基于小波变换和改进Retinex-Net相结合的巡检图像增强方法
作者单位:

1.国网金华供电公司;2.中国计量大学;3.金华八达集团有限公司

基金项目:

国家重点研发计划子课题,项目编号:2021YFF0603702;金华八达集团有限公司科技项目,项目编号:BD2022JH-KXXM007


A Method of Patrol Image Enhancement Based on Wavelet Transform and Retinex-Net
Author:
Affiliation:

1.State Grid Jinhua Power Supply Company;2.School of Mechanical and Electrical Engineering, China Jiliang University;3.Jinhua Bada Group Co., Ltd

Fund Project:

National Key Research and Development Program of China No.2021YFF0603702;Jinhua Bada Group Co.,Ltd. Science and technology project No. Bd2022JH-KXXM007

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [20]
  • | |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    针对无人机拍摄的巡检低质量图像存在的光照异常、杆塔主体特征不明显等问题,提出一种基于小波变换和改进Retinex-Net相结合的图像增强方法。首先,利用小波变换将低质量图像分解为低高频图像,利用改进Retinex-Net网络处理低频图像,引入ASPP模块和SE模块强化网络特征提取能力,利用跳跃链接结构和最近邻插值法进行特征图缩放以减少背景噪声干扰;使用限制对比度自适应直方图均衡算法(CLAHE)增强高频图像,降低高频噪声干扰。最后,利用小波重构得到增强后的低质量图像。在自建数据集上的实验结果表明,相较于HE、MSRCR等算法,本文算法可以增强图像细节边缘特征、还原图像颜色、提高分辨率,帮助输电线路运维人员监测,提高分析工作的准确性。

    Abstract:

    Aiming at the problems of abnormal illumination and unobvious main features of the tower in the low-quality inspection image taken by UAV, an image enhancement method based on wavelet transform and improved Retinex-Net is proposed. Firstly, the low-quality image is decomposed into low-frequency images by wavelet transform, and the low-frequency images are processed by the improved Retinex-Net network. ASPP module and SE module are introduced to enhance the network feature extraction ability, and the feature map is scaled by using jump link structure and nearest neighbor interpolation method to reduce background noise interference. The contrast-limited adaptive histogram equalization algorithm (CLAHE) is used to enhance high-frequency images and reduce high-frequency noise interference. Finally, the enhanced low-quality image is obtained by wavelet reconstruction. Experimental results on self-built data sets show that, compared with HE and MSRCR algorithms, the proposed algorithm can enhance the edge features of image details, restore the image color, improve the resolution, help transmission line operators to monitor and improve the accuracy of analysis.

