改进CycleGAN的图像去雾网络

付星堡; 蔡琼; 陈国清; 赖远哲; 陈玉

Chinese

您当前的位置：

首页 >

文章列表页 >

改进CycleGAN的图像去雾网络

更新时间：2026-09-04

- 改进CycleGAN的图像去雾网络
- Image dehazing network based on improved CycleGAN
- 微电子学与计算机 2021年38卷第8期页码：87-94
- 作者机构：
  
  1. 武汉工程大学机构计算机科学与工程学院,湖北,武汉,430205
  2. 武汉工程大学智能机器人湖北省重点实验室,湖北,武汉,430205
- 作者简介：
- 基金信息：
  
  武汉工程大学教育创新基金 CX2019234
- DOI：
  中图分类号：
- 纸质出版：2021
- 稿件说明：
移动端阅览
付星堡, 蔡琼, 陈国清, 赖远哲, 陈玉. 改进CycleGAN的图像去雾网络[J]. 微电子学与计算机, 2021, 38(8): 87-94.

FU Xingbao, CAI Qiong, CHEN Guoqing, et al. Image dehazing network based on improved CycleGAN[J]. 2021, 38(8): 87-94.
付星堡, 蔡琼, 陈国清, 赖远哲, 陈玉. 改进CycleGAN的图像去雾网络[J]. 微电子学与计算机, 2021, 38(8): 87-94. DOI：

FU Xingbao, CAI Qiong, CHEN Guoqing, et al. Image dehazing network based on improved CycleGAN[J]. 2021, 38(8): 87-94. DOI：

摘要

针对传统CycleGAN在图像去雾后出现模糊和颜色失真等问题，给出了一种改进CycleGAN的图像去雾网络.所提CycleGAN的生成器包括特征提取、特征融合和图像复原三个子网络.图像特征提取子网络用于提取图像的内容特征和风格特征，特征融合子网络利用两种不同的注意力机制分别对提取到的内容特征和风格特征进行融合，图像复原子网络将融合后的图像特征还原成无雾图像.与传统的CycleGAN和已有的去雾网络相比，所提网络对合成图像和真实图像均可取得理想的去雾结果，有效解决了传统CycleGAN在图像去雾后出现的模糊和颜色失真的问题.

Abstract

In view of the blur and color distortion in traditional CycleGAN image dehazing

animproved CycleGANImage dehazing network is proposed. The generator of present CycleGAN consists three parts: feature extraction sub-network

feature fusion sub-network and image restoration sub-network. The image feature extraction sub network is used to extract the content features and style features of the image.The feature fusion sub network uses two different attention mechanisms to fuse the extracted content features and style features. The image complex atom network restores the fused image features to a haze free image.Compared with traditional CycleGAN and existing defogging networks

the proposed network can achieve ideal dehazing results for both synthetic images and real images

and effectively solves the problems of blurring and color distortion in traditional CycleGAN images after dehazing.

关键词

Keywords

references

QIN X, WANG Z L, BAI Y C, et al. FFA-Net: feature fusion attention network for single image Dehazing[C]//Proceedings of the 34th AAAI Conference on Artificial Intelligence, AAAI 2020, the 32nd Innovative Applications of Artificial Intelligence Conference, IAAI 2020, the 10th AAAI Symposium on Educational Advances in Artificial Intelligence. New York, NY, USA: AAAI, 2020: 11908-11915.

吴迪, 朱青松. 图像去雾的最新研究进展[J]. 自动化学报, 2015, 41(2): 221-239. DOI: 10.16383/j.aas.2015.c131137. WU D, ZHU Q S. The latest research progress of image Dehazing[J]. Acta Automatica Sinica, 2015, 41(2): 221-239. DOI: 10.16383/j.aas.2015.c131137.

温立民, 巨永锋, 王会峰, 等. 基于双边滤波暗原色透射率细化的图像去雾算法[J]. 测控技术, 2020, 39(4): 62-65. DOI:10.19708/j.ckjs.2020.04.013. WEN L M, JU Y F, WANG H F, et al. Algorithm for Image fog removal of dark channel transmissivity refinement based on bilateral filtering[J]. Measurement % Control Technology, 2020, 39(4): 62-65. DOI:10.19708/j.ckjs.2020.04.013.

