(49-6) 14 * << * >> * Russian * English * Content * All Issues

Adaptive compression algorithm based on JPEG2000 with neural network corrector of decoded image definition
S.V. Sai 1, V.S. Nikonov 1, E.S. Fomina 1

1 Pacific National University,
Tikhookeanskaya Str. 136, Khabarovsk, 680035, Russia

 PDF, 1713 kB

DOI: 10.18287/2412-6179-CO-1586

Pages: 986-993.

Full text of article: Russian language.

Abstract:
The article describes features of digital image processing in the process of adaptive compression based on the discrete wavelet transform (DWT) used in the JPEG2000 compression standard. Unlike the well-known compression algorithm, the new algorithm uses a reduction (scaling) of the sizes of the matrices of the DWT coefficients of the Y, U, and V signals at the first DWT iteration, after which a reduced copy of the low-frequency range image is used for further processing. Thus, at this stage the volume of video data is reduced by four times. To obtain the original resolution, in contrast to the well-known interpolation methods, the inverse DWT transform with added zero coefficients in the high-frequency subranges of the first iteration is used. The proposed algorithm allows compressing image files on average by 40...50 times with satisfactory quality. To restore high image quality, an original neural network sharpness corrector based on a convolutional model with training of sample image blocks by the brightness index in the Lab space is developed. Based on the ResNet architecture, a proprietary deep neural network model is developed, based on combining several techniques used to solve image reconstruction problems in other architectures. An optimal neural network training option is selected, allowing the trained model to be used to correct the clarity of decoded images to high objective quality indicators with subjective assessments of "good" and "excellent".

Keywords:
definition corrector, image analysis, distortion metric, discrete wavelet transform, compression efficiency, neural network.

Citation:
Sai SV, Nikonov VS, Fomina ES. Adaptive compression algorithm based on JPEG2000 with neural network corrector of decoded image definition. Computer Optics 2025; 49(6): 986-993. DOI: 10.18287/2412-6179-CO-1586.

Acknowledgements:
This work was funded by the Russian Science Foundation under grant No 24-11-20024 and the Ministry of Education and Science of Khabarovsk Krai under agreement No 124C/2024.

References:
  1. Svoboda P, Hradis M, Barina D, Zemcik P. Compression artifacts removal using convolutional neural networks. arXiv Preview. 2016. Source: <https://arxiv.org/abs/1605.00366>. DOI:      10.48550/arXiv.1605.00366.
  2. Sai SV. Method of adaptive quantization for the coefficients of a discrete wavelet transformation. Optoelectron Instrum Data Process 2023; 59(2): 167-176. DOI: 10.3103/S8756699023020115.
  3. Sai SV. Metric of fine structures distortions of compressed images. Computer Optics 2018; 42(5): 829-837. DOI: 10.18287/2412-6179-2018-42-5-829-837.
  4. Wang Z, Chen J, Hoi SCH. Deep learning for image super-resolution: A survey. arXiv Preview. 2020. Source: <https://arxiv.org/abs/1902.06068>. DOI: 10.48550/arXiv.1902.06068.
  5. Top 15 best image enlarger review. 2024. Source: <https://topten.ai/image-enlargers-review/>.
  6. Lai W-S, Huang J-B, Ahuja N, Yang M-H. Fast and accurate image super-resolution with deep Laplacian pyramid networks. IEEE Trans Pattern Anal Mach Intell 2019; 41(11): 2599-2613. DOI: 10.1109/TPAMI.2018.2865304.
  7. DIV2K dataset: DIVerse 2K resolution high quality images as used for the challenges @ NTIRE (CVPR 2017 and CVPR 2018) and @ PIRM (ECCV 2018). 2025. Source: <https://data.vision.ee.ethz.ch/cvl/DIV2K/>.
  8. Zhang K, Liang J, Van Gool L, Timofte R. Designing a practical degradation model for deep blind image super-resolution. arXiv Preview. 2021. Source: <https://arxiv.org/abs/2103.14006>. DOI: 10.48550/arXiv.2103.14006.
  9. Wang X, Xie L, Dong C, Shan Y. Real-Esrgan: Training real-world blind super-resolution with pure synthetic data. arXiv Preview. 2021. Source: <https://arxiv.org/abs/2107.10833>. DOI:           10.48550/arXiv.2107.10833.
  10. Liang J, Cao J, Sun G, Zhang K, Van Gool L, Timofte R. SwinIR: Image restoration using swin transformer. arXiv Preview. 2021. Source: <https://arxiv.org/abs/2108.10257>. DOI:    10.48550/arXiv.2108.10257.
  11. Tai Y, Yang J, Liu X, Xu C. MemNet: A persistent memory network for image restoration. 2017 IEEE Int Conf on Computer Vision (ICCV) 2017: 4549-4557. DOI: 10.1109/ICCV.2017.486.
  12. Zhang Y, Tian Y, Kong Y, Zhong B, Fu Y. Residual dense network for image super-resolution. arXiv Preview. 2021. Source: <https://arxiv.org/abs/1802.08797>. DOI: 10.48550/arXiv.1802.08797.
  13. Lim B, Son S, Kim H, Nah S, Lee K. Enhanced deep residual networks for single image super-resolution. arXiv Preview. 2017. Source: <https://arxiv.org/abs/1707.02921>. DOI:    10.48550/arXiv.1707.02921.
  14. He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. arXiv Preview. 2015. Source: <https://arxiv.org/abs/1512.03385>. DOI: 10.48550/arXiv.1512.03385.
  15. Li J, Fang F, Mei K, Zhang G. Multi-scale residual network for image super-resolution. In Book: Ferrari V, Hebert M, Sminchisescu C, Weiss Y, eds. Computer vision – ECCV 2018. 15th European Conference, Munich, Germany, September 8-14, 2018, Proceedings, Part VIII. Cham, Switzerland: Springer Nature Switzerland AG: 2018: 527-542. DOI: 10.1007/978-3-030-01237-3_32.
  16. Fairchild MD. Color appearance models. John Wiley and Sons Ltd; 2005. ISBN: 0-470-01216-1.
  17. Keras 3 API documentation. 2025. Source: <http://keras.io/>.
  18. CUDA Toolkit. 2025. Source: <http://developer.nvidia.com>.
  19. TensorFlow. 2025. Source: <http://tensorflow.org>.
  20. Bychkov IV, Ruzhnikov GM, Fedorov RK, Popova AK, Avramenko YV. Classification of Sentinel-2 satellite images of the Baikal Natural Territory. Computer Optics 2022; 46(1): 90-96. DOI: 10.18287/2412-6179-CO-1022.

© 2009, IPSI RAS
151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: journal@computeroptics.ru ; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20