计算机科学 ›› 2021, Vol. 48 ›› Issue (6A): 581-586.doi: 10.11896/jsjkx.200500026

• 交叉&应用 • 上一篇    下一篇

局部细粒度信息引导的双循环一致性绝缘子缺陷样本生成

赵潇1, 李仕林2, 李凡3, 余正涛3, 张林华1, 杨勇2   

  1. 1 云南电网有限责任公司楚雄供电局 云南 楚雄675000
    2 云南电网有限责任公司电力科学研究院 昆明650217
    3 昆明理工大学信息工程与自动化学院 昆明650500
  • 出版日期:2021-06-10 发布日期:2021-06-17
  • 通讯作者: 李凡(478263823@qq.com)
  • 作者简介:81878618@qq.com
  • 基金资助:
    国家重点研发计划项目(2018YFC0830105,2018YFC0830100);云南电网公司科技项目(YNKJXM20190729)

Double-cycle Consistent Insulator Defect Sample Generation Method Based on Local Fine-grainedInformation Guidance

ZHAO Xiao1, LI Shi-lin2, LI Fan3, YU Zheng-tao3, ZHANG Lin-hua1, YANG Yong2   

  1. 1 Chuxiong Power Supply Bureau of Yunnan Power Grid Co.,Ltd.Chuxiong,Yunnan 675000,China
    2 Eleictric Power Reasearch Institute of Yunnan Power Grid Co.,Ltd.,Kunming 650217,China
    3 Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,China
  • Online:2021-06-10 Published:2021-06-17
  • About author:ZHAO Xiao,born in 1982,undergraduate,engineer.His main research interests include transmission line operation and maintenance.
    LI Fan,born in 1986,Ph.D,lecturer,master's supervisor,is a member of China Computer Federation.His main research interests include picture processing and image recognition
  • Supported by:
    National Key R&D Program of China(2018YFC0830105,2018YFC0830100) and Science and Technology Project of Yunnan Power Grid Company(YNKJXM20190729).

摘要: 针对绝缘子缺陷样本数据缺乏,现有生成方法又要求训练样本的规模庞大,且在生成过程中绝缘子缺陷的细节常常被丢失或扭曲,提出了一种基于局部细粒度信息引导的双循环一致性绝缘子缺陷样本生成方法。该方法利用粗糙绝缘子图像作为网络输入,提出通过循环一致性生成对抗方法向精细缺陷绝缘子样本学习,生成较为逼真的缺陷样本。为使生成的样本具有丰富的缺陷特征,提出将生成图像中的缺陷区域图像作为判别网络的输入,并利用对抗约束的方式引导生成网络重点关注缺陷的细粒度信息,从而进一步提升生成绝缘子缺陷样本的真实性和多样性。与现有方法相比,所提方法构建的绝缘子缺陷样本数据集具有逼真、多样化等特点,为提升绝缘子缺陷自动识别的准确性提供了重要的数据基础。

关键词: 绝缘子, 样本生成, 局部细粒度信息, 双循环一致性, 对抗式网络

Abstract: In view of the lack of data for insulator defects samples,the existing generation methods require a large number of training samples,and the details of insulator defects are often lost or distorted during the generation process.This paper presents a double-cycle consistent insulator defect sample generation method based on local fine-grained information guidance (LFGI-DCC).The approach uses the rough insulator image as the network input,and learns from the fine defect insulator sample through the cycle consistency generative adversarial method to generate more realistic defect insulator samples.At the same time,the image of the defect area in the generated image is used as the input to discriminator,and the generation network is guided to focus on the fine-grained information of the defect by the method of resisting constraints,thereby further improving the authenti-city and diversity of the insulator defect samples.Compared with the existing methods,the insulator defect dataset constructed by the proposed method has the characteristics of fidelity and diversification,which provides an important foundation for improving the accuracy of insulator defect automatic identification.

Key words: Insulator, Sample generation, Local fine-grained information, Double-cycle consistency, Adversarial network

中图分类号: 

