基于无人机巡检的渠道衬砌破损图像识别方法研究

Research on Recognition Method of Damaged Channel Lining Images based on UAV Patrol Inspection

渠道衬砌的完整对于安全高效运营具有重要的意义。针对现有渠道衬砌人工损毁定位效率低的缺点,提出了一种基于无人机巡检的渠道衬砌破损信息提取及破损情况评估方法。利用无人机获取渠道研究区域图像,基于双窗Otsu阈值化方法对图像进行二值化分类,识别图像中的渠道衬砌部分和非衬砌部分(包括非渠道部分与衬砌损毁部分),采用Canny边缘检测结合最大连通阈及最小外接矩等方法,从非衬砌部分获得渠道的轮廓,继而评估渠道衬砌破损情况及破损比例。对某引水工程渠道采用无人机技术及图像识别方法识别破损情况,得到了渠道衬砌的破损情况,该过程耗时短、操作便捷,结果证明本文所提方法的有效性。

The integrity of channel lining is important for safe and efficient operation.Aiming at the shortcomings of the existing artificial channel damage location,this paper proposes a method for extracting and analyzing channel lining damage based on UAV patrol inspection.The images of the research area were acquired by UAV,and the images were binary classified based on the double-window Otsu thresholding method to identify the channel lining part and non-lining part(including non-channel part and damaged lining part)in the image.Canny edge detection combined with the maximum connectivity threshold and the minimum external moment was used to obtain the contour of the channel from the non-lining part,and then the damage condition and damage ratio of the channel lining were evaluated.The UAV technology and image recognition method are used to identify the damage of the channel for a water diversion project,and the damage of the channel lining is obtained.The process is time-consuming and easy to operate.The results prove the effectiveness of the method proposed in this paper.