| 基于主动式全景视觉的管道内部缺陷检测方法 |
| |
| 引用本文： | 吴 挺,鲁少辉,韩旺明,胡克钢,汤一平.基于主动式全景视觉的管道内部缺陷检测方法[J].仪器仪表学报,2015,36(10):2258-2264. |
| |
| 作者姓名： | 吴 挺 鲁少辉 韩旺明 胡克钢 汤一平 |
| |
| 作者单位： | 浙江工业大学信息工程学院杭州310023 |
| |
| 基金项目： | 国家自然科学基金(61070134)项目资助 |
| |
| 摘 要： | 针对管道内表面检测空间受限的约束以及同时需要检测管道内壁功能性缺陷和结构性缺陷等需求，基于主动式全景视觉检测原理提出了一种既能视觉检测管道横截面几何尺寸又能获取和分析管道内壁表面缺陷的管道内部全方位视觉检测方法。利用携带有主动式全景视觉传感器的爬行机器人进入管道内部，分别实时获取内壁全景图像及激光横断面扫描全景图像。本文重点介绍基于管道内壁全景图像的缺陷检测方法，对管道内壁全景图像进行展开、预处理并提取管道病害区域的几何特征，最后判定缺陷类别及危害程度。经本文方法获取的缺陷几何特征实验数据验证，管道内表面存在的较为普遍的裂缝、腐蚀缺陷几何特征差异明显：腐蚀缺陷的圆形度较大，而裂缝缺陷的圆形度趋于0；腐蚀缺陷的凸度普遍大于裂缝缺陷；裂缝缺陷由于弯曲程度不同其边界离心率变化率大，而腐蚀缺陷的边界离心率趋于1。实验结果表明：该系统能够有效提取疑似缺陷区域并计算各几何特征，可通过合适的阈值判定其所属类别，为管道内表面缺陷检测提供了一种新的手段。
|
| 关 键 词： | 管道缺陷检测 主动式全景视觉 视觉三维测量 全景激光视觉 |
In pipe internal defect inspection method based on active stereo
omni directional vision sensor |
| |
| Wu Ting,Lu Shaohui,Han Wangming,Hu Kegang,Tang Yiping.In pipe internal defect inspection method based on active stereo
omni directional vision sensor[J].Chinese Journal of Scientific Instrument,2015,36(10):2258-2264. |
| |
| Authors: | Wu Ting Lu Shaohui Han Wangming Hu Kegang Tang Yiping |
| |
| Affiliation: | School of Information Engineering, Zhejiang University of Technology, Hangzhou 310023, China |
| |
| Abstract: | To ensure the safety and serviceability of underground pipelines, it is essential to inspect and assess the structural and functional faults at the same time. The underground pipe inner surface defect inspection technique is constrained by the space accessibility. Thus, in this paper we propose a comprehensive in pipe visual inspection method based on the active stereo omnidirectional vision principle, which can not only visually detect the geometrical dimension of the pipe cross section, but also acquire and analyze the surface defect on the pipe internal wall. In our system, a crawling robot equipped with an active stereo omnidirectional vision sensor travels along the pipeline, and takes the panoramic image of the pipe internal surface, and the panoramic image of the laser streaks on the pipe cross section projected from an omnidirectional laser in real time. In this paper, the defect detection method based on the internal surface panoramic image is introduced emphatically. The panoramic image of the pipe internal surface is unwrapped, pre processed, and the geometrical features of the defect region is extracted; finally the defect category and harm degree are determined. The geometrical characteristic experiment data obtained with the proposed method verify that the geometric characteristics of the crack, corrosion defects generally existing on the pipe inner surface are significantly different: the circularity of the corrosion defect is greater while that of the crack defect tends to 0; the convexity of the corrosion defect is generally greater than that of the crack defect; the border eccentricity of the crack defect changes significantly due to different bending degrees, and the border eccentricity of the corrosion defect tends to 1. Experiment results show that this system is able to extract the suspected defect region effectively and compute the geometrical characteristics, ultimately determines the category of the defect with appropriate threshold, which provides a new approach for the inspection of in pipe internal surface defects. |
| |
| Keywords: | in pipe defect inspection active stereo omni directional vision visual 3D measurement panoramic laser vision |
|
| 点击此处可从《仪器仪表学报》浏览原始摘要信息 |
|
点击此处可从《仪器仪表学报》下载全文 |
|
