基于Hough变换的管道漏水病害探地雷达图像检测方法与应用  

Image Detection Method and Application of GPR Based onHough Transform for Pipeline Leakage Defects

作　　者：施烨辉[1] 许佳毅 沈宇鹏[2] Shi Yehui;Xu Jiayi;Shen Yupeng(Jiangsu Nanjing Institute of Geological Engineering Survey;School of Civil Engineering,Beijing Jiaotong University)

机构地区：[1]江苏南京地质工程勘察院,南京市100068 [2]北京交通大学土木建筑工程学院,北京市100044

出　　处：《勘察科学技术》2024年第4期48-52,58,共6页Site Investigation Science and Technology

基　　金：江苏省隧道与地下工程技术研究开放基金(2019);国家重点研发计划(2017YFC0805305)。

摘　　要：图像边缘检测是计算机视觉领域一个基础而重要的技术。该文基于图像边缘检测技术Hough变换方法,对不同方位管道漏水病害探地雷达模拟图像的振幅、相对波阻抗和瞬时振幅图进行自动检测,实验和实际雷达数据检测结果显示,这种图像处理方法可以更好地识别探地雷达图像中的病害特征,对人工识别探地雷达管道漏水病害起到了辅助性作用,并提供了机器学习识别病害的基础方法。Image edge detection is a basic and crucial technology in the field of computer vision.It utilizes image edge detection technology based on Hough transform method to automatically detect the amplitude,relative wave impedance and instantaneous amplitude maps of GPR simulation images of pipeline leakage defects in different directions in this paper.The experimental and actual radar data detection results show that this image processing method can better intelligently identify disease features in GPR images,play an auxiliary role in artificial detection of GPR pipeline leakage defects,and provide a basic method for machine learning-based disease defect detection.

关 键 词：HOUGH变换 探地雷达 图像检测 管道病害 

分 类 号：TU990.3[建筑科学—市政工程] TN957.52[电子电信—信号与信息处理]