王小霞, 黄颖雯, 谭庭均, 等. 基于Retinex的汽车牌照图像去雾研究[J]. 内蒙古农业大学学报(自然科学版), 2018, 39(1): 64-70. DOI: 10.16853/j.cnki.1009-3575.2018.01.011. WANG X X, HUANG Y W, TAN T J, et al. Research on automobile license images defogging based on Retinex algorithm[J]. Journal of Inner Mongolia AgriculturalUniversity(Natural Science Edition), 2018, 39(1): 64-70. DOI: 10.16853/j.cnki.1009-3575.2018.01.011.

ZHU J Y, PARK T, ISOLA P, et al. Unpaired image-to-image translation using cycle-consistent adversarial networks[C]//Proceedings of 2017 IEEE International Conference on Computer Vision. Venice, Italy: IEEE, 2017: 2223-2232. DOI: 10.1109/ICCV.2017.244.

雍杨, 王敬儒, 张启衡. 弱小目标低对比度图像增强算法研究[J]. 激光与红外, 2005, 35(5): 370-373. DOI:10.3969/j.issn.1001-5078.2005.05.018. YONG Y, WANG J R, ZHANG Q H. Enhancement of low contrast image contain small target[J]. Laser % Infrared, 2005, 35(5): 370-373. DOI:10.3969/j.issn.1001-5078.2005.05.018.

KIM J, KIM M, KANG H, et al. U-GAT-IT: unsupervised generative attentional networks with adaptive layer-instance normalization for image-to-image translation[J]. arXiv: 1907.10830, 2019

WANG W C, YUAN X H, WU X J, et al. Fast image dehazing method based on linear transformation[J]. IEEE Transactions on Multimedia, 2017, 19(6): 1142-1155. DOI: 10.1109/Tmm。2017.2652069.

LI B Y, REN W Q, FU D P, et al. Benchmarking single-image dehazing and beyond[J]. IEEE Transactions on Image Processing, 2019, 28(1): 492-505. DOI: 10.1109/TIP.2018.2867951.

LI B Y, PENG X L, WANG Z Y, et al. AOD-Net: all-in-one Dehazing network[C]//Proceedings of 2017 IEEE international conference on computer vision. Venice, Italy: IEEE, 2017: 4770-4778. DOI: 10.1109/ICCV.2017.511.

CHEN Y, LAI Y K, LIU Y J. CartoonGAN: generative adversarial networks for photo cartoonization[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA: IEEE, 2018: 9465-9474. DOI: 10.1109/CVPR.2018.00986.

ZHU Q S, MAI J M, SHAO L. A fast single image haze removal algorithm using color attenuation prior[J]. IEEE Transactions on Image Processing, 2015, 24(11): 3522-3533. DOI: 10.1109/TIP.2015.2446191.

ZHANG Y L, TIAN Y P, KONG Y, et al. Residual dense network for image super-resolution[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA: IEEE, 2018: 2472-2481. DOI: 10.1109/CVPR.2018.00262.

WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Proceedings of the 15th European Conference on Computer Vision (ECCV). Munich, Germany: Springer, 2018: 3-19. DOI: 10.1007/978-3-030-01234-2_1.

TAN R T. Visibility in bad weather from a single image[C]//Proceedings of 2008 IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, AK, USA: IEEE, 2008: 1-8. DOI: 10.1109/CVPR.2008.4587643.

CAI B L, XU X M, JIA K, et al. Dehazenet: an end-to-end system for single image haze removal[J]. IEEE Transactions on Image Processing, 2016, 25(11): 5187-5198. DOI: 10.1109/TIP.2016.2598681.

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. DOI: 10.1109/TPAMI.2010.168.

浏览量

141

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MSS-YOLO-MPHands融合架构的装配手部动作识别研究

一种用于极低压boost系统的高效率电荷泵

基于多尺度融合与小目标增强的航拍车辆检测算法

基于反向体偏置的高增益稳定性高温运放设计

AlN/GaN混合缓冲层双异质结p-GaN HEMT直流特性仿真