一种自适应的变电站设备红外图像分割方法 - 红外技术.pdf

第38 卷 第9 期 红 外 技 术 Vol.38 No.9 2016 年9 月 Infrared Technology Sep. 2016 一种自适应的变电站设备红外图像分割方法 1 1 2 王启银 ，薛建东 ，任新辉 （1.国网山西省电力公司大同供电公司，山西 大同 037008 ；2.西南交通大学 电气工程学院，四川 成都 610031 ） 摘要：为了较好地实现变电站电气设备红外图像的分割，采用了一种自适应的变电站电气设备红外图 像分割方法。通过采用基于形态学的权重自适应算法对变电站电气设备红外图像进行增强处理，然后 采用基于加权切比雪夫距离的K-means 算法对变电站电气设备红外图像进行分割，最后对分割得到的 二值图像采用形态学方法进行处理。通过实验验证了该方法的有效性和适应性，方便了后续的特征提 取和识别。 关键词：变电站电气设备；红外图像；形态学；图像分割；K-means 中图分类号：TP301.6 文献标识码：A 文章编号：1001-8891(2016)09-0770-04 An Adaptive Segmentation Method of Substation Equipment Infrared Image 1 2 3 WANG Qiyin ，XUE Jiandong ，REN Xinhui （1.State Grid Datong Power Supply Company, Datong, Shanxi, 037008, China; 2. College of Electrical Engineering, Southwest Jiaotong University, Chengdu, 610031, China） Abstract ：In order to achieve the segmentation of substation electrical equipment infrared image better, an adaptive substation electrical equipment infrared image enhancement and segmentation algorithm was adopted in the paper. An adaptive weight algorithm based on morphological was used to enhance the infrared image of substation electrical equipment, then the K-means algorithm based on weighted Chebyshev distance was used for segmentation of substation electrical equipment infrared image, lastly the morphological method was used to deal with the segmented binary image. The arithmetic was validated by experiment later, and this method is convenient for follow-up feature extraction and recognition. Key words ：substation electrical equipment，infrared image，morphology