首页 > 论文 > 激光与光电子学进展 > 56卷 > 4期(pp:41004--1)

基于改进生成对抗网络的水下激光图像后向散射光修复方法

Backscattered Light Repairing Method for Underwater Laser Image Based on Improved Generative Adversarial Network

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

为提高水下激光图像的质量,改进了生成对抗网络的生成网络,使其成为一种包含跳跃结构和空洞卷积的深度卷积神经网络。利用该网络从自建数据集中学习待修复图像到目标图像的端到端映射参数,再对带有强后向散射光的水下激光图像进行修复。实验结果表明,所提方法能够快速对后向散射光区域进行填充修复,相比传统去噪和增强对比度方法联合处理的结果,所提方法的峰值信噪比平均提高了9.10 dB,特征相似度平均提高了0.11,实现了水下激光图像的去噪、对比度增强和非均匀性照明改善,较好地去除了后向散射光。

Abstract

To improve the qualities of underwater laser images, the generator network of the generative adversarial network is changed to be a deep convolution neural network which contains the jumping structure and the dilated convolution. The network is used to learn the end-to-end parameters which map the unrepaired images to the target images from the self-built data set and repair underwater laser images with strong backscatter light. Compared with the experiment results of the joint processing of the classic denoising method and the contrast enhancement method, the proposed method can fill and repair the backscattered light area. The peak signal to noise ratio obtained by the proposed method increases by an average of 9.10 dB and the feature similarity increases by an average of 0.11. The denoising, enhancing contrast, improving the non-uniform illumination are achieved and the backscattered light is removed well.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TP183

DOI:10.3788/lop56.041004

所属栏目:图像处理

基金项目:军内科研项目(417210751)

收稿日期:2018-08-23

修改稿日期:2018-09-02

网络出版日期:2018-09-06

作者单位    点击查看

张清博:海军工程大学兵器工程学院, 湖北 武汉 430033
张晓晖:海军工程大学兵器工程学院, 湖北 武汉 430033
韩宏伟:海军工程大学兵器工程学院, 湖北 武汉 430033

联系人作者:张清博(527992400@qq.com)

【1】Wang L, Xu Z Y, Zhang Q H, et al. Design and experiment of range-gated underwater laser imaging system[J]. Laser & Optoelectronics Progress, 2012, 49(5): 051101.
王磊, 徐智勇, 张启衡, 等. 距离选通水下激光成像系统设计及实验[J]. 激光与光电子学进展, 2012, 49(5): 051101.

【2】Lu M T, Su X Y, Cao Y P, et al. 3D shape reconstruction algorithms for modulation measuring profilometry with synchronous scanning[J]. Chinese Journal of Lasers, 2016, 43(3): 0308006.
卢明腾, 苏显渝, 曹益平, 等. 同步扫描的调制度测量轮廓术三维面形重建算法[J]. 中国激光, 2016, 43(3): 0308006.

【3】Yang L M, Liang J, Zhang W F, et al. Underwater polarimetric imaging target enhancement technology based on unpolarized illumination[J]. Acta Optica Sinica, 2018, 38(6): 0611003.
杨力铭, 梁健, 张文飞, 等. 基于非偏振光照明的水下偏振成像目标增强技术[J]. 光学学报, 2018, 38(6): 0611003.

【4】Huang Z J, Xiao J, Lu A J, et al. Comparison and research on several image denoising methods[J]. Communications Technology, 2017, 50(11): 2465-2471.
黄子吉, 肖杰, 陆安江, 等. 几种图像去噪方法的比较研究[J]. 通信技术, 2017, 50(11): 2465-2471.

【5】Dabov K, Foi A, Katkovnik V, et al. Image denoising by sparse 3D transform-domain collaborative filtering[J]. IEEE Transactions on Image Processing, 2007, 16(8): 2080-2095.

【6】Han H W, Zhang X H. Study of enhancement technology in underwater laser imaging[J]. Laser & Infrared, 2007, 37(10): 1105-1108.
韩宏伟, 张晓晖. 水下激光图像增强方法研究[J]. 激光与红外, 2007, 37(10): 1105-1108.

