Image Super-resolution Reconstructing based on Generative Adversarial Network

被引:1
|
作者
Nan Jing [1 ]
Bo Lei [1 ]
机构
[1] Wuhan Natl Lab Optoelect, Huazhong Inst Electroopt, 717th Yangguang Ave, Wuhan, Hubei, Peoples R China
来源
关键词
Image super-resolution; Generative Adversarial Network; Reconstructing; ResNet;
D O I
10.1117/12.2547435
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Image super-resolution refers to the technique of reconstructing a high-resolution image by processing one or more complementary low-resolution images. It is widely used in medical imaging, video surveillance, remote sensing imaging and other fields. The learning-based super-resolution algorithm obtains a mapping relationship between the high-resolution image and the low-resolution image by learning, and then guides the generation of the high-resolution image according to the obtained mapping relationship. The generative adversarial network (GAN) is composed of a generative network model and a discriminator network model, and the two play each other until the Nash equilibrium is reached, and the texture information and the high-frequency details of the downsampled image can be restored based on the super-resolution method of generative adversarial network. However, super-resolution algorithms based on generative adversarial network can only be used for one kind of magnification time, and the versatility is insufficient. Despite convolutional neural networks has achieved breakthroughs in accuracy and speed of traditional single-frame super-resolution reconstruction, and can achieve a higher peak signal-to-noise ratio (PSNR). Most of them use Mean Square Error (MSE) as the minimum optimization objective function, so although a higher peak signal-to-noise ratio can be achieved, when the image downsampling factors is higher, the reconstructed image will be too smooth, lack high-frequency details and perceptually unsatisfy in the sense that they fail to match the fidelity expected at the higher resolution. When dealing with complex data of real scenes, the model's representation ability is not high; and the generative adversarial network training is very unstable, seriously affecting the model training process. This paper is based on generative adversarial network, improving the network structure and optimizing the training method to improve the quality of generating images. The following improvements have been made to the generator model: the multi-level structure is used to enlarge the image step by step, so that the model can simultaneously generate multiple images with a larger scale, and also ensure that the image obtained at a larger magnification has higher quality; ResNet model is improved by recursive learning and residual learning, and the batch normalization structure in the model is removed. On the basis of ensuring the image quality, the efficiency of the model is effectively improved. The recursive and residual learning methods can effectively improve the feature expression ability of the model, and thus significantly improve the quality of the generated image. The Expand-Squeeze method is proposed to generate images. The basic idea is to expand the dimension of the last layer of the convolution layer of the model. In this way, more context information is obtained, and then the image is generated by using the lx1 convolution kernel. The Expand-Squeeze method can effectively reduce the checkerboard effect and improve the quality of the generated image to some extent. This paper improves the discriminator network loss function. Measure the similarity between generated image and real image by introducing Wasserstein distance. The loss function proposed consists of two parts: the loss function of resistance and the loss function of content. The experimental results verify that the improved generation of the generative adversarial network can effectively improve the quality of the generated image and effectively improve the stability of the model training.
引用
收藏
页数:8
相关论文
共 50 条
  • [1] Mars image super-resolution based on generative adversarial network
    Wang, Cong
    Zhang, Yin
    Zhang, Yongqiang
    Tian, Rui
    Ding, Mingli
    [J]. IEEE Access, 2021, 9 : 108889 - 108898
  • [2] Image super-resolution based on conditional generative adversarial network
    Gao, Hongxia
    Chen, Zhanhong
    Huang, Binyang
    Chen, Jiahe
    Li, Zhifu
    [J]. IET IMAGE PROCESSING, 2020, 14 (13) : 3006 - 3013
  • [3] Image Super-Resolution Reconstruction Based on a Generative Adversarial Network
    Wu, Yun
    Lan, Lin
    Long, Huiyun
    Kong, Guangqian
    Duan, Xun
    Xu, Changzhuan
    [J]. IEEE ACCESS, 2020, 8 : 215133 - 215144
  • [4] Mars Image Super-Resolution Based on Generative Adversarial Network
    Wang, Cong
    Zhang, Yin
    Zhang, Yongqiang
    Tian, Rui
    Ding, Mingli
    [J]. IEEE ACCESS, 2021, 9 : 108889 - 108898
  • [5] Image super-resolution reconstruction based on improved generative adversarial network
    Wang, Ying-Li
    Li, Xiao-Jing
    Ma, Hong-Bin
    Ding, Qun
    Pirouz, Matin
    Ma, Qi-Tao
    [J]. Journal of Network Intelligence, 2021, 6 (02): : 155 - 163
  • [6] Super-Resolution of Text Image Based on Conditional Generative Adversarial Network
    Wang, Yuyang
    Ding, Wenjun
    Su, Feng
    [J]. ADVANCES IN MULTIMEDIA INFORMATION PROCESSING, PT III, 2018, 11166 : 270 - 281
  • [7] Image Super-resolution Reconstruction Based on an Improved Generative Adversarial Network
    Liu, Han
    Wang, Fan
    Liu, Lijun
    [J]. 2019 1ST INTERNATIONAL CONFERENCE ON INDUSTRIAL ARTIFICIAL INTELLIGENCE (IAI 2019), 2019,
  • [8] Improved generative adversarial network for retinal image super-resolution
    Qiu, Defu
    Cheng, Yuhu
    Wang, Xuesong
    [J]. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2022, 225
  • [9] Spatial Transformer Generative Adversarial Network for Image Super-Resolution
    Rempakos, Pantelis
    Vrigkas, Michalis
    Plissiti, Marina E.
    Nikou, Christophoros
    [J]. IMAGE ANALYSIS AND PROCESSING, ICIAP 2023, PT I, 2023, 14233 : 399 - 411
  • [10] A lightweight generative adversarial network for single image super-resolution
    Lu, Xinbiao
    Xie, Xupeng
    Ye, Chunlin
    Xing, Hao
    Liu, Zecheng
    Cai, Changchun
    [J]. VISUAL COMPUTER, 2024, 40 (01): : 41 - 52