High-altitude Multi-object Detection and Tracking based on Drone Videos

Limei Peng

¹ School of Computer Science and Engineering, Kyungpook National University, Daegu, 41566, Republic of Korea

∗ Corresponding author

Drone videos have more extensive shooting ranges, more angles, and no geographical limitations. Thus the object detection algorithm based on drone videos is increasingly playing a role in various fields, such as military surveillance, space remote sensing, smart city, disaster monitoring scenes, etc. Compared to low-altitude object detection and tracking (LA-ODT), high-altitude object detection and tracking (HA-ODT) are receiving increasing attention, especially in modern cities with massive high buildings, because of their higher flying h eight, w ider v iewing a ngle, a nd t he a bility t o t rack multiple f ast-moving o bjects s imultaneously. However, high-altitude aerial videos (HA-AVs) are constrained by small objects that can be measured, fewer feature points, occlusions, and light changes. Therefore, HA-AVs suffer from blurry images with fewer feature points of objects and missed detection due to occlusion, degrading the ODT accuracy. Since the accessible HA datasets are very limited, not to mention featured datasets considering angles, weather, etc., this paper directly uses drones to collect HA pictures and videos of different angles, different illuminations, and different heights for self-labeling training. Regarding this, we adopt super-resolution reconstruction to increase the data diversity and add artificial o cclusions t o e nhance t he c ollected d ata t o improve t he a ccuracy o f HA-ODT.

Drone videos, neural network, Multi-object detection and tracking

Qiang Zhao, Limei Peng (2022). High-altitude Multi-object Detection and Tracking based on Drone Videos. Journal of Networking and Network Applications, Volume 2, Issue 1, pp. 36–42. https://doi.org/10.33969/J-NaNA.2022.020103.

[1] Zou Z, Shi Z, Guo Y, et al. Object detection in 20 years: A survey. arXiv preprint arXiv:1905.05055, 2019.

[2] Yilmaz A, Javed O, Shah M. Object tracking: A survey. Acm computing surveys (CSUR), 2006, 38(4): 13-es.

[3] Gordon R, Herman G. Three-dimensional reconstruction from projec-tions: A review of algorithms. International review of cytology, 1974, 38: 111-151.

[4] Abderrahmane A, Bin Y, and Tarik T. Generalized Traffic Flow Model for Multi-Services Oriented UAV System. Journal of Networking and Network Applications,Vol.1, Iss.1, pp.1–8. https://doi.org/10.33969/J-NaNA.2021.010101

[5] Viola P,Jones M. Rapid object detection using a boosted cascade of simple features. Computer Vision and Pattern Recognition(CVPR), 2001. vol.1. IEEE, 2001, pp.1-1.

[6] Viola P, Jones M. Robust real-time face detection. International journal of computer vision, vol.57, no.2, pp.137–154, 2004.

[7] Ali S, Shah M. COCOA: tracking in aerial imagery.Airborne Intelli-gence, Surveillance, Reconnaissance (ISR) Systems and Applications

III. International Society for Optics and Photonics, 2006, 6209: 62090D.

[8] Girshick R, Donahue J, Darrell T. Regionbased convolutional networks for accurate object detection and segmentation. IEEE transactions on pattern analysis and machine intelligence, vol.38, no.1, pp.142– 158, 2016.

[9] Redmon J, Divvala S, Girshick R.You only look once: Unified, real-time object detection. IEEE conference on computer vision and pattern recognition, 2016, pp.779–788.

[10] Ren S, He K, Girshick R, and Sun J. Faster r-cnn: Towards real-time object detection with region proposal networks. in Advances in neural information processing systems, 2015, pp. 91–99.

[11] Abdelwahab M A, Abdelwahab M M. A novel algorithm for vehicle de-tection and tracking in airborne videos. IEEE International Symposium on Multimedia (ISM). IEEE, 2015: 65-68.

[12] Park S C, Park M K, Kang M G. Super-resolution image reconstruction: a technical overview. IEEE signal processing magazine, 2003, 20(3): 21-36.

[13] Girshick R. Fast r-cnn. IEEE international conference on computer vision. 2015: 1440-1448.

[14] Khokhlov I, Davydenko E, Osokin I, et al. Tiny-YOLO object detection supplemented with geometrical data. 2020 IEEE 91st Vehicular Tech-nology Conference (VTC2020-Spring). IEEE, 2020: 1-5.

[15] Zhang P, Zhong Y, Li X. SlimYOLOv3: Narrower, faster and better for real-time UAV applications. International Conference on Computer Vision Workshops. 2019: 0-0.

[16] Redmon J,Farhadi A. YOLOv3: an incremental improve-ment (2018). arXiv preprint arXiv:1804.02767 (1804). https://pjreddie.com/darknet/yolo/.

[17] Tan M, Pang R, Le Q V. Efficientdet: Scalable and efficient object detec-tion. IEEE CVF conference on computer vision and pattern recognition. 2020: 10781-10790.

[18] Tan M, Le Q. Efficientnet: Rethinking model scaling for convolutional neural networks. International Conference on Machine Learning. PMLR, 2019: 6105-6114.

[19] Dong C, Loy C C, He K, et al. Image super-resolution using deep convolutional networks. IEEE transactions on pattern analysis and machine intelligence, 2015, 38(2): 295-307.

[20] Glasner D, Bagon S, Irani M. Super-resolution from a single image. 2009 IEEE 12th international conference on computer vision. IEEE, 2009: 349-356.

[21] Ledig C, Theis L, Husz´ar F, et al. Photo-realistic single image super-resolution using a generative adversarial network. Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 4681-4690.

[22] Ji X, Cao Y, Tai Y, et al. Real-world super-resolution via kernel estimation and noise injection. Conference on Computer Vision and Pattern Recognition Workshops. 2020: 466-467.

[23] Grewal M S, Andrews A P. Kalman filtering: Theory and Practice with MATLAB. John Wiley & Sons, 2014.