首页 > 论文 > 激光与光电子学进展 > 57卷 > 20期(pp:201004--1)

基于注意力机制和Retinex的低照度图像增强方法

Low-Illumination Image Enhancement Method Based on Attention Mechanism and Retinex

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

摘要

低照度图像增强的主要目的是提升图像的整体光照度,进而呈现更多有用的信息。针对低照度图像的整体照度低、对比度弱和噪声较高的问题,提出基于注意力机制和Retinex算法的低照度图像增强方法。该方法首先将低照度图像分解为不变性反射图和缓变平滑性光照图;再通过注意力机制提取图像的空间和局部物体信息,从而能够保证增强过程中利用空间和局部物体信息进行约束;同时增加色彩损失函数改善图像饱和度,用以补偿和校准增强过程中的对比度细节;改进低照度图像和合成方法,加入真实噪声有效扩充训练数据集。最终在LOL和SID数据集上实验表明,所提方法的主观感受和客观评价指标均有所提升。

Abstract

The goal of low-illuminance image enhancement is to increase the overall illuminance of the image, thereby presenting more useful information. Aiming at the problems of low illumination, low contrast and high noise in low-illumination images, a method of image enhancement method based on attention mechanism and Retinex algorithm is proposed. This method first decomposes the low-illumintion image into an invariant reflection map and a slowly-varying smooth illumination map. Then, it uses the attention mechanism to extract the spatial and local object information of the image, so as to ensure that the spatial and local object information is used for constraints during the enhancement process. Moreover, it increases the color loss function to improve the image saturation to compensate and calibrate the contrast details in the enhancement process. Furthermore, it improves the low-illumintion image and synthesis method, add real noise, and efficiently expands the training data set. Finally, the experiments on the LOL and SID data sets show that the subjective and objective evaluation indicators of the proposed method improved.

广告组1 - 空间光调制器+DMD
补充资料

中图分类号:TN911.73

DOI:10.3788/LOP57.201004

所属栏目:图像处理

收稿日期:2020-01-16

修改稿日期:2020-02-24

网络出版日期:2020-10-01

作者单位    点击查看

黄辉先:湘潭大学信息工程学院, 湖南 湘潭 411105
陈凡浩:湘潭大学信息工程学院, 湖南 湘潭 411105

联系人作者:陈凡浩(chenfanhaohaooo@163.com)

【1】Ren Y F. Application of histogram equalization in image processing Science and Technology Information[J]. 0, 2007(4): 37-38.
任艳斐. 直方图均衡化在图像处理中的应用 科技信息[J]. 0, 2007(4): 37-38.

【2】Pan Q, Yin J. An image enhancement algorithm based on weighted constraint decision [J]. Control Engineering of China. 2018, 25(11): 2017-2021.
潘强, 印鉴. 基于权重约束决策的图像增强算法 [J]. 控制工程. 2018, 25(11): 2017-2021.

【3】He K M, Sun J, Tang X O. Single image haze removal using dark channel prior [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2011, 33(12): 2341-2353.

【4】Mei Y J, Ning Y, Chen J J. Block-adjusted image enhancement algorithm combining dark channel prior with MSRCR [J]. Acta Photonica Sinica. 2019, 48(7): 0710005.
梅英杰, 宁媛, 陈进军. 融合暗通道先验和MSRCR的分块调节图像增强算法 [J]. 光子学报. 2019, 48(7): 0710005.

【5】Li L, Wang H G, Liu X. Underwater image enhancement based on improved dark channel prior and color correction [J]. Acta Optica Sinica. 2017, 37(12): 1211003.
李黎, 王惠刚, 刘星. 基于改进暗原色先验和颜色校正的水下图像增强 [J]. 光学学报. 2017, 37(12): 1211003.

【6】Wang X, Zhu X C, Ning C, et al. Combination of dark-channel prior with sparse representation for underwater image restoration [J]. Journal of Electronics & Information Technology. 2018, 40(2): 264-271.
王鑫, 朱行成, 宁晨, 等. 融合暗原色先验和稀疏表示的水下图像复原 [J]. 电子与信息学报. 2018, 40(2): 264-271.

【7】Sima Z L, Hu F. Low-light image enhancement method based on simulating multi-exposure fusion [J]. Journal of Computer Applications. 2019, 39(6): 1804-1809.
司马紫菱, 胡峰. 基于模拟多曝光融合的低照度图像增强方法 [J]. 计算机应用. 2019, 39(6): 1804-1809.

【8】Zotin A. Fast algorithm of image enhancement based on multi-scale Retinex [J]. Procedia Computer Science. 2018, 131: 6-14.

【9】Zhang H Y, Hu W B. Spatio-temporal context tracking algorithm based on Retinex-enhanced gray information and color information [J]. Journal of Computer-Aided Design & Computer Graphics. 2017, 29(12): 2323-2329.
张红颖, 胡文博. 基于Retinex灰度增强和颜色信息的时空上下文跟踪算法 [J]. 计算机辅助设计与图形学学报. 2017, 29(12): 2323-2329.

【10】Lore K G, Akintayo A, Sarkar S. LLNet: a deep autoencoder approach to natural low-light image enhancement [J]. Pattern Recognition. 2017, 61: 650-662.

【11】Wei C, Wang W J, Yang W H, et al. -08-01) https:∥www.researchgate.net/publication/327033239_Deep_Retinex_Decomposition_for_Low-Light_ [2020-01-15]. Enhancement. 2018.

【12】LeCun Y, Boser B E, Denker J S, et al. Backpropagation applied to handwritten zip code recognition [J]. Neural Computation. 1989, 1(4): 541-551.

【13】Shen X, Cheng X H, Wang X Z. Infrared dim-small object detection algorithm based on adaptive scale local contrast enhancement combined with visual attention mechanism [J]. Infrared Technology. 2019, 41(8): 764-771.
沈旭, 程小辉, 王新政. 结合视觉注意力机制基于尺度自适应局部对比度增强的红外弱小目标检测算法 [J]. 红外技术. 2019, 41(8): 764-771.

【14】Wan C, You Q J, Sun J, et al. Retinal image quality assessment based on FA-Net [J]. Chinese Journal of Experimental Ophthalmology. 2019, 37(8): 608-612.
万程, 游齐靖, 孙晶, 等. 基于FA-Net的视网膜眼底图像质量评估 [J]. 中华实验眼科杂志. 2019, 37(8): 608-612.

【15】Stentiford F W M. Attention-based similarity measure with application to content-based information retrieval [J]. Proceedings of SPIE. 2003, 5021: 221-232.

【16】Dabov K, Foi A, Katkovnik V, et al. Image denoising with block-matching and 3D filtering [J]. Proceedings of SPIE. 2006, 6064: 606414.

【17】Ronneberger O, Fischer P, Brox T. U-net: convolutional networks for biomedical image segmentation [M]. ∥Navab N, Hornegger J, Wells W, et al. Medical image computing and computer- MICCAI 2015. Lecture notes in computer science. Cham: Springer. 2015, 9351: 234-241.

【18】Dony R D, Wesolkowski S. Edge detection on color images using RGB vector angles . [C]∥1999 IEEE Canadian Conference on Electrical and Computer Engineering, May 9-12, 1999, Edmonton, Alberta, Canada. New York: IEEE. 1999, 6505379.

【19】Ma K D, Zeng K, Wang Z. Perceptual quality assessment for multi-exposure image fusion [J]. IEEE Transactions on Image Processing. 2015, 24(11): 3345-3356.

【20】Guo X J, Li Y, Ling H B. LIME: low-light image enhancement via illumination map estimation [J]. IEEE Transactions on Image Processing. 2017, 26(2): 982-993.

【21】Lee C, Lee C, Kim C S. Contrast enhancement based on layered difference representation . [C]∥2012 19th IEEE International Conference on Image Processing, September 30-October 3, 2012, Orlando, FL, USA. New York: IEEE. 2012, 965-968.

【22】Wang S H, Zheng J, Hu H M, et al. Naturalness preserved enhancement algorithm for non-uniform illumination images [J]. IEEE Transactions on Image Processing. 2013, 22(9): 3538-3548.

引用该论文

Huang Huixian,Chen Fanhao. Low-Illumination Image Enhancement Method Based on Attention Mechanism and Retinex[J]. Laser & Optoelectronics Progress, 2020, 57(20): 201004

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): 201004

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