【7】Gan Y L, Li F D. Noise filtering and simulation for underwater image[J]. Laser & Infrared, 2013, 43(3): 306-310.
淦元柳, 李富栋. 水下图像的噪声滤波与仿真[J]. 激光与红外, 2013, 43(3): 306-310.

【8】Xie J Y, Xu L L, Chen E H. Image denoising and inpainting with deep neural networks[C]∥NIPS''12 Proceedings of the 25th International Conference on Neural Information Processing Systems, December 03-06, 2012, Lake Tahoe, Nevada. New York: ACM, 2012: 341-349.

【9】Han H W, Zhang X H, Ge W L. Study of the mode and maximum detecting performance of underwater laser range-gated imaging system[J]. Chinese Journal of Lasers, 2011, 38(1): 0109001.
韩宏伟, 张晓晖, 葛卫龙. 水下激光距离选通成像系统的模型与极限探测性能研究[J]. 中国激光, 2011, 38(1): 0109001.

【10】Liu X M, Zhang M D, Sun X H. A new kind of underwater laser imaging system[J]. Chinese Journal of Lasers, 2000, 27(3): 206-210.
刘雪明, 张明德, 孙小菡. 一种新型水下激光成像系统[J]. 中国激光, 2000, 27(3): 206-210.

【11】Wen D H, Jiang Y S, Zhang X G, et al. Speckle characteristics of laser radar polarization imaging and despecking algorithm[J]. Acta Optica Sinica, 2013, 33(3): 0311003.
闻东海, 江月松, 张绪国, 等. 激光偏振成像散斑统计特性和抑制方法研究[J]. 光学学报, 2013, 33(3): 0311003.

【12】Goodfellow I J, Pouget-Abadie J, Mirza M, et al. Generative adversarial nets[C/OL]∥International Conference on Neural Information Processing Systems, June 10, 2014.[S.l.:s.n.], 2014: 2672-2680.

【13】Yeh R A, Chen C, Lim T Y,et al. Semantic image inpainting with deep generative models[C]∥IEEE Conference on Computer Vision and Pattern Recognition, July 21-26, 2017, Honolulu, HI, USA. New York: IEEE, 2017: 6882-6890.

【14】Iizuka S, Simo-Serra E, Ishikawa H. Globally and locally consistent image completion[J]. ACM Transactions on Graphics, 2017, 36(4): 1-14.

【15】Liu Y, Pan J, Su Z. Deep blind image inpainting[EB/OL]. (2017-12-27)[2018-08-11].

【16】Ratliff L J, Burden S A, Sastry S S. Characterizationand computation of local nash equilibria in continuous games[C]∥Annual Allerton Conference on Communication Control and Computing, October 2-4, 2013,Monticello, IL, USA. New York: IEEE, 2013: 917-924.

【17】Yu F, Koltun V. Multi-scale context aggregation by dilated convolutions[J]. International Conference on Learning Representations, 2016, 13(8): 1123-1130.

【18】Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[EB/OL]. (2015-04-10)[2018-08-12].

【19】Li J, Wang J P, Wan G T, et al. Novel algorithm for image enhancement with histogram equalization and MSRCR[J]. Journal of Xidian University(Natural Science Edition), 2014, 41(3): 103-109.
李锦, 王俊平, 万国挺,等. 一种结合直方图均衡化和MSRCR的图像增强新算法[J]. 西安电子科技大学学报(自然科学版), 2014, 41(3): 103-109.

【20】Zhang L, Zhang L, Mou X Q, et al. A comprehensive evaluation of full reference image quality assessment algorithms[C]∥IEEE International Conference on Image Processing, September 30- October 3, 2012, Orlando, FL, USA. New York: IEEE, 2012: 1477-1480.

引用该论文

Zhang Qingbo,Zhang Xiaohui,Han Hongwei. Backscattered Light Repairing Method for Underwater Laser Image Based on Improved Generative Adversarial Network[J]. Laser & Optoelectronics Progress, 2019, 56(4): 041004

张清博,张晓晖,韩宏伟. 基于改进生成对抗网络的水下激光图像后向散射光修复方法[J]. 激光与光电子学进展, 2019, 56(4): 041004

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF