Issue |
ITM Web Conf.
Volume 70, 2025
2024 2nd International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2024)
|
|
---|---|---|
Article Number | 01013 | |
Number of page(s) | 7 | |
Section | Traffic Prediction and Analysis | |
DOI | https://doi.org/10.1051/itmconf/20257001013 | |
Published online | 23 January 2025 |
Scheme Analysis for Enhancing Autonomous Driving Based on Computer Vision
1 School of Economics & Management, Beijing Forestry University, 100083 Beijing, China
2 Chengdu Foreign Languages School, 611731 Chengdu, China
3 Nanjing Foreign Language School, 210008 Nanjing, China
* Corresponding author: chenxinyi2021@bjfu.edu.cn
This article explores the application and challenges of computer vision technology in autonomous driving, a critical component for the advancement of this field. Thesis adopted both literature review and technical analysis, focusing on recent developments in key technologies such as image processing, hybrid convolutional neural network (CNN)-transformer models, object detection, and multi-sensor fusion. The principles, benefits, limitations, and practical challenges of each technology were examined in detail. Thesis findings indicate that CNNs and their variants excel in tasks like object detection and semantic segmentation, significantly enhancing system perception and accuracy. Additionally, multi-sensor fusion technology boosts the reliability and the robustness of autonomous driving systems. However, challenges remain, including high computational demands, environmental perception accuracy, multi-sensor data fusion efficiency, and the high costs associated with implementation. Future research will prioritize developing highly effective deep learning models and optimizing cognitive computing visual systems to ameliorate the efficiency and ensure the safety of autonomous driving. The insights from this study offer valuable references for advancing autonomous driving technology and guide future research directions.
© The Authors, published by EDP Sciences, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.