Improved Pedestrian Detection Algorithm Based on YOLOv5s

Zhihua Li; Yuanbiao Zhang; Chao Wang; Guopeng Tan; Yahui Yan

doi:10.20965/jaciii.2024.p0768

single-jc.php

« previous

JACIII Vol.28 No.4 pp. 768-775

doi: 10.20965/jaciii.2024.p0768

(2024)

Research Paper:

Views over last 60 days: 275

Improved Pedestrian Detection Algorithm Based on YOLOv5s

Zhihua Li^*,**, Yuanbiao Zhang^*,**, Chao Wang^*,**,†, Guopeng Tan^,, and Yahui Yan^

^*School of Electronic and Information Engineering, Hebei University of Engineering
No.19 Taiji Road, Economic and Technological Development Zone, Handan, Hebei 056038, China

^**Hebei Key Laboratory of Security & Protection Information Sensing and Processing
No.19 Taiji Road, Handan Economic and Technological Development Zone, Handan, Hebei 056038, China

^***Xinxing Hebei Engineering and Research Inc., Ltd.
No.309 Xunzi North Street, Economic Development Zone, Handan, Hebei 056008, China

^†Corresponding author

Received:

June 22, 2023

Accepted:

December 20, 2023

Published:

July 20, 2024

Keywords:

pedestrian detection, multiscale detection, lightweight convolutions, YOLOv5s

Abstract

In this study, we propose YOLOv5s-PGD algorithm for dense pedestrian detection, which can improve the recall and reduce the number of parameters compared with YOLOv5s. First, a minimum scale detection layer has been added to deepen the pyramid’s depth and enhance detection accuracy. Second, ghost convolution has been employed to replace standard convolution to increase real-time performance of the algorithm. Finally, depth separable convolution has been used to address issues of high parameters and large computational complexity that lower the efficiency of the algorithm. Experiment results demonstrate that the detection accuracy of the YOLOv5s-PGD algorithm on the CrowdHuman public dataset is up to 85.1%, which is 2.2% higher than that of YOLOv5s. Furthermore, the number of parameters has decreased by 19.7%, and the calculation burden has decreased by 2.5%. Consequently, the proposed YOLOv5s-PGD algorithm better satisfies the requirements of real-time detection, model optimization, and terminal deployment in dense pedestrian scenarios.

Cite this article as:

Z. Li, Y. Zhang, C. Wang, G. Tan, and Y. Yan, “Improved Pedestrian Detection Algorithm Based on YOLOv5s,” J. Adv. Comput. Intell. Intell. Inform., Vol.28 No.4, pp. 768-775, 2024.

Data files:

