SIA OpenIR  > 机器人学研究室
Alternative TitleMulti-input fusion adversarial network for underwater image enhancement
林森1,2,3; 刘世本1; 唐延东2,3
Source Publication红外与激光工程
Indexed ByEI ; CSCD
EI Accession number20202408812394
Contribution Rank1
Funding Organization国家自然科学基金(61473280,91648118) ; 辽宁省自然科学基金面上项目(2015020100) ; 辽宁省教育厅科研项目(LJ2019JL022)
Keyword深度学习 生成对抗网络 多输入融合 编码解码框架 水下图像增强

针对水下图像出现对比度低、颜色偏差和细节模糊等问题,提出了多输入融合对抗网络进行水下图像增强。该方法主要特点是生成网络采用编码解码结构,通过卷积层滤除噪声,利用反卷积层恢复丢失的细节并逐像素进行细化图像。首先,对原始图像进行预处理,得到颜色校正和对比度增强两种类型图像。其次,利用生成网络学习两种增强图像与原始图像之间差异的置信度图。然后,为减少在生成网络学习过程中两种增强算法引入的伪影和细节模糊,添加了纹理提取单元对两种增强图像进行纹理特征提取,并将提取的纹理特征与对应的置信度图进行融合。最后,通过构建多个损失函数,反复训练对抗网络,得到增强的水下图像。实验结果表明,增强的水下图像色彩鲜明并且对比度提升,评价指标UCIQE均值为0.639 9,NIQE均值为3.727 3。相比于其他算法有显著优势,证明了该算法的良好效果。

Other Abstract

For underwater image of low contrast, color deviation and blurred details and other issues, the multiinput fusion adversarial networks was proposed to enhance underwater images. The main feature of this method was that the generative network used encoding and decoding structure, filtering noise through convolution layer, recovering lost details through deconvolution layer and refining the image pixel by pixel. Firstly, the original image was preprocessed to obtain two types of images: color correction and contrast enhancement. Secondly, the confidence graph of the difference between the two enhanced images and the original image was learned by using the generated network. Then, in order to reduce artifacts and details blur introduced by the two enhancement algorithms in the process of generating network learning, the texture extraction unit was added to extract texture features from the two enhanced images, and the extracted texture features were fused with the corresponding confidence map. Finally, the enhanced underwater image was obtained by constructing multiple loss functions and training the adversarial network repeatedly. The experimental results show that the enhanced underwater image has bright color and improved contrast, the average value of UCIQE and NIQE is 0.639 9 and 3.727 3 respectively. Compared with other algorithms, the algorithm has significant advantages and proves its good effect.

Citation statistics
Document Type期刊论文
Corresponding Author林森
Recommended Citation
GB/T 7714
林森,刘世本,唐延东. 多输入融合对抗网络的水下图像增强[J]. 红外与激光工程,2020,49(5):1-9.
APA 林森,刘世本,&唐延东.(2020).多输入融合对抗网络的水下图像增强.红外与激光工程,49(5),1-9.
MLA 林森,et al."多输入融合对抗网络的水下图像增强".红外与激光工程 49.5(2020):1-9.
Files in This Item:
File Name/Size DocType Version Access License
多输入融合对抗网络的水下图像增强.pdf(2774KB)期刊论文出版稿开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[林森]'s Articles
[刘世本]'s Articles
[唐延东]'s Articles
Baidu academic
Similar articles in Baidu academic
[林森]'s Articles
[刘世本]'s Articles
[唐延东]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[林森]'s Articles
[刘世本]'s Articles
[唐延东]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 多输入融合对抗网络的水下图像增强.pdf
Format: Adobe PDF
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.