State-of-the-Art on IoV-Based Deep Learning Framework for Enhanced Driving Behavior Recognition: Recent Progress, Technology Updates, Challenges, and Future Direction
In recent years, the application of deep learning (DL) models to identify dangerous driving behaviors has emerged as a novel approach to enhance road safety and detect high-risk driving behaviors. However, these advanced algorithms still face several challenges. We analyze relevant literature from 2...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11104155/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In recent years, the application of deep learning (DL) models to identify dangerous driving behaviors has emerged as a novel approach to enhance road safety and detect high-risk driving behaviors. However, these advanced algorithms still face several challenges. We analyze relevant literature from 2015 to 2024, to uncover key trends and gaps in the development of applications for identifying dangerous driving behaviors, with a primary focus on optimizing integration to improve the accuracy and real-time capability of driving behavior prediction. The analysis reveals that data quality, computational demands, and privacy issues are currently hindering the full realization of this technology. The Internet of Vehicles (IoV) collects data based on real driving environments, offering unique advantages in terms of data timeliness, sufficiency, and diversity. Therefore, this paper proposes an active IoV based DL framework, emphasizing the feasibility of enhancing data processing techniques and algorithmic improvements to boost model accuracy and generalization ability, and highlighting the potential of integrating edge and cloud computing to support real-time data analysis and decision-making. Future research should concentrate on multimodal data fusion and federated learning to further enhance recognition performance and ensure broader applicability of deep learning in the internet of vehicles environments. |
|---|---|
| ISSN: | 2169-3536 |