|本期目录/Table of Contents|

[1]汪然然,娄联堂*.基于图像分析和深度学习的复合绝缘子憎水性分级[J].武汉工程大学学报,2021,43(05):580-585.[doi:10.19843/j.cnki.CN42-1779/TQ. 202106003]
 WANG Ranran,LOU Liantang*.Hydrophobicity Classification of Composite Insulators Based on Image Analysis and Deep Learning[J].Journal of Wuhan Institute of Technology,2021,43(05):580-585.[doi:10.19843/j.cnki.CN42-1779/TQ. 202106003]
点击复制

基于图像分析和深度学习的复合绝缘子憎水性分级(/HTML)
分享到:

《武汉工程大学学报》[ISSN:1674-2869/CN:42-1779/TQ]

卷:
43
期数:
2021年05期
页码:
580-585
栏目:
机电与信息工程
出版日期:
2021-10-31

文章信息/Info

Title:
Hydrophobicity Classification of Composite Insulators Based on Image Analysis and Deep Learning
文章编号:
1674 - 2869(2021)05 - 0580 - 06
作者:
汪然然娄联堂*
中南民族大学数学与统计学学院,湖北 武汉 430074
Author(s):
WANG Ranran LOU Liantang*
College of Mathematics and Statistics, South-Central University for Nationalities, Wuhan 430074, China
关键词:
图像分析深度学习神经网络U-Net网络
Keywords:
image analysis deep learning neural network U-Net network
分类号:
TP391.4
DOI:
10.19843/j.cnki.CN42-1779/TQ. 202106003
文献标志码:
A
摘要:
为了更加方便快捷地检测大量复合绝缘子憎水性等级,提出一种基于图像分析和深度学习的复合绝缘子憎水性分级方法。首先为提高图像对比度,对复合绝缘子憎水性图像进行灰度化和图像增强处理;其次利用图像分析技术和U-Net网络提取水珠轮廓,得到水珠轮廓图像;接着引入深度卷积神经网络,将这些水珠轮廓图作为神经网络的输入,以相应的憎水性等级作为输出向量,训练网络得到分级模型;最后将分级模型用于憎水性分级,得到分级结果。实验结果表明:该方法的分级结果已达到实际应用要求,水珠轮廓提取的精度达到了92.96%,分级准确率达到了90.2%,预测一幅图像的憎水性等级平均耗时0.1 s。
Abstract:
To detect the hydrophobicity level of a large number of composite insulators more conveniently and quickly, a method for grading the hydrophobicity of composite insulators based on image analysis and deep learning was proposed. First, to improve the image contrast, the water droplet images of composite insulators were gray-scaled and image-enchanced. Then the image analysis technology and U-Net network were used to extract the contour of the water droplet to obtain the contour image of the water droplets. After that, the deep convolutional neural network was introduced, while these water droplets contours acted as the input of the neural network and the corresponding hydrophobic level as the output vectors, and the network was trained to obtain the classification model. Finally, the classification model was used for the hydrophobicity classification to obtain the classification results. The experimental results showed that the classification results of this method have reached the practical application requirements, the accuracy of water droplets contour extraction and the grading accuracy rate reach 92.96% and 90.2%, respectively; and it takes an average of 0.1 s to predict the hydrophobicity of an image.

参考文献/References:

