한국융합학회논문지 (Journal of the Korea Convergence Society)

  • 제11권3호
  • /
  • Pages.19-26
  • /
  • 2020
  • /
  • 2233-4890(pISSN)
  • /
  • 2713-6353(eISSN)
한국융합학회 (Korea Convergence Society)
권호 표지
DOI QR코드

DOI QR Code

손실 영역 분석 기반의 학습데이터 매핑 기법을 이용한 초해상도 연구

Super Resolution using Dictionary Data Mapping Method based on Loss Area Analysis

  • 한현호 (울산대학교 교양학부) ;
  • 이상훈 (광운대학교 인제니움학부)
  • Han, Hyun-Ho (College of General Education, University of Ulsan) ;
  • Lee, Sang-Hun (Ingenium College of Liberal Arts, Kwangwoon University)
  • 투고 : 2019.11.20
  • 심사 : 2020.03.20
  • 발행 : 2020.03.28
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 학습된 사전 기반 초해상도 결과를 개선하기 위해 분석한 손실 영역을 기반으로 학습 데이터를 적용하는 방법을 제안하였다. 기존의 학습된 사전 기반 방법은 입력 영상의 특징을 고려하지 않는 학습된 영상의 형태로 출력할 수 있으며, 이 과정에서 인공물이 발생할 수 있다. 제안하는 방법은 입력 영상과 학습된 영상의 일치하지 않는 특징으로 인한 인공물 발생을 줄이기 위해 1차 복원 결과를 분석함으로써 손실 정보를 추정하였다. 추정된 결과의 잡음 및 화소 불균형을 가우시안 기반의 커널로 개선하여 생성된 특징 맵에 따라 학습 데이터를 매핑하였다. 결과 비교를 위해 기존의 초해상도 방법과 제안 방법의 결과를 고화질 영상과 PSNR(Peak Signal to Noise Ratio), SSIM(Structural SIMilarity Index) 으로 비교한 결과 각각 4%와 3%의 향상된 결과를 확인하였다.

In this paper, we propose a method to analyze the loss region of the dictionary-based super resolution result learned for image quality improvement and to map the learning data according to the analyzed loss region. In the conventional learned dictionary-based method, a result different from the feature configuration of the input image may be generated according to the learning image, and an unintended artifact may occur. The proposed method estimate loss information of low resolution images by analyzing the reconstructed contents to reduce inconsistent feature composition and unintended artifacts in the example-based super resolution process. By mapping the training data according to the final interpolation feature map, which improves the noise and pixel imbalance of the estimated loss information using a Gaussian-based kernel, it generates super resolution with improved noise, artifacts, and staircase compared to the existing super resolution. For the evaluation, the results of the existing super resolution generation algorithms and the proposed method are compared with the high-definition image, which is 4% better in the PSNR (Peak Signal to Noise Ratio) and 3% in the SSIM (Structural SIMilarity Index).

키워드

참고문헌

  1. H. H. Han, J. Y. Lee, K. D. Jung & S. H. Lee. (2018). Patch Information based Linear Interpolation for Generating Super-Resolution Images in a Single Image. Journal of the Korea Convergence Society, 9(6), 45-52. DOI : 10.15207/JKCS.2018.9.6.045
  2. H. J. Kim, Y. S. Park, K. B. Kim & S. H. Lee. (2019). Modified HOG Feature Extraction for Pedestrian Tracking. Journal of the Korea Convergence Society, 10(3), 39-47. DOI : 10.15207/JKCS.2019.10.3.039
  3. D. I. Kim, G. S. Lee, K. H. Han & S. H. Lee. (2019). A Study on the Improvement of Skin Loss Area in Skin Color Extraction for Face Detection. Journal of the Korea Convergence Society, 10(5), 1-8. DOI : 10.15207/JKCS.2019.10.5.001
  4. G. O. Kim, G. S. Lee & S. H. Lee. (2014). An Edge Extraction Method Using K-Means Clustering In Image, Journal of Digital Convergence, 12(11), 281-288. DOI : 10.14400/JDC.2014.12.11.281
  5. T. H. Yoo, G. S. Lee & S. H. Lee. (2012). Window Production Method based on Low-Frequency Detection for Automatic Object Extraction of Grabcut, Journal of Digital Convergence, 10(8), 211-217. DOI : 10.14400/JDPM.2012.10.8.211
  6. E. Rhee. (2017). Security Algorithm for Vehicle Type Recognition. Journal of Convergence for Information Technology, 7(2), 77-82. DOI : 10.22156/CS4SMB.2017.7.2.077
  7. M. K. Kwon & H. S. Yang. (2017). A scene search method based on principal character identification using convolutional neural network. Journal of Convergence for Information Technology, 7(2), 31-36. DOI : 10.22156/CS4SMB.2017.7.2.031
  8. Y. Tai, J. Yang & X. Liu. (2017). Image super-resolution via deep recursive residual network. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 3147-3155). DOI : 10.1109/cvpr.2017.298
  9. R. W. Gerchberg. (1974). Super-resolution through error energy reduction. Optica Acta: International Journal of Optics, 21(9), 709-720. DOI : 10.1080/713818946
  10. L. C. Pickup, S. J. Roberts & A. Zisserman. (2004). A sampled texture prior for image super-resolution. In Advances in neural information processing systems (pp. 1587-1594). DOI : 10.1109/icip.2012.6467332
  11. A. Corduneanu & J. C. Platt. (2005, September). Learning spatially-variable filters for super-resolution of text. In IEEE International Conference on Image Processing 2005 (1, pp. I-849). IEEE. DOI : 10.1109/icip.2005.1529884
  12. W. Liu, D. Lin & X. Tang. (2005, July). Neighbor combination and transformation for hallucinating faces. In 2005 IEEE International Conference on Multimedia and Expo (pp. 4-pp). IEEE. DOI : 10.1109/icme.2005.1521381
  13. D. Datsenko & M. Elad. (2007). Example-based single document image super-resolution: a global MAP approach with outlier rejection. Multidimensional Systems and Signal Processing, 18(2-3), 103-121. DOI : 10.1007/s11045-007-0018-z
  14. M. Elad & D. Datsenko. (2008). Example-based regularization deployed to super-resolution reconstruction of a single image. The Computer Journal, 52(1), 15-30. DOI : 10.1093/comjnl/bxm008
  15. L. Li & Y. Wang. (2008, October). Face super-resolution using a hybrid model. In 2008 9th International Conference on Signal Processing (pp. 1153-1156). IEEE. DOI : 10.1109/icosp.2008.4697334
  16. J. Yang, Z. Wang, Z. Lin, S. Cohen & T. Huang. (2012). Coupled dictionary training for image super-resolution. IEEE transactions on image processing, 21(8), 3467-3478. DOI : 10.1109/tip.2012.2192127
  17. D. Glasner, S. Bagon & M. Irani. (2009, September). Super-resolution from a single image. In 2009 IEEE 12th international conference on computer vision (pp. 349-356). IEEE. DOI : 10.1109/iccv.2009.5459271
  18. 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/icip.2011.6115635
  19. D. Martin, C. Fowlkes, D. Tal & J. Malik. (2001, July). A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics. Vancouver:: Iccv. DOI : 10.1109/iccv.2001.937655