Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Automation and Machine Learning
  4. Vol 6, Issue 2, 2025

Review on Machine Vision-Based Detection of Pedestrians and Non-Motorized Vehicles in Autonomous Driving

Download as PDF

DOI: 10.23977/autml.2025.060206 | Downloads: 2 | Views: 72

Author(s)

Xueju Hao 1

Affiliation(s)

1 School of Electronic and Information Engineering, University of Science and Technology Liaoning, Anshan, China

Corresponding Author

Xueju Hao

ABSTRACT

Vulnerable Road Users (VRUs), including pedestrians, bicycles, and electric bikes, are the primary targets for collision risk prevention in autonomous driving systems due to their random motion and weak protection. Machine vision, as a core environmental perception technology, enables real-time detection and early warning of VRUs, which is crucial for the safety of autonomous driving. This paper focuses on the application of machine vision in VRU detection, systematically elaborates on the technical logic of detection and early warning, and emphasizes the characteristics and application values of mainstream datasets such as KITTI and Waymo Open Dataset. It deeply analyzes the detection bottlenecks in complex scenarios like nighttime driving and fast-moving pedestrians, and proposes corresponding technical optimization paths. Additionally, the integration ideas of machine vision and radar sensors are briefly discussed to improve the robustness of the detection system. The research shows that deep learning models (e.g., YOLOv8, DETR) and multi-sensor fusion technologies effectively enhance the accuracy and reliability of VRU detection. This review provides a comprehensive technical reference for the performance improvement of environmental perception systems in autonomous driving. 

KEYWORDS

Machine Vision; Autonomous Driving; Vulnerable Road Users; Pedestrian Detection; Non-Motorized Vehicle Detection; Sensor Fusion; KITTI Dataset

CITE THIS PAPER

Xueju Hao, Review on Machine Vision-Based Detection of Pedestrians and Non-Motorized Vehicles in Autonomous Driving. Automation and Machine Learning (2025) Vol. 6: 43-49. DOI: http://dx.doi.org/10.23977/autml.2025.060206.

REFERENCES

[1] Redmon, Joseph, and Ali Farhadi. "Yolov3: An incremental improvement." arxiv preprint arxiv:1804.02767 (2018).
[2] Ren, Shaoqing, et al. "Faster R-CNN: Towards real-time object detection with region proposal networks." IEEE transactions on pattern analysis and machine intelligence 39.6 (2016): 1137-1149.
[3] Carion, Nicolas, et al. "End-to-end object detection with transformers." European conference on computer vision. Cham: Springer International Publishing, 2020.
[4] Geiger, Andreas, Philip Lenz, and Raquel Urtasun. "Are we ready for autonomous driving? the kitti vision benchmark suite." 2012 IEEE conference on computer vision and pattern recognition. IEEE, 2012.
[5] Sun, Pei, et al. "Scalability in perception for autonomous driving: Waymo open dataset." Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2020.
[6] Howard, Andrew G., et al. "Mobilenets: Efficient convolutional neural networks for mobile vision applications." arxiv preprint arxiv:1704.04861 (2017).
[7] Xizhou Zhu, et al. "Deformable detr: Deformable transformers for end-to-end object detection." arxiv preprint arxiv:2010.04159 (2020).

Subscription
E-Mail Alert
Most popular papers
Call for Special Issue
Publication Ethics & OA Statement
Plagiarism Policy
Contact Us
Downloads: 4318
Visits: 199344

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.