Real-Time Pedestrian Recognition in Low-Light Conditions for Enhanced Surveillance

Pedestrian recognition plays a crucial role in modern surveillance systems, particularly in applications related to public safety and security monitoring. However, recognizing pedestrians in low-light or night-time conditions remains a significant challenge due to poor illumination, noise, and low contrast. Traditional surveillance systems often fail to provide reliable results under such conditions. This paper presents a real-time pedestrian recognition framework specifically designed for lowlight environments. The proposed system integrates low-light image enhancement techniques with deep learning-based pedestrian detection models to improve recognition accuracy. Contrast enhancement and noise reduction are applied as preprocessing steps, followed by a real-time object detection model to identify pedestrians efficiently. Experimental observations indicate that the proposed approach improves detection accuracy while maintaining real-time performance, making it suitable for enhanced surveillance applications.