This advanced autonomous system is designed for the detection of suspicious objects in busy railway environment. Leveraging the technology advancement and the power of artificial intelligence, it employs cutting-edge computer vision techniques and deep learning algorithms in a novel approach to identify potential threats from CCTV footage. One of the key strengths of this system is its ability to overcome challenges arisen in busy scenes for conventional object detection and tracking algorithms, due to dynamic foreground occlusions and complex backgrounds. Railway specific datasets were collected, specialized deep learning models, and novel integration of algorithms were developed to overcome these challenges and enhanced the detection performance that suits the busy railway environment. Through continuous real time analysis of video streams, the system promptly detects any prohibited items and unattended objects that may pose a risk, then immediately alerts station operators, enabling them to take appropriate and timely actions to address the situation and mitigate the safety and security risk. This uplifts the security levels in railway environment and hence ensures the safety of passengers and staff. With its autonomous operation and advanced detection capabilities, the system becomes a crucial tool in safeguarding public spaces and transportation, and preventing potential security incidents.
