improved Sobel method based on absolute gradient value was proposed. The edge information was enhanced, the noise and the false edges were reduced. After the follow - up image processing, the crack information was tracked and identified to the pavement image.