LiDAR-based estimation of bounding box coordinates using Gaussian process regression and particle swarm optimization

Abstract

;Camera-based object tracking systems in a given closed environment lack privacy and confidentiality. In this study, light detection and ranging (LiDAR) was applied to track objects similar to the camera tracking in a closed environment, guaranteeing privacy and confidentiality. The primary objective was to demonstrate the efficacy of the proposed technique through carefully designed experiments conducted using two scenarios. In Scenario I, the study illustrates the capability of the proposed technique to detect the locations of multiple objects positioned on a flat surface, achieved by analyzing LiDAR data collected from several locations within the closed environment. Scenario II demonstrates the effectiveness of the proposed technique in detecting multiple objects using LiDAR data obtained from a single, fixed location. Real-time experiments are conducted with human subjects navigating predefined paths. Three individuals move within an environment, while LiDAR, fixed at the center, dynamically tracks and identifies their locations at multiple instances. Results demonstrate that a single, strategically positioned LiDAR can adeptly detect objects in motion around it. Furthermore, this study provides a comparison of various regression techniques for predicting bounding box coordinates. Gaussian process regression (GPR), combined with particle swarm optimization (PSO) for prediction, achieves the lowest prediction mean square error of all the regression techniques examined at 0.01. Hyperparameter tuning of GPR using PSO significantly minimizes the regression error. Results of the experiment pave the way for its extension to various real-time applications such as crowd management in malls, surveillance systems, and various Internet of Things scenarios.

Keywords

LiDAR / Data acquisition / Bounding box / Gaussian process regression / Particle swarm optimization (PSO)

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

[1] H. Lvwen, C. Siyuan, Z. Jianfeng, C. Bang, L. Mingqing, Real-time motion tracking for indoor moving sphere objects with a LiDAR sensor, Sens. Transp. (2017).

[2] B. Drazen, K. Takayuki, I. Tetsushi, M. Takahiro, Person tracking in large public spaces using 3-D range sensors, IEEE Trans. Hum.-Mach. Syst. 43 (2013).

[3] C.H. Alejandra, G. Clara, C. Jonathan, B. Ramón, Object detection applied to indoor environments for mobile robot navigation, Phys. Sens. (2016).

[4] MahmudulHasan J. Hanawa R. Goto R. Suzuki H. Fukuda Y. Kuno Y. Kobayashi, LiDAR-based detection, tracking, and property estimation: A contemporary review, Neurocomputing 506 (2022) 393-405.

[5] Y. Wu, Y. Wang, S. Zhang, H. Ogai, Deep 3D object detection networks using LiDAR data: A review, IEEE Sens. J. 21 (2021) 1152-1171.

[6] S. Demetri, P. Gian, L. Bruzzone, Laps: Lidar-assisted placement of wireless sensor networks in forests, ACM Trans. Sensor Netw. 15 (2019) 1-40.

[7] C. Benedek, A. Majdik, B. sNagy, Z. Rozsa, T. Sziranyi, Positioning and perception in LIDAR point clouds, Digit. Signal Process. 119 (2021).

[8] W. Liu, S. Chen, E. Hauser, Lidar-based bridge structure defect detection, Exper. Tech. 35 (2017) 27-34.

[9] B. Behroozpour, A.M.S. Phillip, M.C. Wu, E.B. Bernhard, Lidar system architectures and circuits, in: IEEE Communications Magazine. Vol. 55, 2017, pp. 135-142.

[10] Y. Yao, S. Fatholahi, Y. Chen, J. Li, Indoor LiDAR point clouds upsampling for object detection enhancement, in: Joint Urban Remote Sensing Event, JURSE, 2023.

[11] R. Wang, M. An, S. Shao, M. Yu, S. Wang, X. Xu, Lidar sensor-based object recognition using machine learning, J. Russ. Laser Res. 42 (2021) 484-493.

[12] F. Bu, T. Le, X. Du, R. Vasudevan, R. Vasudevan, M. Johnson-Roberson, Pedestrian planar LiDAR pose (PPLP) network for oriented pedestrian detection based on planar LiDAR and monocular images, IEEE Robot. Autom. Lett. 5 (2020) 1626-1633.

[13] K. Sohn, A. Bernardi, O.R. Neiroukh, Effects of input resolution on LIDAR pedestrian localization using deep learning, in: International Conference on Electronics, Information, and Communication, ICEIC, 2021.

[14] J. Li, R. Stevenson, 2D LiDAR and camera fusion using motion cues for indoor layout estimation, in: IEEE 24th International Conference on Information Fusion, FUSION, 2021.

[15] H. Miawarni, T.A. Sardjono, E. Setijadi, Wijayanti, D. Arraziqi, A.B. Gumelar, M.H. Purnomo, Fall detection system for elderly based on 2D LiDAR: A preliminary study of fall incident and activities of daily living (ADL) detection,in:International Conference on Computer Engineering, Network, and Intelligent Multimedia, CENIM, 2020.

[16] S. Pintea, S. Karaoğlu, J. van Gemert, A. Smeulders, Large scale Gaussian process for overlap-based object proposal scoring, Comput. Vis. Image Understand. 150 (2016) 95-108.

[17] J. Kennedy, R. Eberhart,Particle swarm optimization, in:Proceedings of ICNN’95 - International Conference on Neural Networks, 1995.

[18] E.S. Gopi, Pattern recognition and computational intelligence tech-niques using matlab, in: Transactions on Computational Science and Computational Intelligence, TRACOSCI, Springer, 2020.

[19] S. Miao, Z.J. Wang, L. Rui, A CNN regression approach for real-time 2D/3D registration, IEEE Trans. Med. Imaging 35 (2016) 1352-1363.

[20] M.B. Christopher, Pattern recognition and machine learning, in: Informa-tion Science and Statistics. Vol. 1, Springer New York, NY, 2006, p. XX, 738.