面向堤坝监测的无人机自动巡检系统设计

Design of UAV automatic inspection system for dam monitoring

针对人工巡检监测堤坝费时费力,存在安全风险,还可能会造成堤坝损害等问题,设计一种无人机大坝自动巡检系统:首先对搜索区域进行建模,采用基于栅格的技术,为每架无人机划分待搜索区域,然后采用牛耕法对区域进行全覆盖搜索;通过改进的yolov5算法对裂缝图像进行快速识别,利用MAD中值滤波和广义形态学滤波对红外图像进行预处理,实现了对堤坝渗漏区域的快速准确识别。

Aiming at the problems of manual inspection and monitoring,such as time-consuming and labor-intensive,security risk and damage to the dam,an automatic inspection system of UAV dam is designed.Firstly,model the search area and use grid based technology to divide the search area for each drone.Then,use the Boustrophedon method to conduct a full coverage search of the area.Through the improved yolov5 algorithm,crack images were quickly identified,and infrared images were preprocessed using MAD median filtering and generalized morphological filtering,achieving fast and accurate identification of dam leakage areas.