Next Article in Journal
Energy Management Strategy for Hybrid Electric Vehicles Based on Experience-Pool-Optimized Deep Reinforcement Learning
Previous Article in Journal
Review of I–V Electrical Characterization Techniques for Photovoltaic Modules Under Real Installation Conditions
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Article

EAR-CCPM-Net: A Cross-Modal Collaborative Perception Network for Early Accident Risk Prediction

by
Wei Sun
*,
Lili Nurliyana Abdullah
,
Fatimah Binti Khalid
and
Puteri Suhaiza Binti Sulaiman
Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Selangor 43400, Malaysia
*
Author to whom correspondence should be addressed.
Appl. Sci. 2025, 15(17), 9299; https://doi.org/10.3390/app15179299
Submission received: 26 July 2025 / Revised: 21 August 2025 / Accepted: 23 August 2025 / Published: 24 August 2025

Abstract

Early traffic accident risk prediction in complex road environments poses significant challenges due to the heterogeneous nature and incomplete semantic alignment of multimodal data. To address this, we propose a novel Early Accident Risk Cross-modal Collaborative Perception Mechanism Network (EAR-CCPM-Net) that integrates hierarchical fusion modules and cross-modal attention mechanisms to enable semantic interaction between visual, motion, and textual modalities. The model is trained and evaluated on the newly constructed CAP-DATA dataset, incorporating advanced preprocessing techniques such as bilateral filtering and a rigorous MINI-Train-Test sampling protocol. Experimental results show that EAR-CCPM-Net achieves an AUC of 0.853, AP of 0.758, and improves the Time-to-Accident (TTA0.5) from 3.927 s to 4.225 s, significantly outperforming baseline methods. These findings demonstrate that EAR-CCPM-Net effectively enhances early-stage semantic perception and prediction accuracy, providing an interpretable solution for real-world traffic risk anticipation.
Keywords: multimodal fusion; traffic accident prediction; early risk perception; semantic interaction multimodal fusion; traffic accident prediction; early risk perception; semantic interaction

Share and Cite

MDPI and ACS Style

Sun, W.; Abdullah, L.N.; Khalid, F.B.; Suhaiza Binti Sulaiman, P. EAR-CCPM-Net: A Cross-Modal Collaborative Perception Network for Early Accident Risk Prediction. Appl. Sci. 2025, 15, 9299. https://doi.org/10.3390/app15179299

AMA Style

Sun W, Abdullah LN, Khalid FB, Suhaiza Binti Sulaiman P. EAR-CCPM-Net: A Cross-Modal Collaborative Perception Network for Early Accident Risk Prediction. Applied Sciences. 2025; 15(17):9299. https://doi.org/10.3390/app15179299

Chicago/Turabian Style

Sun, Wei, Lili Nurliyana Abdullah, Fatimah Binti Khalid, and Puteri Suhaiza Binti Sulaiman. 2025. "EAR-CCPM-Net: A Cross-Modal Collaborative Perception Network for Early Accident Risk Prediction" Applied Sciences 15, no. 17: 9299. https://doi.org/10.3390/app15179299

APA Style

Sun, W., Abdullah, L. N., Khalid, F. B., & Suhaiza Binti Sulaiman, P. (2025). EAR-CCPM-Net: A Cross-Modal Collaborative Perception Network for Early Accident Risk Prediction. Applied Sciences, 15(17), 9299. https://doi.org/10.3390/app15179299

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop