Article Details

Title · Authors · Abstract · Keywords · DOI · References

Image Super-Resolution Method Based on Deep Learning and Wavelet Transform

HUANG Yufeng1, ZHANG Xiaobo1,2

(1.Xi'an Shiyou University, Xi'an 710065, China; 2.Xianyang Normal University, Xianyang 712000, China)

Section：Computer Technology · Vol. 2025, Issue 6 · Pages：62-66 · 2096-4706（2025）06-0062-05 · CLC number：TP391 · Document code：A · DOI：10.19850/j.cnki.2096-4706.2025.06.012

Abstract

In recent years, single image Super-Resolution methods based on Deep Learning have achieved remarkable achievements. However, many methods only study the image spatial domain, ignoring the importance of high-frequency information in the image frequency domain, resulting in a relatively smooth image. Because wavelet transform can extract image details, this paper proposes a single image Super-Resolution method combining wavelet transform and Deep Learning. The method processes the image through dynamic convolution, uses Haar wavelet to decompose the image into low-frequency subgraphs and high-frequency subgraphs in three directions, and employs connected residual dense blocks and residual blocks to construct the inference network. Then, the local features are enhanced by inverse wavelet transform and residual dense blocks to obtain Super-Resolution images. The experimental results show that the method performs well in both subjective visual effects and objective evaluation indicators.

Keywords

Super-Resolution; wavelet transform; Deep Learning; residual dense block; residual block

References

  1. [1] Yang Wenxia Analysis of Medical Image Super Resolution Reconstruction Based on Deep Learning [J]. Information Technology and Informatization, 2023 (8): 176-179
  2. [2] Xie Meng, Yang Chunlei, Gu Mingjian, etc Deep learning based super-resolution reconstruction of Fengyun-4 satellite images [J]. Infrared, 2023, 44 (7): 46-52
  3. [3] Jiang Shuitao, Wan Wanggen, Wang Xuzhi, etc Facial image super-resolution reconstruction based on semantic perception and spatial attention [J]. Industrial Control Computer, 2023, 36 (10): 81-83
  4. [4] Zhang Siyan, Du Zhounan, Ren Yixin, etc A real-time super-resolution VLSI design using bicubic interpolation [J]. Journal of Southwest University: Natural Science Edition, 2024, 46 (4): 2022-212
  5. [5] Zhong Mengyuan, Jiang Lin Overview of super-resolution image reconstruction algorithms [J]. Computer Science and Exploration, 2022, 16 (5): 972-990
  6. [6] ZHOU S C，ZHANG J W，ZUO W M，et al. Cross-Scale Internal Graph Neural Network for Image Super-Resolution [J/OL].arXiv:2006.16673 [cs.CV].[2024-09-28].https://arxiv.org/abs/2006.16673.
  7. [7] NIU B，WEN W L，REN W Q，et al. Single Image Super-Resolution via A Holistic Attention Network [J/OL].arXiv:2008.08767[eess.IV].[2024-09-28].https://arxiv.org/abs/2008.08767.
  8. [8] MEI Y Q，FAN Y C，ZHOU Y Q. Image Super-Resolution with Non-Local Sparse Attention [C]//2021 IEEE/CVF Computer Vision and Pattern Recognition.Nashville：IEEE，2021：3516-3525.
  9. [9] GUO T T，MOUSAVI H S，VU T H，et al. Deep Wavelet Prediction for Image Super-Resolution [C]//2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops.Honolulu：IEEE，2017：1100-1109.
  10. [10] KIM J，LEE J K，LEE K M. Accurate Image Super-Resolution Using Very Deep Convolutional Networks [C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Las Vegas：IEEE，2016：1646-1654.
  11. [11] Sun Chao, Kou Kunhu, Lv Junwei, etc Research on Image Super Resolution Method Based on Wavelet Deep Network [J]. Computer Application Research, 2020, 37 (S1): 380-384
  12. [12] Duan Lijuan, Wu Chunli, Enqing, etc Deep residual network image super-resolution algorithm based on wavelet domain [J]. Journal of Software, 2019, 30 (4): 941-953
  13. [13] ZHANG Y L，TIAN Y P，KONG Y，et al. Residual Dense Network for Image Super-Resolution [C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.City：IEEE，2018：2472-2481.
  14. [14] HE K M，ZHANG X Y，REN S Q，et al. Deep Residual Learning for Image Recognition [C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas：IEEE，2016：770-778.
  15. [15] Wang Xianghai, Zhao Xiaoyang, Wang Xinying, etc Single image super-resolution reconstruction using non extracted wavelet edge learning deep residual network [J]. Acta Sinica, 2022, 50 (7): 1753-1765
  16. [16] Sun Chao, Lv Junwei, Gong Jian, etc Image super-resolution method combining wavelet transform and deep network [J]. Advances in Laser and Optoelectronics, 2018, 55 (12): 223-230
  17. [17] DONG C，LOY C C，HE K M，et al. Image Super-Resolution Using Deep Convolutional Networks [J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2016，38（2）：295-307.
  18. [18] JEEVAN P，SRINIDHI A，PRATHIBA P，et al. WaveMixSR: Resource-efficient Neural Network for Image Super-resolution [C]//2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV).Waikoloa：IEEE，2024：5872-5880.

About the authors

Huang Yufeng (1998-), female, Han ethnicity, from Nanchong, Sichuan, currently pursuing a master's degree in deep learning; Zhang Xiaobo (1979-), male, Han ethnicity, born in Xianyang, Shaanxi Province, professor, PhD, research interests: machine learning, big data analysis, image processing, natural language processing.