Altitude Prediction Over Water Surface Using CNN Trained With Synthetic Images

• João Vitor da Silva Neto UFPB
• Tiago Pereira do Nascimento UFPB
• Leonardo Vidal Batista UFPB

DOI: https://doi.org/10.5753/wvc.2024.34009

Resumo

The need for redundant odometry in robotic systems benefits from the recent possibilities of fast and efficient real-time image processing using classifications and regression models. The present work proposes an approach using Convolutional Neural Networks (CNN) to detect the distance between the point of capture of an image and a water surface below the capture. Our CNN was trained initially with synthetic images generated by Blender, and evaluated by real images at heights between 50 and 100 cm. Preliminary tests demonstrated an ability of the system to recognize the changing of height in a continuous shot, varying only the vertical position, even with some noise from elements not present in the training test database, such as dirt from leaves and stones at the bottom of the water.

Palavras-chave: UAV, water, machine learning, height estimation, synthetic images, odometry

Referências

H. Shakhatreh, A. H. Sawalmeh, A. Al-Fuqaha, Z. Dou, E. Almaita, I. Khalil, N. S. Othman, A. Khreishah, and M. Guizani, “Unmanned aerial vehicles (uavs): A survey on civil applications and key research challenges,” IEEE Access, vol. 7, pp. 48 572–48 634, 2019.

X.-Z. Cui, Q. Feng, S.-Z. Wang, and J.-H. Zhang, “Monocular depth estimation with self-supervised learning for vineyard unmanned agricultural vehicle,” Sensors, vol. 22, no. 3, 2022. [Online]. Available: [link]

A. Joglekar, D. Joshi, R. Khemani, S. Nair, and S. Sahare, “Depth estimation using monocular camera,” 2011.

X. Wu, W. Li, D. Hong, R. Tao, and Q. Du, “Deep learning for unmanned aerial vehicle-based object detection and tracking: A survey,” IEEE Geoscience and Remote Sensing Magazine, vol. 10, no. 1, pp. 91– 124, 2022.

P. Chakravarty, K. Kelchtermans, T. Roussel, S. Wellens, T. Tuytelaars, and L. Van Eycken, “Cnn-based single image obstacle avoidance on a quadrotor,” in 2017 IEEE International Conference on Robotics and Automation (ICRA), 2017, pp. 6369–6374.

K. D. Julian, S. Sharma, J.-B. Jeannin, and M. J. Kochenderfer, “Verifying aircraft collision avoidance neural networks through linear approximations of safe regions,” 2019. [Online]. Available: [link]

A. Kouris and C.-S. Bouganis, “Learning to fly by myself: A selfsupervised cnn-based approach for autonomous navigation,” in 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018, pp. 1–9.

K. McGuire, G. de Croon, C. De Wagter, K. Tuyls, and H. Kappen, “Efficient optical flow and stereo vision for velocity estimation and obstacle avoidance on an autonomous pocket drone,” IEEE Robotics and Automation Letters, vol. 2, no. 2, pp. 1070–1076, 2017.

X. Hua, X. Wang, T. Rui, F. Shao, and D. Wang, “Light-weight uav object tracking network based on strategy gradient and attention mechanism,” Knowledge-Based Systems, vol. 224, p. 107071, 2021. [Online]. Available: [link]

D. Wierzbicki, “Multi-camera imaging system for uav photogrammetry,” Sensors, vol. 18, no. 8, 2018. [Online]. Available: [link]

M. Gao, C. H. Hugenholtz, T. A. Fox, M. Kucharczyk, T. E. Barchyn, and P. R. Nesbit, “Weather constraints on global drone flyability,” Scientific Reports, vol. 11, no. 1, Jun. 2021. [Online]. DOI: 10.1038/s41598-021-91325-w

Q. xing Zhang, G. hua Lin, Y. ming Zhang, G. Xu, and J. jun Wang, “Wildland forest fire smoke detection based on faster r-cnn using synthetic smoke images,” Procedia Engineering, vol. 211, pp. 441–446, 2018, 2017 8th International Conference on Fire Science and Fire Protection Engineering (ICFSFPE 2017). [Online]. Available: [link]

M. G. Prasad, S. Chandran, and M. S. Brown, “A motion blur resilient fiducial for quadcopter imaging,” in 2015 IEEE Winter Conference on Applications of Computer Vision, 2015, pp. 254–261.

D. Acharya, K. Khoshelham, and S. Winter, “Bim-posenet: Indoor camera localisation using a 3d indoor model and deep learning from synthetic images,” ISPRS Journal of Photogrammetry and Remote Sensing, vol. 150, pp. 245–258, 2019. [Online]. Available: [link]

Y. Peng, Z. Tang, G. Zhao, G. Cao, and C. Wu, “Motion blur removal for uav-based wind turbine blade images using synthetic datasets,” Remote Sensing, vol. 14, no. 1, 2022. [Online]. Available: [link]

M. G. Ferrick, “Analysis of river wave types,” Water Resources Research, vol. 21, no. 2, pp. 209–220, 1985. [Online]. Available: [link]

T. Oktay, H. Celik, and I. Turkmen, “Maximizing autonomous performance of fixed-wing unmanned aerial vehicle to reduce motion blur in taken images,” Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering, vol. 232, no. 7, pp. 857–868, Mar. 2018. [Online]. DOI: 10.1177/0959651818765027

S. Dai and Y. Wu, “Motion from blur,” in 2008 IEEE Conference on Computer Vision and Pattern Recognition, 2008, pp. 1–8.

L. Noci, K. Roth, G. Bachmann, S. Nowozin, and T. Hofmann, “Disentangling the roles of curation, data-augmentation and the prior in the cold posterior effect,” in Advances in Neural Information Processing Systems, M. Ranzato, A. Beygelzimer, Y. Dauphin, P. Liang, and J. W. Vaughan, Eds., vol. 34. Curran Associates, Inc., 2021, pp. 12 738–12 748. [Online]. Available: [link]

S. Ruder, “An overview of gradient descent optimization algorithms,” 2016. [Online]. Available: [link]

T. Ergen, A. Sahiner, B. Ozturkler, J. M. Pauly, M. Mardani, and M. Pilanci, “Demystifying batch normalization in relu networks: Equivalent convex optimization models and implicit regularization,” CoRR, vol. abs/2103.01499, 2021. [Online]. Available: [link]