HUMAN DETECTION IN SEARCH AND RESCUE OPERATIONS USING EMBEDDED ARTIFICIAL INTELLIGENCE
DOI:
https://doi.org/10.11113/jurnalteknologi.v86.19497Keywords:
Unmanned aerial vehicles, deep learning, transfer learning, TensorFlow, quantizationAbstract
The paper discusses the use of unmanned aerial vehicles (drones) in search and rescue operations to detect humans in disaster areas where rescue teams cannot reach. The paper highlights the limitations of current methods, including high computational power, high cost, and dependence on internet connectivity. The paper proposes using transfer learning to develop a human detection model with a mean average precision (mAP@0.5) above 90% and compares two deep learning models, MobileNet v2 and EfficientDet. The study uses multi-datasets of aerial images of humans, namely SeaDronesee and SARD, and the TensorFlow version 2.8 framework. MobileNet v2 required less GPU usage for training and yielded a relatively high accuracy of 95.5%, while EfficientDet achieved higher accuracy (97.3%). The trained MobileNet v2 model size is compressed using quantization from 25.5 MB to 4.15 MB, making it suitable for deployment on an edge device for on-chip inference. The paper concludes that the proposed method can improve the efficiency and effectiveness of search and rescue operations.
References
L. P. Osco et al. 2021. A Review on Deep Learning in UAV Remote Sensing. [Online]. Available: http://arxiv.org/abs/2101.10861.
M. M. Hasan et al. 2021. Search and Rescue Operation in Flooded Areas: A Survey on Emerging Sensor Networking-enabled IoT-oriented Technologies and Applications. Cogn Syst Res. 67: 104-123. Doi: 10.1016/j.cogsys.2020.12.008.
S. Mantowsky, F. Heuer, S. Saqib Bukhari, M. Keckeisen, and G. Schneider. 2021. ProAI: An Efficient Embedded AI Hardware for Automotive Applications - A Benchmark Study. IEEE International Conference on Computer Vision (ICCV). [Online]. Available: https://www.raspberrypi.org/.
R. David et al. 2020. TensorFlow Lite Micro: Embedded Machine Learning on TinyML Systems. [Online]. Available: http://arxiv.org/abs/2010.08678.
B. Sudharsan et al. 2021. TinyML Benchmark: Executing Fully Connected Neural Networks on Commodity Microcontrollers. 7th IEEE World Forum on Internet of Things, WF-IoT 2021. Institute of Electrical and Electronics Engineers Inc. 883-884. Doi: 10.1109/WF-IoT51360.2021.9595024.
H. Han and J. Siebert. 2022. TinyML: A Systematic Review and Synthesis of Existing Research. Institute of Electrical and Electronics Engineers (IEEE). 269-274. Doi: 10.1109/icaiic54071.2022.9722636.
A. R. Pathak, M. Pandey, and S. Rautaray. 2018. Application of Deep Learning for Object Detection. Procedia Computer Science. 1706-1717. Doi: 10.1016/j.procs.2018.05.144.
W. Ren, Y. Sun, H. Luo, and M. Guizani. 2022. A Demand-driven Incremental Deployment Strategy for Edge Computing in IoT Network. IEEE Trans Netw Sci Eng. 9(2): 416-430. Doi: 10.1109/TNSE.2021.3120270.
T. Marasović and V. Papić. 2019. Person Classification from Aerial Imagery using Local Convolutional Neural Network Features. Int J Remote Sens. 40(24): 9084-9102. Doi: 10.1080/01431161.2019.1597312.
D. Božić-Štulić, Ž. Marušić, and S. Gotovac. 2019. Deep Learning Approach in Aerial Imagery for Supporting Land Search and Rescue Missions. Int J Comput Vis. 127(9): 1256-1278. Doi: 10.1007/s11263-019-01177-1.
I. Imageryželjko, M. Maruši´c, D. Boži´boži´c-ˇ, S. Stuli´c, S. Gotovac, and T. Ton´co. 2018. Region Proposal Approach for Human Detection on Aerial Imagery; Region Proposal Approach for Human Detection on Aerial Imagery. 2018 3rd International Conference on Smart and Sustainable Technologies.
M. K. Vasić and V. Papić. 2020. Multimodel Deep Learning for Person Detection in Aerial Images. Electronics (Switzerland). 9(9): 1-15. Doi: 10.3390/electronics9091459.
N. M. K. Dousai and S. Loncaric. 2021. Detection of Humans in Drone Images for Search and Rescue Operations. ACM International Conference Proceeding Series. Association for Computing Machinery. 69-75. Doi: 10.1145/3449365.3449377.
S. Ghamari et al. 2021. Quantization-guided Training for Compact TinyML Models. [Online]. Available: http://arxiv.org/abs/2103.06231.
W. Liu et al. 2016. SSD: Single Shot MultiBox Detector. Computer Vision and Pattern Recognition. [Online]. Available: https://github.com/weiliu89/caffe/tree/ssd.
N. M. K. Dousai and S. Loncaric. 2021. Detection of Humans in Drone Images for Search and Rescue Operations. ACM International Conference Proceeding Series. Association for Computing Machinery. 69-75. Doi: 10.1145/3449365.3449377.
Ž. M. S. G. Dunja Božić-Štulić. 2019. Deep Learning Approach on Aerial Imagery in Supporting Land Search and Rescue Missions. International Journal of Computer Vision.
D. Božić-Štulić, Ž. Marušić, and S. Gotovac. 2019. Deep Learning Approach in Aerial Imagery for Supporting Land Search and Rescue Missions. Int J Comput Vis. 127(9): 1256-1278. Doi: 10.1007/s11263-019-01177-1.
E. Lygouras, N. Santavas, A. Taitzoglou, K. Tarchanidis, A. Mitropoulos, and A. Gasteratos. 2019. Unsupervised Human Detection with an Embedded Vision System on a Fully Autonomous UAV for Search and Rescue Operations. Sensors (Switzerland). 19(16). Doi: 10.3390/s19163542.
Y. Ampatzidis, V. Partel, and L. Costa. 2020. Agroview: Cloud-based Application to Process, Analyze and Visualize UAV-collected Data for Precision Agriculture Applications Utilizing Artificial Intelligence. Comput Electron Agric. 174. Doi: 10.1016/j.compag.2020.105457.
M. Tan and Q. v Le, “EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks.”
A. G. Howard et al. 2017. MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. [Online]. Available: http://arxiv.org/abs/1704.04861.
M. del Carmen Rodríguez-Hernández and S. Ilarri. 2021. AI-based Mobile Context-aware Recommender Systems from an Information Management Perspective: Progress and Directions. Knowl Based Syst. 215. Doi: 10.1016/j.knosys.2021.106740.
R. Girshick, J. Donahue, T. Darrell, and J. Malik. 2014. Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE Computer Society. 580-587. Doi: 10.1109/CVPR.2014.81.
G. Crocioni, D. Pau, J. M. Delorme, and G. Gruosso. 2020. Li-Ion Batteries Parameter Estimation with Tiny Neural Networks Embedded on Intelligent IoT Microcontrollers. IEEE Access. 8: 122135-122146. Doi: 10.1109/ACCESS.2020.3007046.
V. Janapa Reddi et al. 2022. Widening Access to Applied Machine Learning with TinyML. Harv Data Sci Rev. Doi: 10.1162/99608f92.762d171a.
Downloads
Published
Issue
Section
License
Copyright of articles that appear in Jurnal Teknologi belongs exclusively to Penerbit Universiti Teknologi Malaysia (Penerbit UTM Press). This copyright covers the rights to reproduce the article, including reprints, electronic reproductions, or any other reproductions of similar nature.
