PERSON RE-IDENTIFICATION FOR PUBLIC SAFETY IN INDIAN RAILWAYS USING DEEP LEARNING

Abstract

With the rapid growth of railway transportation in India, ensuring passenger safety and effective surveillance has become a major concern. Indian Railways handles millions of passengers daily, making it challenging to monitor and track individuals across multiple stations and platforms. Traditional surveillance systems rely heavily on manual monitoring, which is time-consuming and prone to human error. This project proposes a deep learning-based person re-identification system to enhance public safety in Indian Railways by automatically recognizing and tracking individuals across different camera views. The proposed system utilizes advanced deep learning techniques such as Convolutional Neural Networks (CNNs) to extract distinctive features from images of individuals captured by surveillance cameras. These features include appearance-based attributes such as clothing, body structure, and movement patterns. The system compares these features across multiple camera feeds to identify whether the same person appears in different locations. This enables continuous tracking of individuals without relying on facial recognition alone, which may not always be reliable in crowded or low-quality video conditions. The model is trained using large-scale datasets containing diverse images of people under varying conditions such as lighting, angles, and occlusions. Techniques such as data augmentation and feature normalization are used to improve model robustness and accuracy. The system can be integrated with existing CCTV infrastructure to provide real-time monitoring and alerts for suspicious activities or missing persons. Overall, the proposed solution enhances surveillance capabilities and improves public safety by enabling efficient tracking of individuals in complex environments. It can assist law enforcement agencies in crime prevention, investigation, and crowd management. The system provides a scalable and intelligent approach to modern railway security challenges using deep learning technologies.