INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
Accident detection is an essential application in intelligent transportation systems for the safety of drivers and passengers. Deep learning-based object identification algorithms have significantly improved in recent years in spotting objects in real time. YOLO (You Only Look Once) is one such model that has gained popularity due to its real-time performance and high accuracy. We propose an accident detection system in this paper using YOLOv3, the most recent version of YOLO. The proposed system is designed to detect three types of accidents, namely vehicle rollover, rear-end collision, and head-on collision. The system uses a pre-trained YOLOv3 model trained on the COCO dataset, which is fine-tuned on a custom dataset of accident images. The proposed system achieves an average precision of 0.94 for vehicle rollover detection, 0.93 for rear-end collision detection, and 0.92 for head-on collision detection. The system also shows promising results in terms of real-time performance, with an average processing time of 0.03 seconds per frame on an NVIDIA GeForce GTX 1080 Ti GPU. The proposed system can be integrated into intelligent transportation systems to provide real-time accident detection and alerting, improving the safety of drivers and passengers on the road.
"Real Time Vehicle Collision Detection Using Deep Learning", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.8, Issue 7, page no.323 - 329, July-2023, Available :http://www.ijsdr.org/papers/IJSDR2307044.pdf
Downloads:
000338536
Publication Details:
Published Paper ID: IJSDR2307044
Registration ID:206377
Published In: Volume 8 Issue 7, July-2023
DOI (Digital Object Identifier):
Page No: 323 - 329
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn