多尺度空间下的隧道裂缝与渗水区域检测

基于点云影像提出了多尺度空间下的裂缝渗水区域识别检测算法.根据不同尺度隧道裂缝的物理特点,通过定义尺度空间,提出了一种融合型图像检测算子,保持对带状裂缝检测的稳定性,同时对于灰度值较高的小裂缝,可以恢复对其的敏感特性,得到较多的响应边缘像素,实现不同尺度裂缝渗水区域的识别和检测.并通过实例证实了该算法可以有效排除伪裂缝信息的干扰,对不同尺度的裂缝进行准确识别定位和提取.

Tunnel Crack and Seepage Detection in Multi-scale Space

Based on the image of point cloud, a detection algorithm for cracks and water seepage area identification in multi-scale space is proposed. According to the physical features of tunnel cracks in different scale, meanwhile, with the definition of scale space, a fusion image detection operator is developed to maintain the stability of ribbon like cracks detection, at the same time, to restore the sensitivity to the small cracks in high gray value for obtaining more edge pixels. Therefore, the identification and detection of multi-scale cracks and seepage are achieved with the application of proposed algorithm. It is proved by an example that the algorithm can effectively eliminate the interference of fake cracks, and accurately identify and locate cracks in different scales.