视觉注意机制的注意残差稠密神经网络弱光照图像增强

邓箴; 王一斌; 刘立波

doi:10.37188/CJLCD.2021-0098

您当前的位置：

首页 >

文章列表页 >

视觉注意机制的注意残差稠密神经网络弱光照图像增强

图像增强与分割 | 更新时间：2021-11-08

- 视觉注意机制的注意残差稠密神经网络弱光照图像增强
- Attentive residual dense network of visual attention mechanism for weakly illuminated image enhancement
- 液晶与显示 2021年36卷第11期页码：1463-1473
- 作者机构：
  
  1.宁夏大学信息工程学院, 宁夏银川 750021
  2.四川师范大学工学院, 四川成都 610000
- 作者简介：
  
  [ "邓箴(1984-)，女，河南三门峡人，博士研究生，讲师，2006年于北京科技大学获得硕士学位，主要从事图像处理、机器视觉方面的研究。E-mail: dengzhen1025@163.com" ]
  [ "王一斌(1982-)，男，河南漯河人，博士，讲师，2016年于西北工业大学获得博士学位，主要从事机器视觉、模式识别方面的研究。E-mail: yibeen.wong@gmail.com" ]
  [ "刘立波(1974-)，女，宁夏平罗人，博士，教授，2010年于中国农业科学院获得博士学位，2012年于北京理工大学博士后出站，主要从事计算机视觉、智能信息处理等方面的研究。E-mail: liulib@163.com" ]
- 基金信息：
  
  宁夏回族自治区自然科学基金(2020AAC03031);宁夏回族自治区青年托举人才工程(2017);国家自然基金(61862050)
- DOI：10.37188/CJLCD.2021-0098
  中图分类号： TP391.4
- 收稿日期：2021-04-13，
  
  修回日期：2021-05-14，
  
  纸质出版日期：2021-11
- 稿件说明：
移动端阅览
邓箴, 王一斌, 刘立波. 视觉注意机制的注意残差稠密神经网络弱光照图像增强[J]. 液晶与显示, 2021,36(11):1463-1473.

Zhen DENG, Yi-bin WANG, Li-bo LIU. Attentive residual dense network of visual attention mechanism for weakly illuminated image enhancement[J]. Chinese journal of liquid crystals and displays, 2021, 36(11): 1463-1473.
邓箴, 王一斌, 刘立波. 视觉注意机制的注意残差稠密神经网络弱光照图像增强[J]. 液晶与显示, 2021,36(11):1463-1473. DOI： 10.37188/CJLCD.2021-0098.

Zhen DENG, Yi-bin WANG, Li-bo LIU. Attentive residual dense network of visual attention mechanism for weakly illuminated image enhancement[J]. Chinese journal of liquid crystals and displays, 2021, 36(11): 1463-1473. DOI： 10.37188/CJLCD.2021-0098.

摘要

针对传统的弱光照图像增强算法鲁棒性差，基于神经网络的图像增强算法直接从弱光照图像中估计增强结果，并未注入视觉注意机制，不能有效注意弱光照区域，导致算法增强结果的精度不高等问题，本文提出了注意残差稠密神经网络的弱光照图像增强算法来提高弱光照图像的增强精度和视觉效果。该算法主要包括注意循环网络和残差稠密网络，注意循环网络在光照图的引导下，利用循环网络结构逐步关注图像中的弱光照区域，从而产生由粗到细，逐步优化的光照注意图。而光照注意图则进一步联合弱光照图像作为后续的残差稠密网络的输入，引导残差稠密网络为弱光照区域分配更多的计算资源，更好地学习弱光照图像与增强图像的映射关系，得到准确的图像增强结果。实验表明，本文算法在合成图像及真实图像上均较常用算法有更好的增强效果。

