A Holistic Segmentation Method for Faulty Electrical Equipment under Complex Background
-
摘要: 针对变电站电气设备红外监测过程中，获取的红外图像背景复杂而导致故障设备定位不准确、分割难度较大等问题，提出了一种在复杂背景下对故障设备进行定位与整体分割的方法。首先，通过SLIC（Simple Linear Iterative Clustering）超像素算法对图像进行分割，并对超像素块进行Lab颜色空间转换，根据阈值判断是否存在故障并获取故障区域。然后，选取故障图像中最大联通量的较亮点作为种子，利用最大类间方差原理控制种子数目，通过改进区域生长法获取目标主体设备。最后，将故障区域与目标主体设备进行交集运算，完成对故障电气设备的整体分割。研究结果表明，该方法能有效完成复杂背景下的故障电气设备定位与整体分割。与其他分割方法相比，该方法获取的故障电气设备更加完整准确。Abstract: A method of positioning and integral segmentation of faulty equipment in infrared images acquired during the process of infrared monitoring of electrical equipment in substations with complex backgrounds is proposed to contribute to solving problems including inaccurate positioning and difficult segmentation of faulty equipment. First, the image was segmented using the SLIC superpixel algorithm and the superpixel block was transformed into the Lab color space. The faulty area was obtained after the fault was determined based on the threshold value. Second, relatively bright spots with the maximum connectivity in the image, including faulty equipment, were selected as the original seeds. The number of seeds was controlled based on the principle of maximum variance between classes. Accordingly, primary equipment was obtained using an improved regional growth method. Finally, the overall segmentation of the faulty electrical equipment was completed through an intersection calculation between the faulty area and the main equipment. The results show that the positioning and overall segmentation of faulty electrical equipment under complex backgrounds can be successfully completed using the proposed method. Compared with other segmentation methods, identification of faulty electrical equipment using this method is more complete and accurate.
-
Key words:
- faulty electrical equipment /
- complex background /
- infrared image /
- SLIC algorithm /
- image segmentation
-
图 1 基于SLIC故障区域分割算法流程图
Figure 1. Flow chart for fault region segmentation based on SLIC algorithm
图 4 不同种子数分割后的类间方差
Figure 4. Variances between classes after segmentation of different seed numbers
表 1 4种算法的实验结果
Table 1. Experimental results due to four algorithms
Segmentation method OTSU Marked watershed Region growing Ours Precision Recall Precision Recall Precision Recall Precision Recall Picture1 0.4259 0.9271 0.3190 0.4037 0.9360 0.7686 0.9596 0.9240 Picture2 0.8438 0.8425 0.8060 0.9148 0.7244 0.8680 0.9324 0.9660 Picture3 0.4325 0.8945 0.4327 0.9192 0.7109 0.8136 0.9394 0.9393 Picture4 0.4924 0.9582 0.5474 0.9306 0.7032 0.9454 0.9485 0.9218 Picture5 0.2527 0.9322 0.4624 0.9261 0.6862 0.9244 0.9682 0.9388 
下载: 导出CSV
-
[1] 王启银, 薛建东, 任新辉. 一种自适应的变电站设备红外图像分割方法[J]. 红外技术, 2016, 38(9): 770-773. http://hwjs.nvir.cn/article/id/hwjs201609010WANG Qiyin, XUE Jiandong, REN Xinhui. An adaptive infrared image segmentation method for substation equipment[J]. Infrared Technology, 2016, 38(9): 770-773. http://hwjs.nvir.cn/article/id/hwjs201609010 [2] 王小芳, 毛华敏. 一种复杂背景下的电力设备红外图像分割方法[J]. 红外技术, 2019, 41(12): 1111-1116. http://hwjs.nvir.cn/article/id/hwjs201912004WANG Xiaofang, MAO Huamin. Infrared image segmentation method for power equipment under complex background[J]. Infrared Technology, 2019, 41(12) : 1111-1116. http://hwjs.nvir.cn/article/id/hwjs201912004 [3] 林颖, 郭志红, 陈玉峰. 基于卷积递归神经网络的电流互感器红外故障图像诊断[J]. 电力系统保护与控制, 2015, 43(16): 87-94. doi: 10.7667/j.issn.1674-3415.2015.16.013LIN Ying, GUO Zhihong, CHEN Yufeng. Convolutional-recursive network based current transformer infrared fault image diagnosis[J]. Power System Protection and Control, 2015, 43(16): 87-94. doi: 10.7667/j.issn.1674-3415.2015.16.013 [4] 陈跃伟, 彭道刚, 夏飞, 等. 基于区域生长法和BP神经网络的红外图像识别[J]. 激光与红外, 2018, 48(3): 401-408. doi: 10.3969/j.issn.1001-5078.2018.03.024CHEN Yuewei, PENG Daogang, XIA Fei, et al. Infrared image recognition based on region growing method and BP neural network[J]. Laser & Infrared, 2018, 48(3): 401-408. doi: 10.3969/j.issn.1001-5078.2018.03.024 [5] 贾鑫, 张惊雷, 温显斌. 