螺栓状态无人机可见光图像目标检测方法  

Target detection method in visible light image of bolt state UAV

作　　者：刘鑫明 徐驰名 郭韩金 马三江 LIU Xin-ming;XU Chi-ming;GUO Han-jin;MA San-jiang(State Grid Zhejiang Electric Power Co.,Ltd.,Hangzhou Fuyang District Power Supply Company,Hangzhou 311400,China;State Grid Zhejiang Electric Power Co.,Ltd.,Jiande Power Supply Company,Hangzhou 311600,China;Zhongneng Juchuang(Hangzhou)Energy Technology Co.,Ltd.,Hangzhou 311100,China)

机构地区：[1]国网浙江省电力有限公司杭州市富阳区供电公司,杭州311400 [2]国网浙江省电力有限公司建德市供电公司,杭州311600 [3]中能聚创(杭州)能源科技有限公司,杭州311100

出　　处：《信息技术》2025年第2期104-109,115,共7页Information Technology

摘　　要：为有效判断输电铁塔螺栓当前应用状态,排除人工干扰。文中提出基于无人机可见光图像的输电铁塔螺栓状态检测方法。利用无人机搭载可见光摄像机采集输电铁塔螺栓可见光图像,图像经旋转、平移与缩放处理后,校正螺栓的方向与格式大小;利用中值滤波方法增强校正后螺栓图像,构建输电铁塔螺栓状态检测双高斯分布拟合模型,按照优先级排列顺序通过拟合判断检测前景目标,实现输电铁塔螺栓状态检测。经实验验证:通过该方法增强后图像对比度及细节明显增加,可有效检测出螺栓的中心区域坐标以及边缘轮廓像素,且目标检测概率相对较高。In order to effectively judge the current application status of transmission tower bolts,eliminate artificial interference.This paper proposed a detection method of bolt state of transmission tower is based on visible light image of UAV.The visible light images of transmission tower bolts are collected by a visible light camera equipped with a UAV.After the images are rotated,translated and scaled,the direction and format size of bolts are corrected.The median filtering method is used to enhance the corrected bolt image,and the double Gaussian distribution fitting model is constructed to detect the bolt condition of transmission tower.The detection foreground target is determined by fitting according to the priority order,and the bolt condition detection is realized.Experiment results show that the enhanced image contrast and detail are significantly increased by this method,and the central coordinates and edge contour pixels of bolts can be effectively detected,and the target detection probability is relatively high.

关 键 词：双高斯分布 输电铁塔 螺栓状态 无人机 可见光图像 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]