Speedy Vision-based Human Detection Using Lightweight Deep Learning Network

Gede Erik Aktama, Franky Manoppo, Rosdiana Simbolon, Adityo Clinton Laloan, Andreas Sumendap, Muhamad Dwisnanto Putro

Abstract


Person detection plays a role as the initial system of video surveillance analysis with various implementations, such as activity analysis, person re-id, behavior analysis, and tracking analysis. The demand for efficient models drives a deep learning architecture with a superficial structure that can operate in real-time. You look only once (YOLO) object detection has been presented as an accurate detector that can operate in real-time. The speed limitation, huge computation cost, and abundant parameters still leave vital issues to improve the efficiency of this architecture. Lightweight human detection is proposed by utilizing the YOLOv5n framework. Modifying layer depth promotes a detection system that can operate fast and without stuttering. As a result, the proposed detector has satisfactory performance and is competitive with existing models. It achieves a mAP of 45.2%, closely competing with other person detectors. Additionally, it can run fast without stumbling at 26 frames per second. The detector's speed offers the advantage of this work that it can be feasibly implemented on a cpu device without a graphics accelerator.

Keywords


Person detection; efficient YOLO; real-time detector; central processing unit; surveillance system

Full Text:

PDF

References


S. Zhu, R. G. Guendel, A. Yarovoy, and F. Fioranelli, “Continuous Human Activity Recognition with Distributed Radar Sensor Networks and CNN-RNN Architectures,” IEEE Transactions on Geoscience and Remote Sensing, vol. 60, 2022, doi: 10.1109/TGRS.2022.3189746.

Y. Tang, X. Yang, X. Jiang, N. Wang, and X. Gao, “Dually Distribution Pulling Network for Cross-Resolution Person Reidentification,” IEEE Trans Cybern, vol. 52, no. 11, pp. 12016–12027, Nov. 2022, doi: 10.1109/TCYB.2021.3077500.

C. Cui and R. Xu, “Multiple Machine Learning Algorithms for Human Smoking Behavior Detection,” in Proceedings - 2022 International Conference on Machine Learning and Intelligent Systems Engineering, MLISE 2022, Institute of Electrical and Electronics Engineers Inc., 2022, pp. 240–244. doi: 10.1109/MLISE57402.2022.00054.

T. Zhou and Y. Liu, “Long-Term Person Tracking for Unmanned Aerial Vehicle Based on Human-Machine Collaboration,” IEEE Access, vol. 9, pp. 161181–161193, 2021, doi: 10.1109/ACCESS.2021.3132077.

Q. Bai, J. Xin, M. Yan, Y. Wang, E. Li, and S. Zhao, “An optimized mask-guided mobile pedestrian detection network with millisecond scale,” in Proceedings - 2020 Chinese Automation Congress, CAC 2020, Institute of Electrical and Electronics Engineers Inc., Nov. 2020, pp. 4975–4980. doi: 10.1109/CAC51589.2020.9326617.

X. Li, X. Luo, H. Hao, F. Chen, and M. Li, “Pedestrian detection method based on multi-scale fusion inception-SSD model,” 2020, pp. 1549–1553. doi: 10.1109/ITAIC49862.2020.9338909.

F. B. Setiawan, C. B. Adipradana, and L. H. Pratomo, “Fruit Ripeness Classification System Using Convolutional Neural Network (CNN) Method,” PROtek : Jurnal Ilmiah Teknik Elektro, vol. 10, no. 1, p. 46, Jan. 2023, doi: 10.33387/protk.v10i1.5549.

K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE Computer Society, Dec. 2016, pp. 770–778. doi: 10.1109/CVPR.2016.90.

K. Simonyan and A. Zisserman, “Very Deep Convolutional Networks for Large-Scale Image Recognition,” Sep. 2014, [Online]. Available: http://arxiv.org/abs/1409.1556

A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet Classification with Deep Convolutional Neural Networks.” [Online]. Available: http://code.google.com/p/cuda-convnet/

X. Zhang, S. Cao, and C. Chen, “Scale-Aware Hierarchical Detection Network for Pedestrian Detection,” IEEE Access, vol. 8, pp. 94429–94439, 2020, doi: 10.1109/ACCESS.2020.2995321.

M. D. Putro, L. Kurnianggoro, and K. H. Jo, “High Performance and Efficient Real-Time Face Detector on Central Processing Unit Based on Convolutional Neural Network,” IEEE Trans Industr Inform, vol. 17, no. 7, pp. 4449–4457, Jul. 2021, doi: 10.1109/TII.2020.3022501.

