一种管道蛇形机器人的裂缝视频检测系统  

A Crack Video Detection System of Pipeline Snake Shaped Robot

作　　者：赵达 王亚慧 陈林林 ZHAO Da;WANG Ya-hui;CHEN Lin-lin(School of Electrical and Information Engineering,Beijing University of Civil Engineering and Architecture,Beijing 100044,China)

机构地区：[1]北京建筑大学电气与信息工程学院,北京100044

出　　处：《科学技术与工程》2023年第6期2492-2498,共7页Science Technology and Engineering

基　　金：国家自然科学基金(51971013);机器人仿生与功能研究北京市重点实验室项目(30080922005)。

摘　　要：为实现蛇形机器人在管道内部快速准确的识别管道内壁裂缝,基于一种改进YOLOv3算法为管道蛇形机器人设计了快速检测管道裂缝的系统。该系统搭载了500万像素相机以及用于辅助标定的两个激光发生器,通过摄像机采集管道内部视频信息,使用改进YOLOv3算法对视频进行检测,若识别出裂缝则输出当前图像。之后结合激光标定和边缘检测算法得到当前裂缝的几何信息。改进YOLOv3算法使用K-means++算法对裂缝数据集进行聚类,得到最佳先验框,并使用距离交并比代替传统的交并比作为损失函数,以提高识别精度和速度。实验表明,改进YOLOv3算法平均精度为87.23%,与原始YOLOv3算法相比提高了5.88%;同时基于激光标定算法的图像处理得到的裂缝几何信息与实际尺寸误差在5%以内,可以用于实际工程。In order to realize the snake-like robot to quickly and accurately identify the cracks in the inner wall of the pipeline,a system for rapid detection of pipe cracks was designed for the pipeline snake shaped robot based on an improved YOLOv3 algorithm.The system is equipped with a 5-megapixel camera and two laser generators for auxiliary calibration.The system collects video information inside the pipeline through the camera,detects the video using the improved YOLOv3 algorithm,and outputs the current image if a crack was identified.After that,the laser calibration and edge detection algorithms were combined to obtain the geometric information of the current cracks.The improved YOLOv3 algorithm used the K-means++algorithm to cluster crack datasets to obtain the best prior box,and used the distance intersection over union instead of the traditional intersection over union as a loss function to improve the recognition accuracy and speed.Experiments show that the average accuracy of the improved YOOlOv3 algorithm is 87.23%,which is 5.88%higher than that of the original YOLOv3 algorithm.At the same time,the error between the geometric information and the actual dimensions of the crack obtained by the image processing based on the laser calibration algorithm is within 5%,which can be used for practical engineering.

关 键 词：管道裂缝 动态目标检测 YOLOv3算法 管道蛇形机器人 距离交并比 K-means++ 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]