Deep Edge-Aware Post-Processing for JPEG Enhancement: CNN-Based Artifact Reduction and Image Quality Restoration
DOI:
https://doi.org/10.3991/ijoe.v22i08.62380Keywords:
CNN-based edge-aware artifact reduction framework (CNN-AR), JPEG enhancement, Holistically Nested Edge Detection (HED), Enhanced Deep Super-Resolution (EDSR)Abstract
Joint photographic experts group (JPEG) is one of the most widely used image compression standards, but its lossy nature often introduces visible artifacts such as blocking, ringing, and blurring, particularly at lower quality factors. These degradations significantly reduce perceptual quality and affect downstream computer vision tasks. To address these limitations, in this study work a CNN-based edge-aware artifact reduction framework (CNN-AR) is proposed that integrates an enhanced deep super-resolution (EDSR) backbone with a holistically nested edge detection (HED) guided loss. This design enforces both pixel fidelity and edge consistency, enabling superior artifact suppression while preserving fine structural details. Extensive experiments conducted on benchmark datasets (LIVE1, Kodak, Set14, Classic5, and CLIC) across quality factors 10–40 demonstrate the effectiveness of the proposed approach. Compared to state-ofthe-art models including ARCNN, DnCNN, and DPW-SDNet, the proposed method consistently achieves higher perceptual scores. On average, CNN-AR improves PSNR by +0.38 dB, Structural Similarity Index (SSIM) by +0.012, MS-SSIM by +0.009, and PSNR-B by +0.41 dB across datasets, shows its ability to deliver both numerically superior and visually sharper reconstructions.
References
[1] G. K. Wallace, “The JPEG still picture compression standard,” Commun. ACM, vol. 34, no. 4, pp. 30–44, 1991.
[2] C. Dong, Y. Deng, C. C. Loy, and X. Tang, “Compression Artifacts Reduction by a Deep Convolutional Network,” presented at the Proceedings of the IEEE International Conference on Computer Vision, 2015, pp. 576–584. Accessed: May 19, 2026. [Online]. Available: https://openaccess.thecvf.com/content_iccv_2015/html/Dong_Compression_Artifacts_Reduction_ICCV_2015_paper.html
[3] K. Zhang, W. Zuo, Y. Chen, D. Meng, and L. Zhang, “Beyond a gaussian denoiser: Residual learning of deep cnn for image denoising,” IEEE Trans. Image Process., vol. 26, no. 7, pp. 3142–3155, 2017.
[4] C. Yim and A. C. Bovik, “Quality assessment of deblocked images,” IEEE Trans. Image Process., vol. 20, no. 1, pp. 88–98, 2010.
[5] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: from error visibility to structural similarity,” IEEE Trans. Image Process., vol. 13, no. 4, pp. 600–612, 2004.
[6] Z. Wang, E. P. Simoncelli, and A. C. Bovik, “Multiscale structural similarity for image quality assessment,” presented at the The thrity-seventh asilomar conference on signals, systems & computers, 2003, Ieee, 2003, pp. 1398–1402.
[7] C. Yim and A. C. Bovik, “Quality assessment of deblocked images,” IEEE Trans. Image Process., vol. 20, no. 1, pp. 88–98, 2010.
[8] S. Xie and Z. Tu, “Holistically-nested edge detection,” presented at the Proceedings of the IEEE international conference on computer vision, 2015, pp. 1395–1403.
[9] J. Kim, J. K. Lee, and K. M. Lee, “Accurate image super-resolution using very deep convolutional networks,” presented at the Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 1646–1654.
[10] B. Lim, S. Son, H. Kim, S. Nah, and K. Mu Lee, “Enhanced deep residual networks for single image super-resolution,” presented at the Proceedings of the IEEE conference on computer vision and pattern recognition workshops, 2017, pp. 136–144.
[11] “Laboratory for Image and Video Engineering - The University of Texas at Austin.” Accessed: May 19, 2026. [Online]. Available: https://live.ece.utexas.edu/research/quality/
[12] “True Color Kodak Images.” Accessed: May 19, 2026. [Online]. Available: https://r0k.us/graphics/kodak/
[13] R. Zeyde, M. Elad, and M. Protter, “On single image scale-up using sparse-representations,” presented at the International conference on curves and surfaces, Springer, 2010, pp. 711–730.
[14] “CLIC • Challenge on Learned Image Compression.” Accessed: May 19, 2026. [Online]. Available: https://clic2025.compression.cc/
[15] A. K. J. Saudagar, “Biomedical Image Compression Techniques for Clinical Image Processing,” Int. J. Online Biomed. Eng. IJOE, vol. 16, no. 12, pp. 133–154, Oct. 2020, doi: 10.3991/ijoe.v16i12.17019.
[16] J. Li, Y. Wang, H. Xie, and K.-K. Ma, “Learning a single model with a wide range of quality factors for JPEG image artifacts removal,” IEEE Trans. Image Process., vol. 29, pp. 8842–8854, 2020.
[17] J. He, X. He, M. Zhang, S. Xiong, and H. Chen, “Deep dual-domain semi-blind network for compressed image quality enhancement,” Knowl.-Based Syst., vol. 238, p. 107870, 2022.
[18] H. T. Sadeeq, T. H. Hameed, A. S. Abdi, and A. N. Abdulfatah, “Image Compression Using Neural Networks: A Review,” Int. J. Online Biomed. Eng. IJOE, vol. 17, no. 14, pp. 135–153, Dec. 2021, doi: 10.3991/ijoe.v17i14.26059.
[19] “Blind Quality Enhancement for Compressed Video | Request PDF.” Accessed: May 19, 2026. [Online]. Available: https://www.researchgate.net/publication/376256952_Blind_Quality_Enhancement_for_Compressed_Video
[20] C. Dong, C. C. Loy, K. He, and X. Tang, “Learning a deep convolutional network for image super-resolution,” presented at the European conference on computer vision, Springer, 2014, pp. 184–199.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Nupur, Nishant Kumar, Sajal Suhane, Rama Krishna Yellapragada, Ketan Anand
This work is licensed under a Creative Commons Attribution 4.0 International License.
