基于红外图像的高压绝缘子串自动定位方法  

High Voltage Insulator String Automatic Location Method Based on Infrared Image

作　　者：胡淋波 姚建刚[1] 孔维辉[1] 邹品晶 付强[1] 潘雪晴[1] 

机构地区：[1]湖南大学电气与信息工程学院,湖南长沙410082

出　　处：《红外技术》2015年第12期1047-1051,共5页Infrared Technology

基　　金：江西省电力公司科技项目;编号:赣电科201350617

摘　　要：绝缘子串自动定位是实现劣化绝缘子红外智能检测的重要前提。针对红外图像背景复杂的特点,提出了基于灰度阀值分割和Hough变换的自动定位方法。首先采用基于Otsu阀值的灰度拉伸法对图像增强,利用灰度图像的开运算对图像去噪,然后采用灰度Otsu阀值法对图像进行阀值分割,图像二值化后,用最大连通区域过滤法提取绝缘子串区域,最后利用Hough变换进行倾斜度校正,完成绝缘子串的最终定位。试验结果表明:提出的分割方法可以快速完整地将目标提取出来。提出的定位方法可以完成复杂背景下绝缘子串的自动精确定位,为实现劣化绝缘子红外智能检测奠定基础。Insulator string automatic location is an important premise of realizing the intelligent testing of the deteriorated insulator by infrared technology. Aiming at the characteristics of infrared image complex back- ground, this paper proposes an insulator string automatic location method based on gray threshold segmenta- tion and Hough transform. Firstly, the infrared image of insulator string is enhanced by gray level stretching based on Otsu threshold; the gray image open-operation is carried out to denoise the whole image; then gray Otsu threshold method is adopted to segment the image; after the binarization of the image, maximum con- nected region filtration method is carried to retrieve the region of insulator string; at last, Hough transform is used to do the image gradient correction ,achieving insulator string final automatic location. Experiment re- suits show that the proposed segmentation can retrieve the target fast and perfectly; the proposed location method can achieve insulator string location under complicated backgrounds automatically and accurately, and it lays a foundation for the intelligent testing of the deteriorated insulator by infrared technology.

关 键 词：绝缘子串 红外图像 图像增强 阀值分割 HOUGH变换 自动定位 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]