[1] CHEN X, LI C, HUANG X, et al. On-line estimating the level of hydrophobicity of composite insulators using the digital images[C]//IEEE. Proceedings-Electrical Insulation Conference and Electrical Manufacturing Expo. Indianapolis: IEEE, 2005: 216-221. [2] 张鹏, 胡攀峰, 陈潇一, 等. 高聚物复合绝缘子憎水性能的研究[J]. 智慧电力, 2020, 48(12):97-103. [3] 徐志钮,律方成,赵鹏,等. 拟合方法用于硅橡胶静态接触角的测量[J]. 高电压技术, 2009, 35(10):2475-2480. [4] 梁英,李成榕,丁立健,等. 电晕对HTV硅橡胶憎水性恢复的影响[J]. 高电压技术, 2008,34(1):30-32,40. [5] 宋伟,赵林杰,李成榕,等. 复合绝缘子在线检测技术的发展[J]. 高电压技术, 2005, 31(5):28-30. [6] 冈萨雷斯. 数字图像处理[M]. 2版. 阮秋琦,阮宇智,译. 北京:电子工业出版社, 2007. [7] 钱磊. 复合绝缘子憎水性水珠图像识别与分级[D]. 武汉:中南民族大学,2015. [8] 潘龙,张烜,郝斐, 等. 基于深度卷积神经网络的绝缘子憎水性图像识别方法[J]. 电工技术, 2019(13):30-33. [9] 李转. 基于Canny算子和神经网络复合绝缘子憎水性研究[D]. 郑州:华北水利水电大学,2017. [10] RONNEBERGER O, FISCCHER P, BROX T. U-net: convolutional networks for biomedical image segmentation [C]// Springer. Medical Image Computing and Computer-Assisted Intervention- MICCAI 2015. Cham: Springer International Publishing, 2015: 234-241. [11] 朱琳琳,韩璐,杜泓,等. 基于U-Net网络的多主动轮廓细胞分割方法研究[J]. 红外与激光工程, 2020, 49(增刊1):151-159. [12] 汪佛池,闫康,张重远,等. 采用图像分析与神经网络识别绝缘子憎水性[J]. 电机与控制学报,2014,18(11):78-83. [13] SZEGEDY C , WEI L , JIA Y , et al. Going deeper with convolutions[C]//IEEE. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)Computer Society. Boston: IEEE, 2015:1-9. [14] 韩要昌, 王洁, 史通,等. 基于改进GoogLeNet的遥感图像分类方法[J]. 弹箭与制导学报,2019,39(5): 139-142,153. [15] KOLLKALIS C C A, TASAKOS T, KONTARGYRI V T, et al. Hydrophobicity classification of composite insulators based on convolutional neural networks[J]. Engineering Applications of Artificial Intelligence, 2020,91:103613.1-103613.10.

相似文献/References:

[1]洪汉玉,章秀华,程莉,等.道路病害形态特征的图像分析[J].武汉工程大学学报,2014,(04):70.[doi:103969/jissn16742869201404015]
 HONG Han yu,ZHANG Xiu hua,CHENG Li,et al.Image analysis method for road disease morphology characteristic[J].Journal of Wuhan Institute of Technology,2014,(05):70.[doi:103969/jissn16742869201404015]
[2]刘德真,李圆媛*.基于深度学习和多组学数据的肺腺癌分期预测研究[J].武汉工程大学学报,2024,46(02):190.[doi:10.19843/j.cnki.CN42-1779/TQ.202307022]
 LIU Dezhen,LI Yuanyuan*.Stage prediction of lung adenocarcinoma based on deep learning andmulti-omics data[J].Journal of Wuhan Institute of Technology,2024,46(05):190.[doi:10.19843/j.cnki.CN42-1779/TQ.202307022]
[3]蒋珺阳,吴晶华*,赵娜娜.基于大核注意力改进的工业管件位姿估计[J].武汉工程大学学报,2024,46(03):304.[doi:10.19843/j.cnki.CN42-1779/TQ.202310018]
 JIANG Junyang,WU Jinghua*,ZHAO Nana.Pose estimation of industrial pipe fittings based onlarge kernel attention improvement[J].Journal of Wuhan Institute of Technology,2024,46(05):304.[doi:10.19843/j.cnki.CN42-1779/TQ.202310018]
[4]朱柏霖,卢 涛*,王依伊,等.实际场景人脸超分辨率算法综述[J].武汉工程大学学报,2024,46(05):564.[doi:10.19843/j.cnki.CN42-1779/TQ.202211016]
 ZHU Bolin,LU Tao*,WANG Yiyi,et al.Review of real-world face super-resolution algorithms[J].Journal of Wuhan Institute of Technology,2024,46(05):564.[doi:10.19843/j.cnki.CN42-1779/TQ.202211016]

备注/Memo

备注/Memo:
收稿日期:2021-06-04基金项目:国家自然科学基金(60975011)作者简介:汪然然,硕士研究生。 E-mail:[email protected]*通讯作者:娄联堂,博士,教授。Email:[email protected]引文格式:汪然然,娄联堂. 基于图像分析和深度学习的复合绝缘子憎水性分级[J]. 武汉工程大学学报,2021,43(5):580-585.
更新日期/Last Update: 2021-10-27