References

[1] Z. Ma et al., “Spatial distribution, flowing rules and forming mechanism of inter-cities floating population in China,” Geographical Research, Vol.38, No.4, pp. 926-936, 2019 (in Chinese).
[2] R. Stewart, M. Andriluka, and A. Y. Ng, “End-to-end people detection in crowded scenes,” 2016 IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), pp. 2325-2333, 2016. https://doi.org/10.1109/CVPR.2016.255
[3] P. P. Shinde and S. Shah, “A review of machine learning and deep learning applications,” 2018 4th Int. Conf. on Computing Communication Control and Automation (ICCUBEA), 2018. https://doi.org/10.1109/ICCUBEA.2018.8697857
[4] R. Yu, X. Xu, and Z. Wang, “Influence of object detection in deep learning,” J. Adv. Comput. Intell. Intell. Inform., Vol.22, No.5, pp. 683-688, 2018. https://doi.org/10.20965/jaciii.2018.p0683
[5] B. Deng and H. Lv, “Survey of target detection based on neural network,” J. of Physics: Conf. Series, Vol.1952, Article No.022055, 2021. https://doi.org/10.1088/1742-6596/1952/2/022055
[6] S. Ren et al., “Faster R-CNN: Towards real-time object detection with region proposal networks,” IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol.39, No.6, pp. 1137-1149, 2017. https://doi.org/10.1109/TPAMI.2016.2577031
[7] J. Redmon et al., “You only look once: Unified, real-time object detection,” 2016 IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), pp. 779-788, 2016. https://doi.org/10.1109/CVPR.2016.91
[8] W. Liu et al., “SSD: Single shot multibox detector,” Proc. of the 14th European Conf. on Computer Vision (ECCV 2016), Part 1, pp. 21-37, 2016. https://doi.org/10.1007/978-3-319-46448-0_2
[9] Z. Tian et al., “FCOS: Fully convolutional one-stage object detection,” 2019 IEEE/CVF Int. Conf. on Computer Vision (ICCV), pp. 9626-9635, 2019. https://doi.org/10.1109/ICCV.2019.00972
[10] J. Liu and W. Meng, “Review on Single-Stage Object Detection Algorithm Based on Deep Learning,” Aero Weaponry, Vol.27, No.3, pp. 44-53, 2020.
[11] K. Han et al., “Image crowd counting using convolutional neural network and Markov random field,” J. Adv. Comput. Intell. Intell. Inform., Vol.21, No.4, pp. 632-638, 2017. https://doi.org/10.20965/jaciii.2017.p0632
[12] X. Shi et al., “A long-distance pedestrian small target detection method,” Chinese J. of Scientific Instrument, Vol.43, No.5, pp. 136-146, 2022 (in Chinese). https://doi.org/10.19650/j.cnki.cjsi.J2108848
[13] Q. Qin and J. Vychodil, “Pedestrian detection algorithm based on improved convolutional neural network,” J. Adv. Comput. Intell. Intell. Inform., Vol.21, No.5, pp. 834-839, 2017. https://doi.org/10.20965/jaciii.2017.p0834
[14] D. Liu et al., “Research on pedestrian target detection algorithm on VoVNet-FCOS,” Foreign Electronic Measurement Technology, Vol.40, No.11, pp. 64-71, 2021 (in Chinese). https://doi.org/10.19652/j.cnki.femt.2103000
[15] J. Deng and W. Wan, “Dense pedestrian detection based on improved YOLOv3,” Electronic Measurement Technology, Vol.44, No.11, pp. 90-95, 2021 (in Chinese). https://doi.org/10.19651/j.cnki.emt.2106129
[16] Y. Chen et al., “CA-YOLOv5 for crowded pedestrian detection,” Computer Engineering and Applications, Vol.58, No.9, pp. 238-245, 2022 (in Chinese).
[17] S. Shao et al., “CrowdHuman: A benchmark for detecting human in a crowd,” arXiv:1805.00123, 2018. https://doi.org/10.48550/arXiv.1805.00123
[18] X. Li et al., “Improved faster R-CNN for multi-scale object detection,” J. of Computer-Aided Design & Computer Graphics, Vol.31, No.7, pp. 1095-1101, 2019 (in Chinese).
[19] K. Han et al., “GhostNet: More features from cheap operations,” 2020 IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR), pp. 1577-1586, 2020. https://doi.org/10.1109/CVPR42600.2020.00165
[20] F. Chollet, “Xception: Deep learning with depthwise separable convolutions,” 2017 IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), pp. 1800-1807, 2017. https://doi.org/10.1109/CVPR.2017.195

This article is published under a Creative Commons Attribution-NoDerivatives 4.0 Internationa License.

[1] [1] Z. Ma et al., “Spatial distribution, flowing rules and forming mechanism of inter-cities floating population in China,” Geographical Research, Vol.38, No.4, pp. 926-936, 2019 (in Chinese).

[2] [2] R. Stewart, M. Andriluka, and A. Y. Ng, “End-to-end people detection in crowded scenes,” 2016 IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), pp. 2325-2333, 2016. https://doi.org/10.1109/CVPR.2016.255

[3] [3] P. P. Shinde and S. Shah, “A review of machine learning and deep learning applications,” 2018 4th Int. Conf. on Computing Communication Control and Automation (ICCUBEA), 2018. https://doi.org/10.1109/ICCUBEA.2018.8697857

[4] [4] R. Yu, X. Xu, and Z. Wang, “Influence of object detection in deep learning,” J. Adv. Comput. Intell. Intell. Inform., Vol.22, No.5, pp. 683-688, 2018. https://doi.org/10.20965/jaciii.2018.p0683

