High-Speed UAV Image Deblurring Optimization via Optical Flow Estimation Combining Deformable Convolution
PDF

Keywords

UAV image deblurring
optical flow estimation
deformable convolution
motion blur removal
aerial image restoration
deep learning
high-speed UAV imaging
computer vision

Abstract

Unmanned Aerial Vehicles (UAVs) are increasingly  utilized for surveillance, disaster management, and topographic  mapping; however, the utility of the visual data acquired is  frequently compromised by severe motion blur. This degradation  arises from the complex interplay between high-speed flight  maneuvers, engine-induced vibrations, and turbulent environmental  conditions, resulting in non-uniform and spatially variant blur  kernels. Traditional deblurring methods often fail to account for the  dynamic geometric distortions inherent in aerial imagery. This  paper proposes a novel deep learning framework that integrates  optical flow estimation with deformable convolution to address  high-speed UAV image deblurring. By leveraging optical flow to  capture explicit motion information and deformable convolutions to  adaptively adjust the receptive field based on geometric  transformations, the proposed model effectively aligns features and  restores sharp details. We present a comprehensive theoretical foundation for coupling motion estimation with adaptive sampling,followed by a rigorous experimental evaluation on synthetic and  real-world aerial datasets. The results demonstrate that our  approach significantly outperforms existing state-of-the-art  methods in terms of peak signal-to-noise ratio and structural  similarity, proving its efficacy in recovering high-frequency textures  essential for aerial analysis. 

PDF

References

Ma, F., Liu, L., & Cheng, H. V. (2024). TIMA: Text-Image Mutual Awareness for Balancing Zero-Shot Adversarial Robustness and Generalization Ability. arXiv preprint arXiv:2405.17678.

Wang, W., Mi, L., Cen, S., Dai, H., Li, Y., Fu, X., & Liu, Y. (2025). Region-based Content Enhancement for {Efficient} Video Analytics at the Edge. In 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25) (pp. 613-633).

Chen, J., Zhang, K., Zeng, H., Yan, J., Dai, J., & Dai, Z. (2024). Adaptive constraint relaxation-based evolutionary algorithm for constrained multi-objective optimization. Mathematics, 12(19), 3075.

Ma, Y., Qu, D., & Pyrozhenko, M. (2026). Bio-RegNet: A Meta-Homeostatic Bayesian Neural Network Framework Integrating Treg-Inspired Immunoregulation and Autophagic Optimization for Adaptive Community Detection and Stable Intelligence. Biomimetics, 11(1), 48.

Ke, Z., Cao, Y., Chen, Z., Yin, Y., He, S., & Cheng, Y. (2025). Early warning of cryptocurrency reversal risks via multi-source data. Finance Research Letters, 107890.

Zou, Y., & Yin, Z. (2025). MVCM: Enhancing Multi-View and Cross-Modality Alignment for Medical Visual Question Answering and Medical Image-Text Retrieval. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 180-190).

Wang, X., Wang, H., Tian, Z., Wang, W., & Chen, J. (2025). Angle-based dual-association evolutionary algorithm for many-objective optimization. Mathematics, 13(11), 1757.

Ke, Z., Shen, J., Zhao, X., Fu, X., Wang, Y., Li, Z., ... & Mu, H. (2025). A stable technical feature with GRU-CNN-GA fusion. Applied Soft Computing, 114302.

Wang, Y., Zhang, R., & Liu, J. (2023). RLS-DTS: Reinforcement-learning linguistic steganalysis in distribution-transformed scenario. IEEE Signal Processing Letters, 30, 1232-1236.

Ma, F., & Li, H. (2021, July). Underexposed image enhancement via unsupervised feature attention network. In 2021 IEEE International Conference on Multimedia and Expo (ICME) (pp. 1-6). IEEE.

Lu, P., Zhang, Y., Zhang, H., Zheng, J., Tong, K., & Wu, W. (2025, November). Tool-Augmented Hybrid Ensemble Reasoning with Distillation for Bilingual Mathematical Problem Solving. In 2025 4th International Conference on Image Processing, Computer Vision and Machine Learning (ICICML) (pp. 1770-1776). IEEE.

Liu, F., Wang, J., Tian, J., Zhuang, D., Miranda-Moreno, L., & Sun, L. (2022). A universal framework of spatiotemporal bias block for long-term traffic forecasting. IEEE Transactions on Intelligent Transportation Systems, 23(10), 19064-19075.

Wang, R., Guo, T., Li, Y., Meng, D., & Liang, B. (2025). Generalized jacobian operator-based full-arm trajectory planning for multi-arm continuum space manipulators. Aerospace Science and Technology, 111559.

Zhu, D., Xie, C., Wang, Z., & Zhang, H. (2025). RaX-Crash: A Resource Efficient and Explainable Small Model Pipeline with an Application to City Scale Injury Severity Prediction. arXiv preprint arXiv:2512.07848.

