디지털정책학회지 (Journal of Digital Policy)

한국디지털정책학회 (The Society of Digital Policy and Management)
권호 표지
DOI QR코드

DOI QR Code

적응형 가중치 잔차 블록을 적용한 다중 블록 구조 기반의 단일 영상 초해상도 기법

Single Image Super Resolution using Multi Grouped Block with Adaptive Weighted Residual Blocks

  • Hyun Ho Han (College of General Education, University of Ulsan)
  • 투고 : 2024.08.14
  • 심사 : 2024.09.20
  • 발행 : 2024.09.28
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문은 단일 영상 기반의 초해상도에서 결과의 품질을 개선하기 위해 적응형 가중치를 적용한 잔차 블록으로 구성된 다중 블록 구조를 이용하는 방법을 제안하였다. 딥러닝을 이용한 초해상도를 생성하는 과정에서 품질 향상을 위한 가장 중요한 요소는 특징 추출 및 적용이다. 해상도가 낮아 이미 손실된 세부사항을 복원하기 위해 다양한 특징을 추출하는 것이 최우선이지만 네트워크의 구조가 깊어지거나 복잡해지는 등의 문제가 발생하기 때문에 실제 적용에서 제한사항이 있다. 따라서 특징 추출 과정은 효율적으로 구성하고 적용 과정을 개선하여 품질을 개선하였다. 이를 위해 최초 특징 추출 이후 다중 블록 구조를 구성하였고 블록 내부에는 중첩된 잔차 블록을 구성한 뒤 적응형 가중치를 적용하였다. 또한 최종 고해상도 복원을 위해 다중 커널을 이용한 영상 재구성 과정을 적용함으로써 결과물의 품질을 향상시켰다. 평가를 위해 원본 영상 대비 PSNR과 SSIM 값을 구하였고 기존 알고리즘과 비교하여 제안하는 방법의 성능 향상을 확인하였다.

In this paper, proposes a method using a multi block structure composed of residual blocks with adaptive weights to improve the quality of results in single image super resolution. In the process of generating super resolution images using deep learning, the most critical factor for enhancing quality is feature extraction and application. While extracting various features is essential for restoring fine details that have been lost due to low resolution, issues such as increased network depth and complexity pose challenges in practical implementation. Therefore, the feature extraction process was structured efficiently, and the application process was improved to enhance quality. To achieve this, a multi block structure was designed after the initial feature extraction, with nested residual blocks inside each block, where adaptive weights were applied. Additionally, for final high resolution reconstruction, a multi kernel image reconstruction process was employed, further improving the quality of the results. The performance of the proposed method was evaluated by calculating PSNR and SSIM values compared to the original image, and its superiority was demonstrated through comparisons with existing algorithms.

키워드

참고문헌

  1. Y. Zhang, Y. Huang, K. Wang, G. Qi & J. Zhu. (2023). Single image super-resolution reconstruction with preservation of structure and texture details. Mathematics, 11(1), 216. DOI : 10.3390/math11010216 
  2. H. Chen et al. (2022). Real-world single image super-resolution: A brief review. Information Fusion, 79, 124-145. DOI : 10.1016/j.inffus.2021.09.005 
  3. H. Al-Mekhlafi & S. Liu. (2024). Single image super-resolution: a comprehensive review and recent insight. Frontiers of Computer Science, 18(1), 181702. DOI : 10.1007/s11704-023-2588-9 
  4. K. P. Gunasekaran. (2023, May). Ultra sharp: Study of single image super resolution using residual dense network. In 2023 IEEE 3rd International Conference on Computer Communication and Artificial Intelligence (CCAI) (pp. 261-266). IEEE. DOI : 10.1109/CCAI57533.2023.10201303 
  5. J. B. Huang, A., Singh & N. Ahuja. (2015). Single image super-resolution from transformed self-exemplars. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 5197-5206). DOI : 10.1109/CVPR.2015.7299156 
  6. J. Yang, J. Wright, T. S. Huang & Y. Ma. (2010). Image super-resolution via sparse representation. IEEE transactions on image processing, 19(11), 2861-2873. DOI : 10.1109/TIP.2010.2050625 
  7. B. Liu & D. Ait-Boudaoud. (2020). Effective image super resolution via hierarchical convolutional neural network. Neurocomputing, 374, 109-116. DOI : 10.1016/j.neucom.2019.09.035 
  8. S. M. A. Bashir, Y. Wang, M. Khan & Y. Niu. (2021). A comprehensive review of deep learning-based single image super-resolution. PeerJ Computer Science, 7, e621. DOI : 10.7717/peerj-cs.621/retraction 
  9. C. Dong, C. C. Loy, K. He & X. Tang. (2015). Image super-resolution using deep convolutional networks. IEEE transactions on pattern analysis and machine intelligence, 38(2), 295-307. DOI : 10.1109/TPAMI.2015.2439281 
  10. J. Kim, J. K. Lee & K. M. Lee. (2016). Accurate image super-resolution using very deep convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1646-1654). DOI : 10.1109/CVPR.2016.182 
  11. B. Lim, S. Son, H. Kim, S. Nah & K. Mu Lee. (2017). Enhanced deep residual networks for single image super-resolution. In Proceedings of the IEEE conference on computer vision and pattern recognition workshops (pp. 136-144). DOI : 10.48550/arXiv.1707.02921 
  12. Y. Zhang, Y. Tian, Y. Kong, B. Zhong & Y. Fu. (2018). Residual dense network for image super-resolution. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 2472-2481). DOI : 10.48550/arXiv.1802.08797 
  13. Z. Hui, X. Wang & X. Gao. (2018). Fast and accurate single image super-resolution via information distillation network. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 723-731). DOI : 10.48550/arXiv.1803.09454 
  14. G. Wu, J. Jiang & X. Liu. (2023). A practical contrastive learning framework for single-image super-resolution. IEEE Transactions on Neural Networks and Learning Systems. DOI : 10.1109/TNNLS.2023.3290038 
  15. K. S. Reddy, V. P. Vijayan, A. D. Gupta, P. Singh, R. G. Vidhya & D. Kapila. (2022, April). Implementation of super resolution in images based on generative Adversarial network. In 2022 8th International Conference on Smart Structures and Systems (ICSSS) (pp. 01-07). IEEE. DOI: 10.1109/ICSSS54381.2022.9782170