International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 Volume: 11 Issue: 5s DOI: https://doi.org/10.17762/ijritcc.v11i5s.6651 Article Received: 28 February 2023 Revised: 09 April 2023 Accepted: 25 April 2023 ___________________________________________________________________________________________________________________ 250 IJRITCC | May 2023, Available @ http://www.ijritcc.org Real-Time Vehicle Accident Recognition from Traffic Video Surveillance using YOLOV8 and OpenCV Deepak T. Mane 1* , Sunil Sangve 2 , Sahil Kandhare 3 , Saurabh Mohole 4 , Sanket Sonar 5 , Satej Tupare 6 1 Vishwakarma Institute of Technology, Pune-411037, Maharashtra, India 1 dtmane@gmail.com 2,3,4,5,6 JSPM's Rajarshi Shahu College of Engineering, Pune-411033, Maharashtra, India 2 sunilsangve@gmail.com 3 sahilkandhare07@gmail.com, 4 saurabhmohole@gmail.com 5 sanketsonar2002@gmail.com 6 satejtupare@gmail.com Abstract—The automatic detection of traffic accidents is a significant topic in traffic monitoring systems. It can reduce irresponsible driving behavior, improve emergency response, improve traffic management, and encourage safer driving practices. Computer vision can be a promising technique for automatic accident detection because it provides a reliable, automated, and speedy accident detection system that can improve emergency response times and ultimately save lives. This paper proposed an ensemble model that uses the YOLOv8 approach for efficient and precise event detection. The model framework's robustness is evaluated using YouTube video sequences with various lighting circumstances. The proposed model has been trained using the open-source dataset Crash Car Detection Dataset, and its produced precision, recall, and mAP are 93.8% and 98%, 96.1%, respectively, which is a significant improvement above the prior precision, recall, and mAP figures of 91.3%, 87.6%, and 93.8%. The effectiveness of the proposed approach in real-time traffic surveillance applications is proved by experimental results using actual traffic video data. Keywords-Anomal Detection; Accident detection; Computer vision; deep learning; YOLOV8 I. INTRODUCTION All Road traffic accidents have been a significant public safety concern worldwide. The World Health Organization (WHO) estimates those road traffic accidents cause 50 million extra injuries yearly and 1.35 million fatalities. As a result, there is an increasing demand for dependable, reliable, and efficient systems capable of detecting road accidents and responding quickly. Computer vision has developed as a viable tool for road accident detection and response, capturing real- time traffic data and analyzing it for signs of accidents using cameras and sensors. You Only Look Once version 8 (YOLOv8) and OpenCV are widely used tools for object detection and image processing among the computer vision algorithms available. This paper aims to use YOLOv8 and OpenCV to create an accurate and efficient system for the real- time recognition of car crashes in traffic surveillance. The proposed approach can help with rapid accident response, improve emergency response, and reduce the impact of accidents on public safety. There are numerous and diverse applications for accident detection using computer vision, potentially improving traffic management, transportation safety, emergency response, insurance, Autonomous Vehicles, Smart Cities, and many other fields. Accident detection systems based on computer vision have the potential to save lives and improve public safety. • Overall, this research paper contributes to are • To the development of computer vision-based systems for road accident detection and response. The proposed system, which uses YOLOv8 and OpenCV, can enhance emergency response, reduce the adverse effects of accidents on public safety, and even save lives. • The results of this study can be used to create more effective and efficient methods for detecting and responding to traffic accidents, which would eventually increase public safety and help save lives. The paper begins by examining existing computer vision- based road accident detection research, emphasizing YOLOv8 and the OpenCV. The literature survey covers the fundamentals of computer vision and the numerous methodologies that have been utilized in the past, represented in section II. The proposed YOLO8 and OPenCV8 ensemble model is described in section III. The Proposed Architecture section of the paper