双监督信号深度学习的电气设备红外故障识别[J]. 红外与激光工程, 2018, 285(7): 32-38. https://www.cnki.com.cn/Article/CJFDTOTAL-HWYJ201807005.htmJIA Xin, ZHANG Jinglei, WEN Xianbin. Infrared fault identification of electrical equipment based on dual supervised signal deep learning[J]. Infrared and Laser Engineering, 2018, 285(7): 32-38. https://www.cnki.com.cn/Article/CJFDTOTAL-HWYJ201807005.htm [6] 尹阳. 基于红外图像的变电站设备识别与热状态监测系统研究[D]. 西安: 西安科技大学, 2018.YIN Yang. Research on Substation Equipment Identification and Thermal State Monitoring System Based on Infrared Image[D]. Xi'an: Xi'an University of Science and Technology, 2018. [7] 余成波, 曾亮, 张林. 基于OTSU和区域生长的电气设备多点故障分割[J]. 红外技术, 2018, 40(10): 1008-1012. http://hwjs.nvir.cn/article/id/hwjs201810013YU Chengbo, ZENG Liang, ZHANG Lin. Multipoint fault segmentation for electrical equipment based on OTSU and regional growth[J]. Infrared Technology, 2018, 40(10): 1008-1012. http://hwjs.nvir.cn/article/id/hwjs201810013 [8] 赵梦. 基于红外图像的电力设备故障分析研究[D]. 西安: 西安理工大学, 2020.ZHAO Meng. Research on Fault Analysis of Power Equipment Based on Infrared Image [D]. Xi 'an: Xi 'an University of Technology, 2020. [9] 任新辉. 基于红外技术的变电站设备识别与热故障诊断[D]. 成都: 西南交通大学, 2016.REN Xinhui. Substation Equipment Identification and Thermal Fault Diagnosis Based on Infrared Technology[D]. Chengdu: Southwest Jiaotong University, 2016. [10] 张锦文. 变电站电气设备红外图像分割方法研究[D]. 北京: 华北电力大学, 2018.ZHANG Jinwen. Research on Infrared Image Segmentation Method of Electrical Equipment in Substation[D]. Beijing: North China Electric Power University, 2018. [11] 康龙. 基于红外图像处理的变电站设备故障诊断[D]. 北京: 华北电力大学, 2016.KANG Long. Fault Diagnosis of Substation Equipment Based on Infrared Image Processing[D]. Beijing: North China Electric Power University, 2016. [12] 郭铭. 基于红外成像技术的变电站电气设备热故障诊断研究[D]. 阜新: 辽宁工程大学, 2019.GUO Ming. Research on Thermal Fault Diagnosis of Electrical Equipment In Substation Based on Infrared Imaging Technology[D]. Fuxin: Liaoning Engineering University, 2019. [13] 刘辉, 石小龙. 结合显著性和超像素改进的GrabCut图像分割[J]. 红外技术, 2018, 40(1): 55-61. http://hwjs.nvir.cn/article/id/hwjs201801010LIU H, SHI X L. Improved grab cut segmentation based on salience and superpixels[J]. Infrared Technology, 2018, 40(1): 55-61. http://hwjs.nvir.cn/article/id/hwjs201801010 [14] Levinshtein A, Stere A, Kutulakos K N, et al. Turbopixels: fast superpixels using geometric flows[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(12): 2290-2297. doi: 10.1109/TPAMI.2009.96 [15] REN X F, Malik J. Learning a classification model for segmentation[C]//Proceedings of the Ninth IEEE International Conference on Computer Vision, 2003: 10. [16] Veksler O, Boykov Y, Mehrani P. Superpixels and supervoxels in an energy optimization framework[C]//European Conference on Computer Vision(ECCV), 2013: 13-35. [17] 赵文涛, 曹昕鸷, 田志勇. 基于自适应阈值区域生长的红外舰船目标分割方法[J]. 红外技术, 2018, 40(2): 158-163. http://hwjs.nvir.cn/article/id/hwjs201802010ZHAO Wentao, CAO Xinzhi, TIAN Zhiyong. Infrared ship target segmentation based on adaptive threshold region growth[J]. Infrared Technology, 2018, 40(2): 158-163. http://hwjs.nvir.cn/article/id/hwjs201802010 [18] ZUCKER S W. Region growing: Childhood and adolescence[J]. Computer Graphics Image Processing, 1976(5): 382-399. http://ar.newsmth.net/att/dfd6fe6c43255/Region_growing_Childhood_and_adolescence.pdf [19] LI J, QIU M, ZHANG Y, et al. A fast obstacle detection method by fusion of double-layer region growing algorithm and grid-SECOND detector[J]. IEEE Access, 2020, 9: 32053-32063. http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9309211 [20] WANG W, WANG X J, LIU X W, et al. Image segmentation algorithm based on image complexity[J]. Journal of Detection & Control, 2015, 37(3): 5-9. -
点击查看大图
图(5) / 表(1)
计量
- 文章访问数: 332
- HTML全文浏览量: 27
- PDF下载量: 23
- 被引次数: 0
