A HYBRID VAE-HUFFMAN GENERATIVE COMPRESSION FRAMEWORK FOR IMAGE TRANSMISSION IN WIRELESS SENSOR NETWORKS
DOI:
https://doi.org/10.55766/sujst12181Keywords:
Variable autoencoder, Huffman coding, Generative compression, Wireless sensor networksAbstract
This paper proposes a hybrid image compression framework that integrates a tunable β-Variational Autoencoder (β-VAE) with Huffman entropy coding to achieve efficient, low-complexity compression suitable for wireless sensor networks (WSNs) and resource-constrained embedded systems. The model jointly optimizes latent representations and entropy-aware symbol distributions, improving both rate-distortion performance and theoretical coding efficiency. Trained on CIFAR-100 and ImageNet and evaluated on the Kodak dataset, the approach attains over 32 dB PSNR and SSIM above 0.90, and an average compression ratio of 18.22:1. When compared with the JPEG standard, the proposed method provides comparable perceptual quality while enabling deterministic, lightweight decoding that substantially reduces computational and energy demands. The results demonstrate that combining generative latent representation modeling with classical entropy coding offers a scalable, hardware-friendly alternative for next-generation learned image compression systems.
References
Badr, M. S., Elrewainy, A. F., & Elshafey, M. A. (2024). Efficient lossy satellite image compression using hybrid autoencoder model. In Proceedings of the 2024 14th International Conference on Electrical Engineering (ICEENG) (pp. 78-83). IEEE. https://doi.org/10.1109/ICEENG58856.2024.10566288
Ballé, J., Minnen, D., Singh, S., Hwang, S. J., & Johnston, N. (2018). Variational image compression with a scale hyperprior. In Proceedings of the 6th International Conference on Learning Representations (ICLR). https://doi.org/10.48550/arXiv.1802.01436
Cervantes-Sanchez, F., Thoma, A., & Rahman, S. (2024). AI-ready image compression for whole-slide images using bottleneck tensor compression. In Proceedings of the Data Compression Conference (DCC) (pp. 150-159). IEEE. https://doi.org/10.1109/DCC58796.2024.00066
Chen, J., Fang, Y., Khisti, A., Özgür, A., & Shlezinger, N. (2025). Information compression in the AI era: Recent advances and future challenges. IEEE Journal on Selected Areas in Communications, 43(7), 2333-2348. https://doi.org/10.1109/JSAC.2025.3560359
Chitra, P. (2024). 4D image compression in healthcare using convolutional neural networks (CNNs). In Proceedings of the International Conference on Computing and Data Analytics (ICCDA) (pp. 1-7). IEEE. https://doi.org/10.1109/ICCDA64887.2024.10867362
Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press. https://www.deeplearningbook.org/
Kingma, D. P., & Welling, M. (2013). Auto-encoding variational Bayes (arXiv:1312.6114). arXiv. https://doi.org/10.48550/arXiv.1312.6114
Lungisani, B. A., Lebekwe, C. K., Zungeru, A. M., & Yahya, A. (2022). Image compression techniques in wireless sensor networks: A survey and comparison. IEEE Access, 10, 82511-82530. https://doi.org/10.1109/ACCESS.2022.3195891
Muckley, M., Havasi, M., Verbeek, J., & Ullrich, K. (2025). Architecture optimizations for improving neural image compression compute complexity. In Proceedings of the Data Compression Conference (DCC) (pp. 3-12). IEEE. https://doi.org/10.1109/DCC62719.2025.00008
Pathak, T. (2024). Natural images with synthetic noise [Data set]. Kaggle. https://doi.org/10.34740/kaggle/dsv/1958494
Sapkota, R., & Karkee, M. (2025). Ultralytics YOLO evolution: An overview of YOLO26, YOLO11, YOLOv8 and YOLOv5 object detectors for computer vision and pattern recognition (arXiv:2510.09653). arXiv. https://doi.org/10.48550/arXiv.2510.09653
Sebai, D., Sehli, M., & Ghorbel, F. (2023). Learning end-to-end depth maps compression with conditional quality-controllable autoencoder. In Proceedings of the 2023 IEEE International Conference on Visual Communications and Image Processing (VCIP) (pp. 1-5). IEEE. https://doi.org/10.1109/VCIP59821.2023.10402610
Shahila, D. F. F., Stephen, V., Rama, R. S., Prathaban, B. P., Rubia, J., & Reddy, S. R. (2024). Deep computational based blocking effect cancellation for digital image compression. In Proceedings of the 2024 International Conference on Power, Energy, Control and Transmission Systems (ICPECTS) (pp. 1-6). IEEE. https://doi.org/10.1109/ICPECTS62210.2024.10780371
Sullivan, G. J., Ohm, J.-R., Han, W.-J., & Wiegand, T. (2012). Overview of the High Efficiency Video Coding (HEVC) standard. IEEE Transactions on Circuits and Systems for Video Technology, 22(12), 1649-1668. https://doi.org/10.1109/TCSVT.2012.2221191
Tyagi, A., Aditya, H., Khandelwal, R., Singh, J., Pal, Y., & Jaiswal. (2023). Autoencoder image compression for high compression and acceptable quality. In Proceedings of the International Conference on Information Technology (OCIT) (pp. 184-189). IEEE. https://doi.org/10.1109/OCIT59427.2023.10430629
Yu, B. (2024). Digital case image compression algorithm based on deep learning. In Proceedings of the International Conference on Intelligent Algorithms for Computational Intelligence Systems (IACIS) (pp. 1-7). IEEE. https://doi.org/10.1109/IACIS61494.2024.10721668
Zhang, Q., Sleder, S., Hu, X., Bilal, F., Ye, W., & Zhang, Z.-L. (2024). Impact of data compression on downstream AI tasks: A study using teleoperated driving over 5G. In Proceedings of the IEEE International Workshop Technical Committee on Communications Quality and Reliability (CQR) (pp. 25-30). IEEE. https://doi.org/10.1109/CQR62340.2024.10705883
Downloads
Published
How to Cite
Issue
Section
Categories
License
Copyright (c) 2026 Prayoth Kumsawat, Nitchamon Sawangsawai, Kitti Attakitmongcol, Arthit Srikaew
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
