无人机辅助巡视及绝缘子缺陷图像识别研究

Study on power transmission lines inspection using unmanned aerial vehicle and image recognition of insulator defect

针对传统的输电线路人工巡视存在安全、管理、效率方面的不足,引入六旋翼无人机,利用其良好的机动性、操控性和图像识别技术对输电线路进行辅助巡视。论文分析无人机辅助巡视的系统结构、作业模式及搭载设备,针对拍摄的高清细节巡视图像,在采用阈值分割,形态学和边缘检测技术的基础上,设计绝缘子的连通域特征和形状特征,提出能适应航拍的复杂背景和多角度的绝缘子自爆缺陷识别算法,实现绝缘子缺陷的自动侦测,减少人工干预。测试表明,无人机辅助巡视系统有效弥补人工巡视的不足,图像识别技术为缺陷判别提供了智能化的决策依据,与传统基于颜色特征算法相比,在复杂背景下识别准确率达到85%,是高新技术在智能电网中应用的有益探索。

Aiming at the fact that manual inspection of transmission line in traditional inspection method exists some issues on safety,management,efficiency etc.,an assistant inspection method using unmanned aerial vehicle（ UAV） combined with image processing technology is proposed for its remarkable flexibility and controllability in this paper. The system framework,operation mode and loaded devices are discussed. In order to find the defects automatically and reduce human intervention,after taking the HD image by camera on the hexacopter,an algorithm of insulator defect recognition which can adapt to the complex background and multiple angles is presented based on image recognition technology which using threshold segmentation,morphology,edge detection technology and the features of the connected domain and shape of the insulators specially. The tests demonstrate that the assistant inspection system using hexacopter can make up the shortage of manual inspection,and the image recognition technology provides a novel intelligent decision criterion for defect distinguish. Compared with the traditional algorithm based on color feature,the recognition accuracy rate reached 85% under the complex background. It＇s a helpful exploration of high-tech application in smart grid.