Superpixel-Based Improved Fuzzy C-Means Clustering for Electrical Equipment Infrared Image Segmentation
-
摘要:
针对传统模糊C均值（FCM）算法在图像分割中存在分割精度低、收敛速度慢、对初始聚类中心选取不佳而陷入局部最优等问题，提出一种适用于电力设备红外图像的基于超像素的改进FCM分割方法。首先，采用多特征融合的简单非迭代聚类（SNIC）超像素算法对图像进行预分割，用超像素代替像素表达图像特征，降低后续处理复杂度；其次，运用最大类间方差的思想，选取类间方差最大时灰度直方图最大值对应的灰度值作为改进算法的初始聚类中心，避免生成局部最优解；最后，将多特征融合的SNIC算法与FCM算法结合，实现电力设备红外图像分割。实验结果表明：该算法在设备轮廓和局部高温区域的分割上改善了对比算法存在的欠分割现象，提升了运算效率，为后期电力设备故障诊断奠定基础。
Abstract:An improved super-pixel based FCM segmentation method for infrared image of power equipment is presented to solve the problems of low segmentation accuracy, slow convergence, poor selection of initial cluster centers and local optimization in traditional fuzzy C-mean (FCM) algorithm. First, a simple non-iterative clustering (SNIC) superpixel algorithm based on multi-feature fusion is used to pre-segment the image, and superpixels are used instead of pixels to express the image features, which reduces the subsequent processing complexity. Secondly, using the idea of maximizing the variance between classes, the gray value corresponding to the maximum value of gray histogram when the variance between classes is maximized is selected as the initial cluster center of the improved algorithm to avoid generating local optimal solution. Finally, combining the SNIC algorithm of multi-feature fusion with the FCM algorithm, the infrared image of power equipment is segmented. The experimental results show that the algorithm improves the under segmentation of the comparison algorithm on the contour of the device and the local high temperature area, improves the operation efficiency, and lays a foundation for the later fault diagnosis of power equipment.
-
Keywords:
- infrared image /
- Fuzzy c-means clustering /
- super-pixel /
- initial clustering center
-
-
![]()
图 2 各算法对隔离开关的分割结果
Figure 2. The segmentation results of the isolation switch of different algorithms
![]()
图 3 各算法对套管触头的分割结果
Figure 3. The segmentation results of the casing contact of different algorithms
![]()
图 4 各算法对发电机的分割结果
Figure 4. The segmentation results of the generator of different algorithms
![]()
图 5 各算法对设备线夹的分割结果
Figure 5. The segmentation results of the device clamp of different algorithms
表 1 各类红外图像分割效果
Table 1 Various infrared image segmentation effects
Image type FCM FCM_SICM FRFCM SFFCM Ours IOU DICE IOU DICE IOU DICE IOU DICE IOU DICE Isolation switch 0.7996 0.8587 0.9062 0.9368 0.8046 0.8649 0.9219 0.9548 0.9434 0.9765 Casing contact 0.8693 0.8947 0.8436 0.8768 0.9103 0.9471 0.9271 0.9528 0.9326 0.9573 Generator 0.8769 0.8945 0.8503 0.8793 0.8784 0.8972 0.9241 0.9451 0.9472 0.9642 Device clamp 0.8301 0.8794 0.8479 0.8839 0.2511 0.2949 0.2224 0.2648 0.9154 0.9547 
下载: 导出CSV
表 2 各算法平均运行时间
Table 2 Running mean time of different algorithms
Algorithm Mean time/s FCM 7.5763 FCM_SICM 23.6383 FRFCM 11.1746 SFFCM 5.4785 Ours 4.2954
下载: 导出CSV
-
[1] BU C, LIU T, LI R, et al. Infrared image segmentation algorithm based on multi structure morphology—pulse coupled neural network in application to the inspection of aerospace materials[J]. Russian Journal of Nondestructive Testing, 2022, 57(11): 1018-1026.
[2] FENG Hong, SONG Jie, HANG Meng, et al. A novel framework on intelligent detection for module defects of PV plant combining the visible and infrared images[J]. Solar Energy, 2022, 236: DOI: 10.1016/j.solener.2022.03.018
[3] 程子夜. 红外图像处理在架空线路故障诊断中的应用[D]. 淮南: 安徽理工大学, 2020. CHENG Ziye. Application of Infrared Image Processing in Fault Diagnosis of Overhead Lines[D]. Huainan: Anhui University of Science and Technology, 2020.
[4] 王晓飞, 胡凡奎, 黄硕. 基于分布信息直觉模糊c均值聚类的红外图像分割算法[J]. 通信学报, 2020, 41(5): 120-129. WANG Xiaofei, HU Fankui, HUANG Shuo. Infrared image segmentation algorithm based on distribution information intuitionistic fuzzy c-means clustering[J]. Journal on Communications, 2020, 41(5): 120-129.
[5] 刘沛津, 王曦, 贺宁. 改进GSO与二维OTSU融合的红外图像多阈值分割方法[J]. 应用光学, 2021, 42(4): 671-677. LIU Peijin, WANG Xi, HE Ning. Improved infrared image multi threshold segmentation method based on fusion of GSO and two-dimensional OTSU[J]. Journal of Applied Optics, 2021, 42(4): 671-677.
