Pengembangan Model Super Resolution Lightweight Berbasis Hybrid untuk Lingkungan Komputasi Terbatas

Ade Soekarno Putra Santoso; Kahfi Heryandi Suradiradja; Choirul Basir

doi:10.55606/jcsr-politama.v4i4.6537

Authors

Ade Soekarno Putra Santoso Universitas Pamulang
Kahfi Heryandi Suradiradja Universitas Pamulang
Choirul Basir Universitas Pamulang

DOI:

https://doi.org/10.55606/jcsr-politama.v4i4.6537

Keywords:

Hybrid Method, Image Processing, Lightweight Model, Limited Computation, Super-Resolution

Abstract

Computational resource limitations, particularly the use of CPU-based systems without GPU acceleration, require super-resolution models that can achieve a balance between reconstruction quality and computational efficiency. This study proposes a lightweight hybrid super-resolution model designed for resource-constrained computing environments. The proposed architecture integrates lightweight feature extraction, residual learning, and an efficient image reconstruction module to enhance image resolution while maintaining low computational complexity. Model performance is evaluated using Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), computation time, CPU utilization, and memory consumption under different scaling factors. Experimental results show that the proposed model achieves competitive reconstruction performance while maintaining computational efficiency compared with several lightweight benchmark models. These findings indicate that the proposed hybrid architecture is suitable for deployment in resource-limited computing environments. Future work may focus on evaluating and optimizing the model on edge computing platforms, including smartphones and embedded systems, to validate its performance under real-world operating conditions.

Downloads

Download data is not yet available.

References

Ahn, N., Kang, B., & Sohn, K. A. (2018). Fast, accurate, and lightweight super-resolution with cascading residual network. Lecture Notes in Computer Science, 11133, 252-268.

Cai, J., Gu, S., Timofte, R., & Zhang, L. (2019). NTIRE 2019 challenge on real image super-resolution: Methods and results. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 0-0.

Chen, J., Wang, L., & Huang, H. (2023). Lightweight hybrid super-resolution network for resource-constrained devices. Signal Processing: Image Communication, 117, 116987.

Guo, X., Zhang, Y., Li, J., & Wang, H. (2024). Lightweight image super-resolution via adaptive feature distillation. Knowledge-Based Systems, 299, 112145.

Li, Y., Zhao, H., & Chen, X. (2024). Efficient lightweight super-resolution using hybrid feature fusion. Expert Systems with Applications, 255, 124657.

Liu, Z., Wang, Q., & Sun, Y. (2023). Hybrid attention-based lightweight image super-resolution network. Neural Computing and Applications, 35(18), 13241-13258.

Lu, Y., Zhang, H., & Wang, S. (2024). Hybrid attention super-resolution network for lightweight image reconstruction. Neural Computing and Applications, 36(5), 7851-7868.

Ma, W., Li, X., & Zhao, Y. (2024). Deep residual feature aggregation network for lightweight image super-resolution. Knowledge-Based Systems, 302, 112487.

Pan, J., Yang, X., & Li, H. (2025). Efficient hybrid CNN architecture for image super-resolution. Expert Systems with Applications, 266, 125942.

Shi, W., Caballero, J., Huszár, F., Aitken, A., Bishop, R., Rueckert, D., & Wang, Z. (2016). Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1874-1883.

Sun, L., Zhao, X., & Li, H. (2023). Hybrid attention super-resolution network for lightweight image reconstruction. Neurocomputing, 548, 126560.

Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P. (2004). Image quality assessment: From error visibility to structural similarity. Image Processing, 13(4), 600-612.

Yin, P., Zhang, Y., & Liu, Z. (2024). Lightweight feature mixing network for image super-resolution. Information Sciences, 682, 121164.

Zhang, K., Zuo, W., & Zhang, L. (2021). Deep learning for image super-resolution: A survey. International Journal of Computer Vision, 129(10), 2953-2976.

Zhang, Y., Li, K., Li, K., Wang, L., Zhong, B., & Fu, Y. (2018). Image super-resolution using very deep residual channel attention networks. Lecture Notes in Computer Science, 11211, 294-310.

Zhao, X., Chen, Y., & Wang, H. (2025). Adaptive lightweight super-resolution for mobile vision applications. Pattern Recognition, 161, 111221.

Zhou, M., Li, C., & Huang, Y. (2024). Efficient image super-resolution using lightweight residual learning. Multimedia Tools and Applications, 83(17), 50371-50390.

Zhu, L., Chen, H., & Xu, Y. (2024). Efficient residual feature learning for lightweight image super-resolution. Multimedia Tools and Applications, 83(14), 42115-42134.