• Dr. K.Thenmozhi
• A.Thananjeyan
• M.Gokul Raj
• S.Santhosh Kumar

This study will introduce and apply the Object Detection and Tracking System (ODTS) for automatic detection and tracking of unexpected events on CCTVs in tunnels, which are likely to include (1) Wrong-Way Driving (WWD), (2) Stop, (3) Person out of vehicle in tunnel, and (4) Fire. Additionally introduced and used will be the Faster Regional Convolution Neural Network (Faster R-CNN) for Object Detection and the Traditional Object Tracking technique. Using the Bounding Box (box) findings, the Object Detection results for Bounding Box (box) are collected. In order to provide each moving and identified object a distinct ID number, it then compares the Box of the most recent and previous video frames. With this method, a moving object can be tracked in time, which is not currently possible.

Index term— a faster R-CNN for object detection, object tracking, object detection and tracking, unexpected event detection, and tunnel CCTV accident detection

"A DEEP LEARNING ALGORITHM APPLICATION FOR AUTOMATIC DETECTION OF UNPREDICTABLE ACCIDENTS ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 8, page no.c320-c324, August-2023, Available :http://www.jetir.org/papers/JETIR2308235.pdf

"A DEEP LEARNING ALGORITHM APPLICATION FOR AUTOMATIC DETECTION OF UNPREDICTABLE ACCIDENTS ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 8, page no. ppc320-c324, August-2023, Available at : http://www.jetir.org/papers/JETIR2308235.pdf