    参考文献
    [1] 韩冰,尚方.面向无人机输电线路巡检的电力杆塔检测框架模型[J].浙江电力,2016,35(04):6-11.Han Bing, Shang Fang. Power Tower Inspection Framework Model for UAV Transmission Line Inspection [J]. Zhejiang Electric Power, 2016,35(04):6-11.
    [2] Bian J, Hui X, Zhao X, et al. A monocular vision–based perception approach for unmanned aerial vehicle close proximity transmission tower inspection. International Journal of Advanced Robotic Systems, 2019, 16(1): 1729881418820227.
    [3] 戴昊,崔志文,谢志毅,等.复杂天气下的巡检图像增强算法研究[J].机械设计与制造工程,2021,50(10):105-109.Dai Hao, Cui Zhiwen, Xie Zhiyi, et al. Research on patrol image enhancement algorithm in complex weather [J]. Mechanical Design and Manufacturing Engineering, 2021, 50(10): 105-109.
    [4] 白万荣,张驯,朱小琴,等.基于E-FCNN的电力巡检图像增强[J].中国电力,2021,54(05):179-185.Bai Wanrong, Zhang Xun, Zhu Xiaoqin, et al. Image enhancement of power inspection based on E-FCNN [J]. China Electric Power, 2021,54(05):179-185.
    [5] 秦钟,杨建国,王海默,等.基于Retinex理论的低照度下输电线路图像增强方法及应用[J].电力系统保护与控制,2021,49(03):150-157..Qin Zhong, Yang Jianguo, Wang Haimo, et al. Image enhancement method and application of transmission lines in low illumination based on Retinex theory [J]. Power System Protection and Control, 2021,49(03):150-157. ..
    [6] 王浩,张叶,沈宏海,等.图像增强算法综述[J].中国光学,2017,10(04):438-448.Wang Hao, Zhang Ye, Shen Honghai, et al. Overview of image enhancement algorithms [J]. China Optics, 2017,10(04):438-448.
    [7] 马文君,刘金虎,王小鹏,等.结合Lab空间和单尺度Retinex的自适应图像去雾算法[J].应用光学,2020,41(01):100-106.Ma Wenjun, Liu Jinhu, Wang Xiaopeng, et al. Adaptive image defogging algorithm combining Lab space and single-scale Retinex [J]. Applied Optics, 2020,41(01):100-106.
    [8] 史金余,郝明良,邹沛煜.基于颜色校正和引导滤波分层的水下图像增强[J].计算机应用与软件,2022,39(06):203-209.Shi Jinyu, Hao Mingliang, Zou Peiyu. Underwater image enhancement based on color correction and guided filtering [J]. computer applications and software, 2022,39(06):203-209.
    [9] 董丽丽,丁畅,许文海.基于直方图均衡化图像增强的两种改进方法[J].电子学报,2018,46(10):2367-2375.Dong Lili, Ding Chang, Xu Wenhai. Two improvedmethods of image enhancement based on histogram equalization [J]. Acta Electronica Sinica, 2018,46(10):2367-2375.
    [10] 陈小晴. 基于偏微分方程的图像去噪方法研究[D].南京信息工程大学,2013.Chen Xiaoqing. Research on image denoising method based on partial differential equation [D]. Nanjing University of Information Science and Technology, 2013.
    [11] 张振华,陆金桂.基于小波变换和改进的Retniex雾天图像增强[J].计算机应用与软件,2021,38(01):227-231.Zhang Zhenhua, Lu Jingui. Foggy image enhancement based on wavelet transform and improved Retniex [J]. computer applications and software, 2021,38(01):227-231.
    [12] 李庆忠,刘清.基于小波变换的低照度图像自适应增强算法[J].中国激光,2015,42(02):280-286.Li Qingzhong, Liu Qing. Adaptive enhancement algorithm of low illumination image based on wavelet transform [J]. China Laser, 2015,42(02):280-286.
    [13] 于天河,孟雪,潘婷,等.小波变换和自适应变换相结合的图像增强方法[J].哈尔滨理工大学学报,2018,23(06):100-104.Yu Tianhe, Meng Xue, Pan Ting, et al. Image enhancement method combining wavelet transform and adaptive transform [J]. Journal of harbin university of science and technology, 2018,23(06):100-104.
    [14] Huang Huang,Haijun Tao,Haifeng Wang.A Convolutional Neural Network Based Method for Low-illumination Image Enhancement[C].Proceedings of 2019 2nd International Conference on Artificial Intelligence and Pattern Recognition (AIPR 2019),2019:74-79.
    [15] SUN Feng,LI Bo. Low-light color image enhancement based on Retinex[C]//.Proceedings of 5th International Conference on Automation, Control and Robotics Engineering (CACRE 2020).2020:595-599.
    [16] 黄辉先,陈凡浩.基于注意力机制和Retinex的低照度图像增强方法[J].激光与光电子学进展,2020,57(20):53-60.Huang Huixian, Chen Fanhao. Low-illumination image enhancement method based on attention mechanism and Retinex [J]. Progress in Laser and Optoelectronics, 2020, 57(20): 53-60.
    [17] 马红强,马时平,许悦雷,等.基于深度卷积神经网络的低照度图像增强[J].光学学报,2019,39(02):99-108.Ma Hongqiang, Ma Shiping, Xu Yuelei, et al. Low illumination image enhancement based on deep convolution neural network [J]. Acta Optica Sinica, 2019,39(02):99-108.
    [18] Jie WANG,Ming-hua WAN,Yu-hui XU,et al. Image Processing Technology in Fog and Haze Environment Based on Retinex Theory[C]//.Proceedings of 2018 International Conference on Computational, Modeling, Simulation and Mathematical Statistics(CMSMS 2018).,2018:627-631.
    [19] Wei C ,? Wang W ,? Yang W , et al. Deep Retinex Decomposition for Low-Light Enhancement[J].? 2018.
    [20] Chen L C ,? Papandreou G ,? Kokkinos I , et al. DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs[J]. 2016(4).
    相似文献
    引证文献
    引证文献 [0] 您输入的地址无效!
    没有找到您想要的资源,您输入的路径无效!

    网友评论
    网友评论
    分享到微博
    发 布
引用本文
分享
文章指标
  • 点击次数:124
  • 下载次数: 0
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 收稿日期:2023-11-15
  • 最后修改日期:2023-11-27
  • 录用日期:2024-02-22
文章二维码