一种用于图像去雾的改进生成对抗网络

doi:10.19678/j.issn.1000-3428.0061823

摘要/Abstract

摘要： 雾霾天气会使计算机视觉相关系统获取到的图像质量下降并影响系统的正常工作。传统图像去雾算法通过分析大量图像建立模型，并找出清晰图像与模型之间的映射关系，但该类算法大多存在颜色失真和图像失真的问题，且在某些特定场景下可能失效。基于卷积神经网络的去雾算法得到的图像质量相对较好，但是该类算法对数据集要求较高，普遍需要成对数据进行训练，而获取同一时刻和场景下的有雾图像与无雾图像难度较高。提出一种基于循环生成对抗网络（CycleGAN）的改进图像去雾算法，该算法无需使用成对的数据进行训练。通过优化生成器生成的无雾图像与真实无雾图像之间的颜色损失，使得生成器能够生成与无雾图像具有相同颜色分布的图像。同时，通过向2个生成器中分别添加对应目标域图像的输入并引入特征损失函数，以解决经典CycleGAN在图像转换时存在的图像失真问题，从而更好地还原原始图像的细节特征。实验结果表明，相较DCP、CycleGAN、AOD-Net、Cycle-dehaze等算法，该算法的结构相似度提高4.3%~23.0%，峰值信噪比提高2.3%~36.9%，其能取得更好的去雾效果。

关键词: 图像去雾, 图像复原, 图像增强, 深度学习, 生成对抗网络

Abstract: Hazy weather degrades the quality of images obtained by computer vision related systems and affects the normal operation of the system.However, most traditional algorithms establish the relationship between the clear image model and image distortion, and most can determine the relationship between the specific image model and image distortion.The image quality obtained by the dehazing algorithm based on Convolutional Neural Network(CNN) is relatively high, but this type of algorithm has high dataset requirements and typically requires training on the data in pairs.However, it is difficult to obtain the hazy image and the haze-free image of the same scene at the same time.An improved image dehazing algorithm based on Cycle Generative Adversarial Network(CycleGAN) is proposed.The algorithm does not need to use paired data for training.By optimizing the color loss between the haze-free image generated by the generator and the real haze-free image, the generator can generate an image with the same color distribution as the haze-free image.Moreover, by adding the input of the corresponding target domain image to the two generators and introducing the feature loss function, the image distortion problem in the image conversion of classical CycleGAN is solved, improving the restoration of the detailed features of the original image.The results show that the structural similarity of the algorithm improved by 4.3%~23.0% compared with DCP, CycleGAN, AOD-Net, Cycle-dehaze, and other algorithms, and the Peak Signal-to-Noise Ratio(PSNR) improved by 2.3%~36.9%, thereby achieving an improved dehazing effect.

Key words: image dehazing, image restoration, image enhancement, deep learning, Generative Adversarial Network(GAN)

中图分类号:

TP391.4

丁泳钧, 黄山. 一种用于图像去雾的改进生成对抗网络[J]. 计算机工程, 2022, 48(6): 207-212.

DING Yongjun, HUANG Shan. An Improved Generative Adversarial Network for Image Dehazing[J]. Computer Engineering, 2022, 48(6): 207-212.

https://www.ecice06.com/CN/Y2022/V48/I6/207

图/表 7

20220625175831

20220625175835

20220625175838

20220625175842

20220625175846

20220625175850

20220625175854

参考文献

[1] 李彦冬.基于卷积神经网络的计算机视觉关键技术研究[D].成都:电子科技大学, 2017. LI Y D.Convolutional neural network based research on image understanding[D].Chengdu:University of Electronic Science and Technology of China, 2017.(in Chinese)
[2] 雷飞, 朱林, 王雪丽.改进多尺度Retinex色彩灰度的水下彩色图像研究[J].小型微型计算机系统, 2018, 39(1):185-188. LEI F, ZHU L, WANG X L.Improved multi scale retinex enhancement technology on color underwater image[J].Journal of Chinese Computer Systems, 2018, 39(1):185-188.(in Chinese)
[3] KUMAR R, KAUSHIK B K, BALASUBRAMANIAN R.Multispectral transmission map fusion method and architecture for image dehazing[J].IEEE Transactions on Very Large Scale Integration(VLSI) Systems, 2019, 27(11):2693-2697.
[4] 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.
[5] 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.
[6] BERMAN D, TREIBITZ T, AVIDAN S.Non-local image dehazing[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2016:1674-1682.
[7] GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al.Generative adversarial networks[J].Communications of the ACM, 2020, 63(11):139-144.
[8] 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.
[9] ZHANG H, PATEL V M.Densely connected pyramid dehazing network[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:3194-3203.
[10] ZHANG H, SINDAGI V, PATEL V M.Joint transmission map estimation and dehazing using deep networks[J].IEEE Transactions on Circuits and Systems for Video Technology, 2020, 30(7):1975-1986.
[11] ENGIN D, GENC A, EKENEL H K.Cycle-dehaze:enhanced CycleGAN for single image dehazing[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:938-9388.
[12] YANG D, SUN J.Proximal Dehaze-Net:a prior learning-based deep network for single image dehazing[M].Berlin, Germany:Springer, 2018.
[13] LI B Y, PENG X L, WANG Z Y, et al.AOD-Net:all-in-one dehazing network[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C., USA:IEEE Press, 2017:4780-4788.
[14] ZHANG X Y, DONG H, HU Z, et al.Gated fusion network for degraded image super resolution[J].International Journal of Computer Vision, 2020, 128(6):1699-1721.
[15] 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.Washington D.C., USA:IEEE Press, 2017:2242-2251.
[16] SWAMI K, DAS S K.CANDY:conditional adversarial networks based end-to-end system for single image haze removal[C]//Proceedings of the 24th International Conference on Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:3061-3067.
[17] ZHANG H.Learning-based methods for single image restoration and translation[EB/OL].[2021-05-05].https://rucore.libraries.rutgers.edu/rutgers-lib/60116/.
[18] WANG R X, ZHANG Q, FU C W, et al.Underexposed photo enhancement using deep illumination estimation[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2019:6842-6850.
[19] TAIGMAN Y, POLYAK A, WOLF L.Unsupervised cross-domain image generation[EB/OL].[2021-05-05].https://arxiv.org/pdf/1611.02200.pdf.
[20] SILBERMAN N, HOIEM D, KOHLI P, et al.Indoor segmentation and support inference from RGBD images[EB/OL].[2021-05-05].https://cs.nyu.edu/~silberman/papers/indoor_seg_support.pdf.
[21] WANG Z, SIMONCELLI E P, BOVIK A C.Multiscale structural similarity for image quality assessment[C]//Proceedings of the 37th Asilomar Conference on Signals, Systems & Computers.Washington D.C., USA:IEEE Press, 2003:1398-1402.
[22] WANG Z, BOVIK A C, SHEIKH H R, et al.Image quality assessment:from error visibility to structural similarity[J].IEEE Transactions on Image Processing, 2004, 13(4):600-612.
[23] 张文博.基于先验知识的图像去雾算法[D].成都:电子科技大学, 2017. ZHANG W B.Image dehazing algorithm based on prior knowledge[D].Chengdu:University of Electronic Science and Technology of China, 2017.(in Chinese)
[24] HAN K, WANG Y H, TIAN Q, et al.GhostNet:more features from cheap operations[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2020:1577-1586.

选择文件类型/文献管理软件名称

选择包含的内容