中文摘要：

针对道路裂缝损伤检测效率较低、安全性较差、数据标准化不够等实际问题,提出一种非监督方法来实现路面裂缝自动识别与特征测量.首先采用分开-合并方法分割图像,将图像划分为较小的图像块,采用均值漂移方法对图像块进行平滑与分割迭代处理,并将分割图像块合并;然后采用均值漂移方法分2种策略定向跟踪裂缝像素提取裂缝骨架,并对裂缝骨架内插得到完整裂缝,同时计算裂缝形态参数,实现裂缝形态特征测量.实验结果表明,该方法可以对裂缝进行高效识别与准确测量.

英文摘要：

In view of practical problems such as road cracks damage detection with low efficiency, less security, and poor data standardization, an unsupervised method for the automatic recognition and characteristic measurement of road cracks is proposed. Firstly, apply the image to be segmented based on the method called split-and-merge. To be more exact： split the image into smaller blocks, after iterative smoothing and segmentation with the MS method, merge these blocks together. Secondly, two different strategies, both based on the MS method, are used to directionally trace the pixels to extract the skeleton of cracks. And the complete cracks are obtained from the interpolation on the skeleton. Meanwhile, calculate the parameters, the morphological characteristics of cracks are measured. The experimental results show that the proposed algorithm can effectively identify and accurately measure the cracks.