Journal of Digital Policy (디지털정책학회지)

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

DOI QR Code

Single Image Super Resolution using sub-Edge Extraction based on Hierarchical Structure

계층적 보조 경계 추출을 이용한 단일 영상의 초해상도 기법

  • Hyun Ho, Han (College of General Education, University of Ulsan)
  • Received : 2022.11.18
  • Accepted : 2022.12.20
  • Published : 2022.12.28
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we proposed a method using sub-edge information extracted through a hierarchical structure in the process of generating super resolution based on a single image. In order to improve the quality of super resolution, it is necessary to clearly distinguish the shape of each area while clearly expressing the boundary area in the image. The proposed method assists edge information of the image in deep learning based super resolution method to create an improved super resolution result while maintaining the structural shape of the boundary region, which is an important factor determining the quality in the super resolution process. In addition to the group convolution structure for performing deep learning based super resolution, a separate hierarchical edge accumulation extraction process based on high-frequency band information for sub-edge extraction is proposed, and a method of using it as an auxiliary feature is proposed. Experimental results showed about 1% performance improvement in PSNR and SSIM compared to the existing super resolution.

본 논문에서는 단일 영상을 기반으로 초해상도를 생성하는 과정에서 계층 구조를 거쳐 추출된 보조 경계 특징을 이용한 방법을 제안하였다. 초해상도의 품질을 향상시키기 위해서는 영상 내 경계 영역을 선명하게 표현하면서도 각 영역의 형태를 명확하게 구분하여야 한다. 제안하는 방법은 초해상도 과정에서 품질을 결정하는 중요한 요인인 경계 영역을 입력 영상의 구조적 형태를 유지하면서 개선된 초해상도 결과를 생성하기 위해 딥러닝 기반의 초해상도 방법에서 영상의 경계 영역 정보를 보조적으로 활용하는 구조를 사용하였다. 딥러닝 기반의 초해상도를 수행하기 위한 그룹 컨볼루션 구조에 더해 보조 경계 추출을 위한 고주파 대역의 정보를 기반으로 별도의 계층적 구조의 경계 누적 추출 과정을 수행하여 이를 보조 특징으로써 활용하는 방법을 제안하였다. 실험 결과 기존 초해상도 대비 PSNR과 SSIM에서 약 1%의 성능 향상을 보였다.

Keywords

References

  1. Y. Dun, Z. Da, S. Yang, Y. Xue & X. Qian. (2021). Kernel-attended residual network for single image super-resolution. Knowledge-Based Systems, 213, 106663. DOI : 10.3390/electronics10050555
  2. F. Wang, H. Hu & C. Shen. (2021). BAM: a lightweight and efficient balanced attention mechanism for single image super resolution. arXiv preprint arXiv:2104.07566. DOI : 10.48550/arXiv.2104.07566
  3. 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 : 10.1016/j.neucom.2019.09.035
  4. 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
  5. M. S. Sajjadi, B. Scholkopf & M. Hirsch. (2017). Enhancenet: Single image super-resolution through automated texture synthesis. In Proceedings of the IEEE international conference on computer vision (pp. 4491-4500). DOI : 10.48550/arXiv.1612.07919
  6. Y. Zhang, Q. Fan, F. Bao, Y. Liu & C. Zhang. (2018). Single-image super-resolution based on rational fractal interpolation. IEEE Transactions on Image Processing, 27(8), 3782-3797. DOI : 10.1109/TIP.2018.2826139
  7. 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
  8. 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
  9. H. Liu, Z. Fu, J. Han, L. Shao, S. Hou & Y. Chu. (2019). Single image super-resolution using multi-scale deep encoder-decoder with phase congruency edge map guidance. Information Sciences, 473, 44-58. DOI : 10.1016/j.ins.2018.09.018
  10. X. Zhang, H. Zeng & L. Zhang, (2021, October). Edge-oriented convolution block for real-time super resolution on mobile devices. In Proceedings of the 29th ACM International Conference on Multimedia (pp. 4034-4043). DOI : 10.1145/3474085.3475291
  11. G. Suryanarayana et al. (2021). Accurate magnetic resonance image super-resolution using deep networks and Gaussian filtering in the stationary wavelet domain. IEEE Access, 9, 71406-71417. DOI : 10.1109/ACCESS.2021.3077611