D. Chen, S. Xia, and Y. Zhou, “Pedestrian detection via contour fragments,” in Chinese Control Conference, CCC, IEEE Computer Society, Aug. 2016, pp. 4054–4060. doi: 10.1109/ChiCC.2016.7553986.

C. B. Murthy and M. Farukh Hashmi, “Real Time Pedestrian Detection Using Robust Enhanced Tiny-YOLOv3,” in 2020 IEEE 17th India Council International Conference, INDICON 2020, Institute of Electrical and Electronics Engineers Inc., Dec. 2020. doi: 10.1109/INDICON49873.2020.9342082.

J. An, M. D. Putro, A. Priadana, Y. Lee, J. Kim, and K. Jo, “YOLOv5 with Dual Attention Network for Object Detection on Drone,” in 2023 International Workshop on Intelligent Systems (IWIS), IEEE, Aug. 2023, pp. 1–6. doi: 10.1109/IWIS58789.2023.10284592.

C.-Y. Wang, A. Bochkovskiy, and H.-Y. M. Liao, “YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors,” Jul. 2022, [Online]. Available: http://arxiv.org/abs/2207.02696

T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár, “Focal Loss for Dense Object Detection.”

S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” Jun. 2015, [Online]. Available: http://arxiv.org/abs/1506.01497

G. Jocher et al., “ultralytics/yolov5: v3.1 - Bug Fixes and Performance Improvements.” Zenodo, Oct. 2020. doi: 10.5281/zenodo.4154370.

M. and B. S. and H. J. and P. P. and R. D. and D. P. and Z. C. L. Lin Tsung-Yi and Maire, “Microsoft COCO: Common Objects in Context,” in Computer Vision – ECCV 2014, T. and S. B. and T. T. Fleet David and Pajdla, Ed., Cham: Springer International Publishing, 2014, pp. 740–755.

M. Everingham, L. Gool, C. K. Williams, J. Winn, and A. Zisserman, “The Pascal Visual Object Classes (VOC) Challenge,” Int. J. Comput. Vision, vol. 88, no. 2, pp. 303–338, Jun. 2010, doi: 10.1007/s11263-009-0275-4.

M. Xu, Z. Wang, X. Liu, L. Ma, and A. Shehzad, “An Efficient Pedestrian Detection for Realtime Surveillance Systems Based on Modified YOLOv3,” IEEE Journal of Radio Frequency Identification, vol. 6, pp. 972–976, 2022, doi: 10.1109/JRFID.2022.3212907.

H. H. Nguyen, T. N. Ta, N. C. Nguyen, V. T. Bui, H. M. Pham, and D. M. Nguyen, “YOLO Based Real-Time Human Detection for Smart Video Surveillance at the Edge,” in ICCE 2020 - 2020 IEEE 8th International Conference on Communications and Electronics, Institute of Electrical and Electronics Engineers Inc., Jan. 2021, pp. 439–444. doi: 10.1109/ICCE48956.2021.9352144.

M. D. Putro, D. L. Nguyen, and K. H. Jo, “A CPU-based Pedestrian Detector using Deep Learning for Intelligent Surveillance Systems,” in Proceedings of the IEEE International Conference on Industrial Technology, Institute of Electrical and Electronics Engineers Inc., 2022. doi: 10.1109/ICIT48603.2022.10002735.

M. D. Putro, D. L. Nguyen, A. Priadana, and K. H. Jo, “Fast Person Detector with Efficient Multi-level Contextual Block for Supporting Assistive Robot,” in Proceedings - 2022 IEEE 5th International Conference on Industrial Cyber-Physical Systems, ICPS 2022, Institute of Electrical and Electronics Engineers Inc., 2022. doi: 10.1109/ICPS51978.2022.9816965.

W. Y. Hsu and W. Y. Lin, “Adaptive Fusion of Multi-Scale YOLO for Pedestrian Detection,” IEEE Access, vol. 9, pp. 110063–110073, 2021, doi: 10.1109/ACCESS.2021.3102600.




DOI: https://doi.org/10.33387/protk.v11i2.7030

Refbacks

  • There are currently no refbacks.


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



Editorial Office :
Protek : Jurnal Ilmiah Teknik Elektro
Department of Electrical Engineering. Faculty of Engineering. Universitas Khairun.
Address: Jusuf Abdulrahman 53 Gambesi, Ternate City, Indonesia.
Email: protek@unkhair.ac.id, WhatsApp: +6282292852552
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

View Stat Protek