Liang, Z., Wei, W., Zhang, K., & Chen, H. (2025). Research on multi-hop inference optimization of llm based on mquake framework. arXiv preprint arXiv:2509.04770.

Choi, J., Choi, H., Hwang, J., & Park, N. (2022, June). Graph neural controlled differential equations for traffic forecasting. In Proceedings of the AAAI conference on artificial intelligence (Vol. 36, No. 6, pp. 6367-6374).

Huang, Y., Zhang, K., Wang, Y., Du, D., Yuan, Y., & Zhao, Z. (2025, June). Enhancing Open-Vocabulary Panoptic Segmentation with Semantic-Guided Q-Tuning. In 2025 IEEE International Conference on Multimedia and Expo (ICME) (pp. 1-6). IEEE.

Li, Z., Zhang, Y., Pan, T., Sun, Y., Duan, Z., Fang, J., ... & Wang, J. (2025, July). FocusLLM: Precise understanding of long context by dynamic condensing. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 31087-31101).

Lyu, M. R., Ray, B., Roychoudhury, A., Tan, S. H., & Thongtanunam, P. (2025). Automatic programming: Large language models and beyond. ACM Transactions on Software Engineering and Methodology, 34(5), 1-33.

Norris, J., He, Z., Qu, Y., Chen, G., Hertzog, C., & Jin, D. (2025, September). An in-network approach for pmu missing data recovery with data plane programmability. In 2025 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm) (pp. 1-7). IEEE.

Hu, Q., Peng, Y., Shao, Z., & Chen, J. (2026). Scene degradation-aware fusion network for robust infrared and visible image synthesis in extreme conditions. The Visual Computer, 42(1), 48.

Ma, F., Chai, J., & Wang, H. (2019). Two-dimensional compact variational mode decomposition-based low-light image enhancement. IEEE Access, 7, 136299-136309.

Hu, Q., Peng, Y., KinTak, U., & Chen, J. (2025). MSFusion: A Degradation-Correctable Framework for Robust Infrared and Visible Image Fusion. IEEE Sensors Journal, 26(2), 2749-2766.

Linot, A. J., Burby, J. W., Tang, Q., Balaprakash, P., Graham, M. D., & Maulik, R. (2023). Stabilized neural ordinary differential equations for long-time forecasting of dynamical systems. Journal of Computational Physics, 474, 111838.

Zhang, H., Zhao, S., Zhou, Z., Zhang, W., & Meng, Y. (2025, September). Domain-Specific RAG with Semantic Normalization and Contrastive Feedback for Document Question Answering. In 2025 7th International Conference on Internet of Things, Automation and Artificial Intelligence (IoTAAI) (pp. 750-753). IEEE.

Zeng, D., Yang, Y., Tang, Y., Zhao, L., Wang, X., Yun, D., ... & Lin, H. (2025). Shaping school for childhood myopia: the association between floor area ratio of school environment and myopia in China. British Journal of Ophthalmology, 109(1), 146-151.

Zhao, S., Shao, Z., Chen, Y., Zheng, L., & Chen, J. (2025). A self-organizing decomposition based evolutionary algorithm with cooperative diversity measure for many-objective optimization. AIMS Mathematics, 10(6), 13880-13907.

Chen, J., Yin, H., Zhang, K., Ren, Y., & Zeng, H. (2025). Integration of neural networks in brain–computer interface applications: Research frontiers and trend analysis based on Python. Engineering Applications of Artificial Intelligence, 151, 110654.

Wu, J., Sun, Y., Xie, T., Chen, S., Bao, J., Xu, Y., ... & Wang, X. (2026). Cross-Modal Memory Compression for Efficient Multi-Agent Debate. arXiv preprint arXiv:2602.00454.

Ouyang, K., Ke, Z., Fu, S., Liu, L., Zhao, P., & Hu, D. (2024). Learn from global correlations: Enhancing evolutionary algorithm via spectral gnn. arXiv preprint arXiv:2412.17629.

Hayes, C. F., Rădulescu, R., Bargiacchi, E., Källström, J., Macfarlane, M., Reymond, M., ... & Roijers, D. M. (2022). A practical guide to multi-objective reinforcement learning and planning: CF Hayes et al. Autonomous Agents and Multi-Agent Systems, 36(1), 26.

Chen, J., Wang, Y., Shao, Z., Zeng, H., & Zhao, S. (2025). Dual-population cooperative correlation evolutionary algorithm for constrained multi-objective optimization. Mathematics, 13(9), 1441.

Kovachki, N., Li, Z., Liu, B., Azizzadenesheli, K., Bhattacharya, K., Stuart, A., & Anandkumar, A. (2023). Neural operator: Learning maps between function spaces with applications to pdes. Journal of Machine Learning Research, 24(89), 1-97.

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright (c) 2026 Liam O'Connor, Charlotte Dubois (Author)

Downloads

Download data is not yet available.