[6] 聂方彦, 李建奇, 屠添翼. 一种基于信息论距离的复杂图像分割方法[J]. 红外技术, 2018, 40(8): 786-791. http://hwjs.nvir.cn/article/id/hwjs201808011 NIE Fangyan, LI Jianqi, TU Tianyi. A complex image segmentation method based on information theory distance[J]. Infrared Technology, 2018, 40(8): 786-791. http://hwjs.nvir.cn/article/id/hwjs201808011
[7] XIAO Leyi, OUYANG Honglin, FAN Chaodong. An improved Otsu method for threshold segmentation based on set mapping and trapezoid region intercept histogram[J]. Optik, 2019, 196(C): 163106-163106.
[8] GUO H, CHEN P, HUANG S, et al. A threshold segmentation method for non-uniform illumination image based on brightness equalization[J]. IOP Conference Series: Materials Science and Engineering, 2019, 569: 052046. DOI: 10.1088/1757-899X/569/5/052046
[9] 戴若辰, 赵明富, 汤斌, 等. 基于Otsu分割与边缘检测的钢丝绳缺陷检测方法研究[J]. 激光与光电子学进展, 2021, 58(16): 566-573. DAI Ruochen, ZHAO Mingfu, TANG Bin, et al. Research on defect detection method of steel wire rope based on Otsu segmentation and edge detection[J]. Laser & Optoelectronics Progress, 2021, 58(16): 566-573.
[10] 陈露. 无人机载红外管道检测的研究与设计[D]. 芜湖: 安徽工程大学, 2018. CHEN Lu. Research and Design of Infrared Pipeline Detection for Unmanned Aerial Vehicles[D]. Wuhu: Anhui Polytechnic University, 2018.
[11] 黄新宇, 张洋, 王黎明, 等. 基于Mask-RCNN算法的复合绝缘子串红外图像分割与温度读取[J]. 高压电器, 2021, 57(9): 87-94. HUANG Xinyu, ZHANG Yang, WANG Liming, et al. Infrared image segmentation and temperature reading of composite insulator strings based on Mask RCNN algorithm[J]. High Voltage Apparatus, 2021, 57(9): 87-94.
[12] 袁刚, 许志浩, 康兵, 等. 基于DeepLabv3+网络的电流互感器红外图像分割方法[J]. 红外技术, 2021, 43(11): 1127-1134. http://hwjs.nvir.cn/article/id/b9df2f53-2244-471b-b0ec-42159cfaa654 YUAN Gang, XU Zhihao, KANG Bing, et al. Infrared image segmentation method for current transformers based on DeepLabv3+network[J]. Infrared Technology, 2021, 43(11): 1127-1134. http://hwjs.nvir.cn/article/id/b9df2f53-2244-471b-b0ec-42159cfaa654
[13] 赵祥宇, 陈沫涵. 基于聚类的图像分割方法综述[J]. 信息技术, 2018(6): 92-94, 99. ZHAO Xiangyu, CHEN Mohan. A review of clustering-based image segmentation methods[J]. Information Technology, 2018(6): 92-94, 99.
[14] 加小红. 聚类算法及其在图像分割中的应用研究[D]. 兰州: 兰州交通大学, 2017. JIA Xiaohong. Research on Clustering Algorithm and Its Application in Image Segmentation[D]. Lanzhou: Lanzhou Jiaotong University, 2017.
[15] ZADEH LA. Fuzzy sets[J]. Information and Control, 1965, 8(3): 338-353.
[16] WANG Qingsheng, WANG Xiaopeng, FANG Chao, et al. Robust fuzzy c-means clustering algorithm with adaptive spatial & intensity constraint and membership linking for noise image segmentation[J]. Applied Soft Computing Journal, 2020, 92: 106318.
[17] LEI Tao, JIA Xiaohong, ZHANG Yanning, et al. Significantly fast and robust fuzzy C-Means clustering algorithm based on morphological reconstruction and membership filtering[J]. IEEE Transactions on Fuzzy Systems, 2018, 26(5): 3027-3041.
[18] LEI Tao, JIA Xiaohong, ZHANG Yanning, et al. Superpixel-based fast fuzzy C-Means clustering for color image segmentation[J]. IEEE Trans. Fuzzy Systems, 2019, 27(9): 1753-1766.
[19] ACHANTA R, SUSSTRANK S. Superpixels and polygonsusing simple non-iterative clustering[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition, 2017: 4895-4904.
[20] 郑金云. 基于聚类的超像素分割方法研究[D]. 桂林: 桂林电子科技大学, 2021. ZHENG Jinyun. Research on Clustering-based Superpixel Segmentation Method[D]. Guilin: Guilin University Of Electronic Technology, 2021.
-
期刊类型引用(5)
1. 徐航，白晓东，张喜涛. 基于视线角速度精度的导引系统抗干扰性能评估指标研究. 航空兵器. 2024(01): 66-70 . 
百度学术
2. 张楠，陈长胜，孙靖国，梁雪超. 基于桶滚机动和诱饵投射的红外空空导弹对抗策略研究（英文）. 红外技术. 2022(03): 236-248 . 本站查看
3. 孙华为，周俊，冯颖. 基于红外干扰弹特征的目标综合复杂度算法研究. 光电技术应用. 2022(03): 43-47 . 百度学术
4. 孙志朋，赵世明，孙致月. 红外成像制导半实物仿真安装误差分析与修正. 红外技术. 2020(02): 139-143 . 本站查看
5. 张喜涛，白晓东，闫琳，王炜强. 红外空空导弹抗干扰性能评估指标体系研究. 红外技术. 2020(11): 1089-1094+1102 . 本站查看
其他类型引用(0)
计量
- 文章访问数: 34
- HTML全文浏览量: 1
- PDF下载量: 11
- 被引次数: 5
