| 143 | 0 | 130 |
| 下载次数 | 被引频次 | 阅读次数 |
目前图像超分辨率重建技术已在医疗成像、视频监控和工业质检等领域都有了广泛的应用，它不仅能够提高图像的质量，还能增强细节和提高视觉感知，大大提升了低分辨率图像的利用价值。本研究提出了一种基于生成对抗网络（SRGAN）的改进型图像超分辨率重建模型，在生成器中引入了通道和空间注意力机制（CSAB），使其能够有效地利用输入图像的信息增强特征表示并捕捉图像中的重要信息。判别器采用改进后的PatchGAN架构，能够更精确地捕捉图像的局部细节和纹理信息。通过改进的生成器和判别器架构，以及优化的损失函数设计，本研究方法在图像质量评估指标上表现较好。实验结果显示，该方法优于传统方法的性能，同时在视觉效果上也呈现出更为细腻和真实的图像细节。
Abstract：Image super-resolution reconstruction technology is currently widely used in medical imaging, video surveillance, and industrial quality inspection. It not only enhances image quality but also improves details and visual perception, significantly increasing the utility of low-resolution images. In this study, an improved image superresolution reconstruction model based on Generative Adversarial Networks(SRGAN) was proposed. This model introduced a channel and spatial attention mechanism(CSAB) in the generator, allowing it to effectively leverage the information from the input image to enhance feature representations and capture important details. The discriminator was designed with an improved PatchGAN architecture, which more accurately captured local details and texture information of the image. With these enhanced generator and discriminator architectures and an optimized loss function design, this method demonstrated superior performance in image quality assessment metrics. Experimental results showed that this model outperforms traditional methods, presenting more detailed and realistic image details in the visual effects.
[1]CHI J,SUN Z,WANG H,et al.CT Image SuperResolution Reconstruction Based on Global Hybrid Attention[J].Computers in Biology and Medicine,2022,150:106112.
[2]RYU K B,KANG S J,JEONG S I,et al.CN4SRSS:Combined Network for Super-Resolution Reconstruction and Semantic Segmentation in Frontal-Viewing Camera Images of Vehicle[J].Engineering Applications of Artificial Intelligence,2024,130:107673.
[3]ZHU H,LV Y,MENG J,et al.Vehicle Detection in Multisource Remote Sensing Images Based on EdgePreserving Super-Resolution Reconstruction[J].Remote Sensing,2023,15(17):4281.
[4]LEDIG C,THEIS L,HUSZAR F,et al.Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network[DB/OL].(2024-09-29)[2025-05-12].http://arxiv.org/abs/1609.04802.
[5]MEIJERING E,UNSER M.A Note on Cubic Convolution Interpolation[J].IEEE Transactions on Image Processing,2003,12(4):477-479.
[6]DONG C,LOY C C,HE K,et al.Image SuperResolution Using Deep Convolutional Networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,38(2):295-307.
[7]KIM J,LEE J K,LEE K M.Accurate Image SuperResolution Using Very Deep Convolutional Networks[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).2016:1646-1654.
[8]YAZICI Z A,?KSüZ I,EKENEL H K.GLIMS:Attention-Guided Lightweight Multi-scale Hybrid Network for Volumetric Semantic Segmentation[J].Image and Vision Computing,2024,146:105055.
[9]ISOLA P,ZHU J Y,ZHOU T,et al.Image-to-Image Translation with ConditionalAdversarial Networks[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).2017:5967-5976.
[10]RAKOTONIRINA N C,RASOANAIVO A.ESRGAN+:Further Improving Enhanced Super-Resolution Generative Adversarial Network[C]//Proceedings of2020 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP).2020:3637-3641.
[11]WANG X,YU K,WU S,et al.ESRGAN:Enhanced Super-Resolution Generative Adversarial Networks[C]//Computer Vision-ECCV 2018 Workshops.2019,11133:63-79.
[12]ZHANG K,LIANG J,VAN G L L,et al.Designing a Practical Degradation Model for Deep Blind Image Super-Resolution[C]//Proceedings of 2021 IEEE/CVF International Conference on Computer Vision(ICCV).2021:4771-4780.
基本信息:
DOI：10.19370/j.cnki.cn10-1886/ts.2025.05.003
中图分类号:TP391.41
引用信息:
[1]陆欣雅,陈佳怡,司占军,等.基于SRGAN的图像超分辨率重建模型（英文）[J].印刷与数字媒体技术研究,2025,No.238(05):21-28.DOI:10.19370/j.cnki.cn10-1886/ts.2025.05.003.
2025-10-10
2025-10-10