Abstract

Weakly illuminated image enhancement as a kind of pre-processing technologies is widely used in various computer vision tasks. Traditional image enhancement methods has poor robustness. While

methods based on existing convolutional neural networks (CNNs) estimate the enhanced image from weakly illuminated image directly without injecting the visual attention mechanism

ignoring weakly illuminated regions and leading to inaccuracy result. To resolve this problem

we propose an attentive residual dense network for weakly illuminated image enhancement. The proposed network contains two parts: attentive recurrent network and residual dense network. With the guidance of the illumination map

attentive recurrent network focuses more and more on the weakly illuminated regions and generates the attentive illumination map following a coarse-to-fine strategy via the recurrent architecture. This attentive illumination map concatenated with the weakly illuminated image are injected into the subsequent residual dense network to ensure this network assign more computational source to weakly illuminated regions and estimate enhanced image accurately. The experiments demonstrate that our method achieves favorable performance against that of existing image enhancement methods based on synthetic images and real images.

关键词

Keywords

references

S M PIZER , E P AMBURN , J D AUSTIN , 等 . Adaptive histogram equalization and its variations . Computer Vision, Graphics, and Image Processing , 1987 . 39 ( 3 ): 355 - 368 . DOI: 10.1016/S0734-189X(87)80186-X http://doi.org/10.1016/S0734-189X(87)80186-X .

P KANDHWAY , A K BHANDARI , A SINGH . A novel reformed histogram equalization based medical image contrast enhancement using krill herd optimization . Biomedical Signal Processing and Control , 2020 . 56 101677 DOI: 10.1016/j.bspc.2019.101677 http://doi.org/10.1016/j.bspc.2019.101677 .

DONG X, WANG G, PANG Y, et al . Fast efficient algorithm for enhancement of low lighting video[C]// Proceedings of IEEE International Conference on Multimedia and Expo . Barcelona: IEEE, 2011: 1-6.

LI L, WANG R G, WANG W M, et al . A low-light image enhancement method for both denoising and contrast enlarging[C]// Proceedings of IEEE International Conference on Image Processing ( ICIP ). Quebec City: IEEE, 2015: 3730-3734.

YANG J, JIANG X W, PAN C H, et al . Enhancement of low light level images with coupled dictionary learning[C]// International Conference on Pattern Recognition ( ICPR ). Cancun: IEEE, 2016: 751-756.

X M FENG , J J LI , Z HUA . Low-light image enhancement algorithm based on an atmospheric physical model . Multimedia Tools and Applications , 2020 . 79 ( 43/44 ): 32973 - 32997 . DOI: 10.1007/s11042-020-09562-6 http://doi.org/10.1007/s11042-020-09562-6 .

X J GUO , Y LI , H B LING . LIME: low-light image enhancement via illumination map estimation . IEEE Transactions on Image Processing , 2017 . 26 ( 2 ): 982 - 993 . DOI: 10.1109/TIP.2016.2639450 http://doi.org/10.1109/TIP.2016.2639450 .

赵晨 , 吕国强 , 吴磊 , 等 . 基于图像局部亮度的液晶显示器动态调光算法 . 液晶与显示 , 2020 . 35 ( 3 ): 234 - 241 . http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20203503.0234 http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20203503.0234 .

C ZHAO , G Q LV , L WU , 等 . Dynamic dimming algorithm for liquid crystal display based on local brightness of image . Chinese Journal of Liquid Crystals and Displays , 2020 . 35 ( 3 ): 234 - 241 . http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20203503.0234 http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20203503.0234 .

X Y FU , Y H LIAO , D L ZENG , 等 . A probabilistic method for image enhancement with simultaneous illumination and reflectance estimation . IEEE Transactions on Image Processing , 2015 . 24 ( 12 ): 4965 - 4977 . DOI: 10.1109/TIP.2015.2474701 http://doi.org/10.1109/TIP.2015.2474701 .

FU X Y, ZENG D L, HUANG Y, et al . A weighted variational model for simultaneous reflectance and illumination estimation[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 2782-2790.