[5] [5] B. Deng and H. Lv, “Survey of target detection based on neural network,” J. of Physics: Conf. Series, Vol.1952, Article No.022055, 2021. https://doi.org/10.1088/1742-6596/1952/2/022055

[6] [6] S. Ren et al., “Faster R-CNN: Towards real-time object detection with region proposal networks,” IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol.39, No.6, pp. 1137-1149, 2017. https://doi.org/10.1109/TPAMI.2016.2577031

[7] [7] J. Redmon et al., “You only look once: Unified, real-time object detection,” 2016 IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), pp. 779-788, 2016. https://doi.org/10.1109/CVPR.2016.91

[8] [8] W. Liu et al., “SSD: Single shot multibox detector,” Proc. of the 14th European Conf. on Computer Vision (ECCV 2016), Part 1, pp. 21-37, 2016. https://doi.org/10.1007/978-3-319-46448-0_2

[9] [9] Z. Tian et al., “FCOS: Fully convolutional one-stage object detection,” 2019 IEEE/CVF Int. Conf. on Computer Vision (ICCV), pp. 9626-9635, 2019. https://doi.org/10.1109/ICCV.2019.00972

[10] [10] J. Liu and W. Meng, “Review on Single-Stage Object Detection Algorithm Based on Deep Learning,” Aero Weaponry, Vol.27, No.3, pp. 44-53, 2020.

[11] [11] K. Han et al., “Image crowd counting using convolutional neural network and Markov random field,” J. Adv. Comput. Intell. Intell. Inform., Vol.21, No.4, pp. 632-638, 2017. https://doi.org/10.20965/jaciii.2017.p0632

[12] [12] X. Shi et al., “A long-distance pedestrian small target detection method,” Chinese J. of Scientific Instrument, Vol.43, No.5, pp. 136-146, 2022 (in Chinese). https://doi.org/10.19650/j.cnki.cjsi.J2108848

[13] [13] Q. Qin and J. Vychodil, “Pedestrian detection algorithm based on improved convolutional neural network,” J. Adv. Comput. Intell. Intell. Inform., Vol.21, No.5, pp. 834-839, 2017. https://doi.org/10.20965/jaciii.2017.p0834

[14] [14] D. Liu et al., “Research on pedestrian target detection algorithm on VoVNet-FCOS,” Foreign Electronic Measurement Technology, Vol.40, No.11, pp. 64-71, 2021 (in Chinese). https://doi.org/10.19652/j.cnki.femt.2103000

[15] [15] J. Deng and W. Wan, “Dense pedestrian detection based on improved YOLOv3,” Electronic Measurement Technology, Vol.44, No.11, pp. 90-95, 2021 (in Chinese). https://doi.org/10.19651/j.cnki.emt.2106129

[16] [16] Y. Chen et al., “CA-YOLOv5 for crowded pedestrian detection,” Computer Engineering and Applications, Vol.58, No.9, pp. 238-245, 2022 (in Chinese).

[17] [17] S. Shao et al., “CrowdHuman: A benchmark for detecting human in a crowd,” arXiv:1805.00123, 2018. https://doi.org/10.48550/arXiv.1805.00123

[18] [18] X. Li et al., “Improved faster R-CNN for multi-scale object detection,” J. of Computer-Aided Design & Computer Graphics, Vol.31, No.7, pp. 1095-1101, 2019 (in Chinese).

[19] [19] K. Han et al., “GhostNet: More features from cheap operations,” 2020 IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR), pp. 1577-1586, 2020. https://doi.org/10.1109/CVPR42600.2020.00165

[20] [20] F. Chollet, “Xception: Deep learning with depthwise separable convolutions,” 2017 IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), pp. 1800-1807, 2017. https://doi.org/10.1109/CVPR.2017.195

Improved Pedestrian Detection Algorithm Based on YOLOv5s

Zhihua Li*,**, Yuanbiao Zhang*,**, Chao Wang*,**,†, Guopeng Tan*,**, and Yahui Yan***

Zhihua Li^*,**, Yuanbiao Zhang^*,**, Chao Wang^*,**,†, Guopeng Tan^,, and Yahui Yan^