  • TP183
[1] GOODFELLOW I,POUGET-ABADIE J,MIRZAM,et al.Generative adversarial nets[C]//Advances in Neural Information Processing Systems.2014:2672-2680.
[2] MAO X,LI Q,XIE H,et al.Least squares generative adversarial networks[C]//Proceedings of the IEEE International Conference on Computer Vision.2017:2794-2802.
[3] ZHANG H,XU T,LIH,et al.Stackgan:Text to photo-realistic image synthesis with stacked generative adversarial networks[C]//Proceedings of the IEEE International Conference on Computer Vision.2017:5907-5915.
[4] NIE D,TRULLO R,LIAN J,et al.Medical image synthesiswith context-aware generative adversarialnetworks[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention.Springer,Cham,2017:417-425.
[5] YADAV S,CHEN C,ROSS A.Synthesizing Iris Images using RaSGAN with Application in Presentation Attack Detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops.2019:2422-2430.
[6] CHEN X,DUAN Y,HOUTHOOFTR,et al.Infogan:Interpretable representation learning by information maximizing generative adversarial nets[C]//Advances in Neural Information Processing Systems.2016:2172-2180.
[7] ZHANG G,KAN M,SHAN S,et al.Generative adversarial network with spatial attention for face attribute editing[C]//Proceedings of the European Conference on Computer Vision (ECCV).2018:417-432.
[8] ARJOVSKY M,CHINTALA S,BOTTOU L.Wasserstein gan[J].arXiv:1701.07875,2017.
[9] ZHAO J,MATHIEU M,LECUN Y.Energy-based generative adversarial network[J].arXiv:1609.03126,2016.
[10] ZHU J Y,PARK T,ISOLA P,et al.Unpaired image-to-image translation usingcycle-consistent adversarial networks[C]//Proceedings of the IEEE International Conference on Computer Vision.2017:2223-2232.
[11] ZHONG Z,ZHENG L,ZHENG Z,et al.Camera style adaptation for person re-identification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:5157-5166.
[12] MA J,YU W,LIANG P,et al.FusionGAN:A generative adversarial network for infrared and visible image fusion[J].Information Fusion,2019,48:11-26.
[13] MIRZA M,OSINDERO S.Conditional generative adversarialnets[J].arXiv:1411.1784,2014.
[14] ZHAO H,SHI J,QI X,et al.Pyramid scene parsing network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:2881-2890.
[15] GULRAJANI I,AHMED F,ARJOVSKY M,et al.Improvedtraining of wassersteingans[C]//Advances in neural information processing systems.2017:5767-5777.
[16] GAO F,YANG Y,WANG J,et al.A deep convolutional generative adversarial networks (DCGANs)-based semi-supervised method for object recognition in synthetic aperture radar (SAR) images[J].Remote Sensing,2018,10(6):846.
[17] BROCK A,DONAHUE J,SIMONYAN K.Large scale gantraining for high fidelity natural image synthesis[J].arXiv:1809.11096,2018.
[18] LI Y,XIAO N,OUYANG W.Improved boundary equilibriumgenerative adversarial networks[J].IEEE Access,2018,6:11342-11348.
[19] ENGIN D,GENÇ A,KEMAL EKENEL H.Cycle-dehaze:En-hanced cyclegan for single image dehazing[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops.2018:825-833.
[20] DENG W,ZHENG L,YE Q,et al.Image-image domain adaptation with preserved self-similarity and domain-dissimilarity for person re-identification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:994-1003.
[21] WEI L,ZHANG S,GAO W,et al.Person transfer gan to bridge
domain gap for person re-identification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:79-88.
[1] 方挺,韩家明. 航拍图像中绝缘子缺陷的检测与定位[J]. 计算机科学, 2016, 43(Z6): 222-225.
[2] 何洪英,钱艳萍,王玲,罗滇生. 基于像素同龄组和相邻组的绝缘子去噪方法[J]. 计算机科学, 2012, 39(6): 283-284.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 方磊, 武泽慧, 魏强. 二进制代码相似性检测技术综述[J]. 计算机科学, 2021, 48(5): 1 -8 .
[2] 朱雨, 庞建民, 徐金龙, 陶小涵, 王军. 面向SW26010处理器的三维Stencil自适应分块参数算法[J]. 计算机科学, 2021, 48(6): 10 -18 .
[3] 冯芙蓉, 张兆功. 目标轮廓检测技术新进展[J]. 计算机科学, 2021, 48(6A): 1 -9 .
[4] 孙正, 张小雪. 生物光声成像中声反射伪影抑制方法的研究进展[J]. 计算机科学, 2021, 48(6A): 10 -14 .
[5] 周欣, 刘硕迪, 潘薇, 陈媛媛. 自然交通场景中的车辆颜色识别[J]. 计算机科学, 2021, 48(6A): 15 -20 .
[6] 黄雪冰, 魏佳艺, 沈文宇, 凌力. 基于自适应加权重复值滤波和同态滤波的MR图像增强[J]. 计算机科学, 2021, 48(6A): 21 -27 .
[7] 江妍, 马瑜, 梁远哲, 王原, 李光昊, 马鼎. 基于分数阶麻雀搜索优化OTSU肺组织分割算法[J]. 计算机科学, 2021, 48(6A): 28 -32 .
[8] 张子丞, 谭志苇, 张晨瑞, 王旋, 刘晓璇, 俞一彪. 基于高低频带对数能量谱比贝叶斯决策的语音端点检测[J]. 计算机科学, 2021, 48(6A): 33 -37 .
[9] 崔雯昊, 蒋慕蓉, 杨磊, 傅鹏铭, 朱凌霄. 结合MCycleGAN与RFCNN实现太阳斑点图高分辨重建[J]. 计算机科学, 2021, 48(6A): 38 -42 .
[10] 田洋, 毕秀丽, 肖斌, 李伟生, 马建峰. 基于离散切比雪夫变换的图像接缝裁剪篡改检测[J]. 计算机科学, 2021, 48(6A): 43 -50 .