FCM分割和形态学的沥青路面图像裂缝提取-计算机工程与应用.PDF

Computer Engineering and Applications 计算机工程与应用 2013 ，49 （4 ） 31 FCM 分割和形态学的沥青路面图像裂缝提取 宋蓓蓓，韦 娜 SONG Beibei, WEI Na 长安大学 信息工程学院，西安 710064 School of Information Engineering, Chang ’an University, Xi ’an 710064, China SONG Beibei, WEI Na. FCM segmentation and morphology based asphalt pavement image cracks extraction. Computer Engineering and Applications, 2013, 49 （4 ）：31-34. Abstract ：As the brightness, geometry and texture of the crack region is different from that of the road background, an asphalt pavement cracks extraction method using Fuzzy C-Means clustering （FCM ）segmentation and morphology is proposed. A series of preprocessing including denoising, brightness non-uniformity correction and contrast enhancement are implemented, as the aim is to improve the differences between cracks and background. The FCM method is adopted to segment the image to realize the initial cracks extraction. The morphological operations consist of the dilation, erosion and thin are performed on the obtained image, and then the fine cracks extraction is realized by making threshold decision on the cracks connected area. The experimental results show that the proposed method can effectively extract the small cracks and the cracks with the weak contrast, and it has strong robustness and practical value. Key words ：road engineering; pavement cracks; preprocessing; fuzzy c-means clustering; morphology 摘 要：利用裂缝相对于路面的亮度、几何纹理等特征差异，提出了一种基于FCM 图像分割和形态学的沥青路面图像裂缝 提取方法。拍摄的路面图像经过去噪、亮度非均匀校正和对比度增强等一系列预处理，提高图像中裂缝与路面背景的差 异。采用FCM 方法分割图像，初步提取裂缝。利用膨胀、腐蚀和细化等形态学操作，对裂缝连通区域面积进行阈值判决实 现裂缝的精细提取。实验结果表明，该综合方法获得了较好的检测结果，能有效地提取弱对比度的裂缝和细小裂缝。 关键词：道路工程；路面裂缝；预处理；模糊C 均值聚类；形态学 文献标志码：A 中图分类号：TP391 doi ：10.3778/j.issn. 1002-8331.1208-0201 1 引言 值分割方法和形状分析理论运用于路面裂缝的识别中，取 沥青路面长期受到大型车辆荷载的作用以及热胀冷 得了较好的识别效果。但由于路面破损图像具有噪声强、 缩等自然因素的影响导致路面产生裂缝，降低道路的使用 光照不均匀、形式多样等