程亚亚 , 于化东 , 于占江 , 等 . 微铣刀同轴全息图像增强方法 . 中国光学 , 2020 . 13 ( 4 ): 705 - 712 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202004006.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202004006.htm .

Y Y CHENG , H D YU , Z J YU , 等 . Method of enhancing the quality of in-line holographic images for micro-milling tool . Chinese Optics , 2020 . 13 ( 4 ): 705 - 712 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202004006.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202004006.htm .

林剑萍 , 廖一鹏 . 结合分数阶微分及Retinex的NSCT自适应低照度图像增强 . 液晶与显示 , 2020 . 35 ( 4 ): 360 - 373 . http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20203504.0360 http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20203504.0360 .

J P LIN , Y P LIAO . NSCT adaptive low illumination image enhancement combining fractional differential and Retinex . Chinese Journal of Liquid Crystals and Displays , 2020 . 35 ( 4 ): 360 - 373 . http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20203504.0360 http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20203504.0360 .

HUANG H, TAO H J, WANG H F. A convolutional neural network based method for low-illumination image enhancement[C]//2019 2 nd International Conference on Artificial Intelligence and Pattern Recognition . Beijing, China: Association for Computing Machinery, 2019: 72-77.

K G LORE , A AKINTAYO , S SARKAR . LLNet: a deep autoencoder approach to natural low-light image enhancement . Pattern Recognition , 2017 . 61 650 - 662 . DOI: 10.1016/j.patcog.2016.06.008 http://doi.org/10.1016/j.patcog.2016.06.008 .

X D KUANG , X B SUI , Y LIU , 等 . Single infrared image enhancement using a deep convolutional neural network . Neurocomputing , 2019 . 332 119 - 128 . http://www.onacademic.com/detail/journal_1000041592173599_514e.html http://www.onacademic.com/detail/journal_1000041592173599_514e.html .

J FENG , X L FENG , J T CHEN , 等 . Generative adversarial networks based on collaborative learning and attention mechanism for hyperspectral image classification . Remote Sensing , 2020 . 12 ( 7 ): 1149 DOI: 10.3390/rs12071149 http://doi.org/10.3390/rs12071149 .

L He , J C W CHAN , Z M WANG . Automatic depression recognition using CNN with attention mechanism from videos . Neurocomputing , 2021 . 422 165 - 175 . DOI: 10.1016/j.neucom.2020.10.015 http://doi.org/10.1016/j.neucom.2020.10.015 .

CHENG Y, YAN J, WANG Z. Enhancement of weakly illuminated images by deep fusion networks[C]// Processing of 2019 IEEE International Conference on Image Processing ( ICIP ). Taipei: IEEE, 2019: 924-928.

K GU , G T ZHAI , W S LIN , 等 . The analysis of image contrast: from quality assessment to automatic enhancement . IEEE Transactions on Cybernetics , 2016 . 46 ( 1 ): 284 - 297 . DOI: 10.1109/TCYB.2015.2401732 http://doi.org/10.1109/TCYB.2015.2401732 .

K D MA , Z F DUANMU , Q B WU , 等 . Waterloo exploration database: new challenges for image quality assessment models . IEEE Transactions on Image Processing , 2017 . 26 ( 2 ): 1004 - 1016 . DOI: 10.1109/TIP.2016.2631888 http://doi.org/10.1109/TIP.2016.2631888 .

LEE C, LEE C, KIM C S. Contrast enhancement based on layered difference representation[C]// Processing of the IEEE Conference on Image Processing . Orlando: IEEE, 2012: 965-968.

NASA. Retinex image processing[EB/OL]. [2018-10-20] . http://dragon.larc.nasa.gov/retinex/pao/news/ http://dragon.larc.nasa.gov/retinex/pao/news/ .

浏览量

136

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于CNN的异构FPGA硬件加速器设计

残差学习与层注意力相结合的轻量级图像超分辨

基于CNN-Transformer结构的遥感影像变化检测

基于P-U-net的角锥波前探测器的波前校正方法