Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Journal of Artificial Intelligence Practice
  4. Vol 8, Issue 1, 2025

Research on the application of artificial intelligence and multi-scale image fusion technology to pedestrian detection in complex street view

Download as PDF

DOI: 10.23977/jaip.2025.080116 | Downloads: 14 | Views: 414

Author(s)

Gong Li 1

Affiliation(s)

1 Department of Artificial Intelligence, Faculty of Computer Science and Information Technology, University of Malaya, 50603 Kuala Lumpur, Malaysia

Corresponding Author

Gong Li

ABSTRACT

With the increasing face imaging data and the advancement of artificial intelligence (AI) technology, computer-aided monitoring systems are crucial for pedestrian detection in dense street view. However, due to occlusion and small pedestrian scale, pedestrian false alarms and missed detection problems become more and more serious. Therefore, this paper proposes a pedestrian detection model, YOLOv10s-pedestrian. Firstly, CA attention is introduced to redesign the MBConv module, resulting in an efficient MB-CANet backbone for pedestrian feature extraction, enhancing the accurate localization of densely occluded pedestrians. Secondly, a novel C2FN structure was created to reduce the number of parameters while improving the model's accuracy. Additionally, inspired by the BiFPN feature fusion concept, a Bi-C2FN-FPN network structure is proposed to effectively fuse features from different depth sources, strengthening feature fusion and improving pedestrian detection accuracy. Finally, the MPDIOU loss function replaces the original CIoU loss function to enhance anchor box localization. Experimental results demonstrate that the proposed model achieves a mAP50 of 95.6% on the WiderPerson pedestrian detection dataset, which is a 6.1% improvement over the original model, with a recall rate of 86.2%, showcasing excellent detection performance. Compared to several mainstream object detection models, the proposed model also exhibits superior performance. 

KEYWORDS

Pedestrian detection, Computer vision, Multiscale networks, YOLOv10s

CITE THIS PAPER

Gong Li, Research on the application of artificial intelligence and multi-scale image fusion technology to pedestrian detection in complex street view. Journal of Artificial Intelligence Practice (2025) Vol. 8: 119-131. DOI: http://dx.doi.org/10.23977/jaip.2025.080116.

REFERENCES

[1] Azam, S., Munir, F., Kyrki, V., Kucner, T.P., Jeon, M., Pedrycz, 2024. Exploring Contextual Representation and Multi- modality for End-to-end Autonomous Driving. 135, 108767, Engineering Applications of Artificial Intelligence.
[2] Bai, S., Wang, Y., Luo, Z., Tian, 2024. DriveCP: Occupancy-Assisted Scenario Augmentation for Occluded Pedestrian Perception Based on Parallel Vision. IEEE,Journal of Image and Graphics.
[3] Bar-Joseph, M., Ezra, D., Licciardello, G., Catara, A., 2023. Science and Tradition. Springer, pp. 145-215.Diseases of Etrog Citron and Other Citrus Trees, The Citron Compendium: The Citron (Etrog) Citrus medica L.
[4] Gao, H., Huang, S., Li, M., Li, 2024. Multi-scale Structure Perception and Global Context-aware Method for Small-scale Pedestrian Detection. IEEE,Towson University Journal of International Affairs. 
[5] Yuan, Q., Huang, G., Zhong, G., Yuan, X., Tan, Z., Lu, Z., Pun, C., Measurement, 2023. Triangular Chain Closed-Loop. Detection Network for Dense Pedestrian Detection.Transactions on Instrumentation and Measurement.
[6] Gong, W., Yang, S., Guang, H., Ma, B., Zheng, B., Shi, Y., Li, B., & Cao, Y. (2024). An intrusion detection scheme based on multi-order feature interaction to enhance cybersecurity in intelligent connected vehicles. Engineering Applications of Artificial Intelligence, 135, 108815.
[7] Guo, L., Ge, P.-S., Zhang, M.-H., Li, L.-H., & Zhao, Y.-B. (2012). Pedestrian detection in intelligent transportation systems using a combination of AdaBoost and support vector machines. Expert Systems with Applications, Elsevier.
[8] Hou, Q., Zhou, D., & Feng, J. "Designing Efficient Mobile Networks with Coordinate Attention". Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), vol. 2021.
[9] Hsu, W.-Y., & Yang, P.-Y. (2023). Multi-scale structure-enhanced super-resolution for pedestrian detection. IEEE Transactions on Intelligent Transportation Systems. 
[10] Jain, D. K., Zhao, X., Garcia, S., & Neelakandan, S. (2024). A robust deep convolutional neural network-based multimodal pedestrian detection model using ensemble learning. Expert Systems with Applications, Elsevier.
[11] Jiang, H., Liao, S., Li, J., Prinet, V., & Xiang, S. "Semantic Modulation for Urban Scene-Based Pedestrian Detection". Neurocomputing, vol. 474, pp. 1-12, 2022.
[12] Li, J., Bi, Y., Wang, S., Li, Technology, S.f.V., 2023. CFRLA-Net: A Context-Aware Feature Representation Learning Anchor- Free Network for Pedestrian Detection. IEEE.Transactions on Circuits and Systems for Video Technology.
[13] Zheng, A., Wang, H., Wang, J., Huang, H., He, R., Hussain, 2023. Diverse features discovery transformer for pedestrian attribute recognition. 119, 105708.Engineering Applications of Artificial Intelligence.
[14] Kilicarslan, M., Zheng., 2022. DeepStep: Direct detection of walking pedestrian from motion by a vehicle camera. IEEE .Transactions on Intelligent Vehicles.

Subscription
E-Mail Alert
Most popular papers
Call for Special Issue
Publication Ethics & OA Statement
Plagiarism Policy
Contact Us
Downloads: 15127
Visits: 485217

Sponsors, Associates, and Links

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.