압축 왜곡 감소를 위한 CNN 기반 이미지 화질개선 알고리즘 = CNN based Image Restoration Method for the Reduction of Compression Artifacts

다국어 초록 (Multilingual Abstract)

As realistic media are widespread in various image processing areas, image or video compression is one of the key technologies to enable real-time applications with limited network bandwidth. Generally, image or video compression cause the unnecessary compression artifacts, such as blocking artifacts and ringing effects. In this study, we propose a Deep Residual Channel-attention Network, so called DRCAN, which consists of an input layer, a feature extractor and an output layer. Experimental results showed that the proposed DRCAN can reduced the total memory size and the inference time by as low as 47% and 59%, respectively. In addition, DRCAN can achieve a better peak signal-to-noise ratio and structural similarity index measure for compressed images compared to the previous methods.

참고문헌 (Reference)

1 곽내정 ; 신현준 ; 양종섭 ; 송특섭, "변형된 잔차블록을 적용한 CNN" 한국멀티미디어학회 23 (23): 803-811, 2020

2 Google, "Webp—A New Image Format for the Web"

3 R. Olaf, "U-Net : Convolutional Networks for Biomedical Image Segmentation"

4 G. Wallace, "The JPEG Still Picture Compression Standard" 38 (38): 18-34, 1992

5 J. Hu, "Squeeze-and-Excitation Networks" 1-13, 2018

6 S. Kim, "Single Image Super-Resolution Method Using CNN-Based Lightweight Neural Networks" 11 (11): 1092-, 2021

7 G. Sullivan, "Overview of the High Efficiency Video Coding(HEVC)Standard" 22 (22): 1649-1668, 2012

8 N. Ofir, "On Detection of Faint Edges in Noisy Images" 42 (42): 894-908, 2019

9 E. Agustsson, "NTIRE 2017 Challenge on Single Image Super-Resolution:Dataset and Study" 126-135, 2017

10 P. Liu, "Multi-Level Wavelet-CNN for Image Restoration" 886-895, 2018

11 X. Fu, "JPEG Artifacts Reduction via Deep Convolutional Sparse Coding" 2501-2510, 2019

12 W. Shi, "Image Super-Resolution with Global Correspondence Using Multi-Atlas PatchMatch" 9-16, 2013

13 J. Yang, "Image Super-Resolution Via Sparse Representation" 19 (19): 2861-2873, 2010

14 A. Hore, "Image Quality Metrics:PSNR vs. SSIM" 2366-2369, 2010

15 Z. Wang, "Image Quality Assessment : From Error Visibility to Structural Similarity" 13 (13): 600-612, 2004

16 T. Tung, "ICEBIM : Image Contrast Enhancement Based on Induced Norm and Local Patch Approaches" 9 : 23737-23750, 2021

17 박진호 ; 이응주, "Human Activity Recognition Based on 3D Residual Dense Network" 한국멀티미디어학회 23 (23): 1540-1551, 2020

18 K. Srinivas, "Exposure-Based Energy Curve Equalization for Enhancement of Contrast Distorted Images" 30 (30): 4663-4675, 2020

19 Y. Lee, "Enhanced Single Image Super Resolution Method Using Lightweight Multi-Scale Channel Dense Network" 21 (21): 3351-, 2021

20 G. Huang, "Densely Connected Convolutional Networks" 4700-4708, 2017

21 X. Glorot, "Deep Sparse Rectifier Neural Networks" 315-323, 2011

22 K. He, "Deep Residual Learning for Image Recognition" 770-778, 2016

23 C. Dong, "Compression Artifacts Reduction by a Deep Convolutional Network" 576-584, 2015

24 K. Zhang, "Beyond a Gaussian Denoiser : Residual Learning of Deep CNN for Image Denoising" 26 (26): 3142-3155, 2017

25 Y. Lecun, "Backpropagation Applied to handwritten Zip code Recognition" 1 (1): 541-551, 1989

26 J. He, "BDCN : Bi-Directional Cascade Network for Perceptual Edge Detection" 10 : 1-14, 2020

27 J. Wang, "An Improved Enhancement Algorithm Based on CNN Applicable for Weak Contrast Images" 8 : 8459-8476, 2020

28 K. H. Chen, "An Energy-Aware IP Core Design for the Variable-Length DCT/IDCT Targeting at MPEG4 Shape-Adaptive Transforms" 15 (15): 704-715, 2005

29 Y. Liu, "An Adaptive and Robust Edge Detection Method Based on Edge Proportion Statistics" 29 : 5206-5215, 2020