摘要
从具有复杂地物背景的航拍图像中快速检测与提取电力线是无人机电力线巡检的关键问题之一.在充分分析电力线图像特征的基础上,提出一种复杂地物背景下电力线检测与提取的新方法.首先对原始图像进行图像增强处理,增加电力线与背景的对比度,据此改善图像分割效果.接着采用多组形态学处理算法,滤除二值图像中的背景噪声并增强电力线边缘,进一步采用Canny算子检测电力线边缘,并对其做平滑处理.最后利用霍夫变换（Hough Transform,HT）实现对电力线的快速、完整提取.实验结果表明,本方法在复杂地物背景下对电力线提取的准确率较高,检测速度较快,对无人机电力线巡检的推广具有重要意义.
One of the key problems of power line inspection by unmanned aerial vehicle（UAV）is the rapid detection and extraction of power line from the aerialimage,which contains complex ground object background.Based on the sufficient analysis of the characteristics of power line image,a novel method of power line detection and extraction under complex ground object background is proposed.In order to increase the contrast between the power line and the background,the original image is firstly enhanced to improve the effect of image segmentation.Then,a set of morphological processing algorithms are used to filter the background noise of the binary image and enhance the edge of the power line.After that the Canny operator is used to detect the edge of the power line and smoothing operation is conducted later.Finally,the Hough transform（HT）is used to achieve the fast and complete extraction of power line.The experimental results show that the proposed method achieves not only high accuracy but also fast speed in power line extraction from complex ground object background,which is of great significance to the promotion of power line inspection by UAV.
作者
张从新
赵乐
王先培
ZHANG Congxin;ZHAO Le;WANG Xianpei(Office of Scientific Research and Development,Wuhan University,Wuhan 430072,China;School of Electronic Information,Wuhan University,Wuhan 430072,China)
出处
《武汉大学学报（工学版）》
CAS
CSCD
北大核心
2018年第8期732-739,共8页
Engineering Journal of Wuhan University
基金
国家自然科学基金项目(编号:50677047)
湖北省科技支撑计划(编号:2015BCE074)
关键词
复杂地物背景
电力线快速提取
图像增强
形态学处理
CANNY算子
HOUGH变换
complex ground object background
power line rapid extraction
image enhancement
morpho-logical processing
Canny operator
Hough transform （HT）
