基于切比雪夫多项式描述子的复杂背景下输电线路提取算法  

Transmission Line Extraction Algorithm Based on Chebyshev Polynomial Descriptors in Complex Background

作　　者：赵常威 李森林 程登峰 操松元 严波[2] 陈忠 

机构地区：[1]国网安徽省电力公司电力科学研究院,合肥230022 [2]国网安徽省电力公司,合肥230061

出　　处：《计算机测量与控制》2017年第12期170-171,223,共3页Computer Measurement &Control

摘　　要：在电力系统中,利用计算机视觉和图像处理技术对输电线路的进行检测,在保障电力系统的安全运行方面具有非常重要的作用;针对复杂背景下的输电线路,提出了一种基于切比雪夫多项式描述子输电线路提取算法,该算法首先对输入图像进行预处理,其次利用随机Hough变换提取图像中的直线和曲线,然后利用切比雪夫多项式描述子的曲线描述优势,对所提取的对象进行特征描述,为了提高提取的精确度,通过K-means聚类分析方法来训练视觉字典,通过视觉字典的比对,最终确定图像中的输电线路;实验结果说明了所提出的算法可以精确地提取复杂背景下的输电线路。In the power system, there is a very important role in the protection of the safe operation of the power system by using com- puter vision and image processing technology to detect the transmission line. For the complex background, a transmission line detecting algo- rithm based on the Chebyshev polynomial descriptors （CPDs） is proposed. The algorithm firstly preprocesses the input image. Secondly, the random Hough transform is used to extract the straight lines and curves in the image, which are then described by the CPDs. In order to im- prove the accuracy of the extraction, the K--means clustering analysis method is used to train the visual dictionary, and the transmission lines in the image are finally determined by the comparison of the visual dictionary. The experimental results show that the proposed algo- rithm can accurately extract the transmission lines in complex background.

关 键 词：切比雪夫多项式描述子 随机HOUGH变换 输电线路 K-MEANS聚类 曲线描述 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]