International Journal Of Imaging

Abstract

Night Time Pedestrian Detection Based on A Fusion of Visual Information and Millimeter-Wave Radar

Author: B. Siva Krishna, Sana Anjum, N. Vamshi Krishna, Dr. B. Laxmi Kantha

Abstract:

Pedestrian detection has been a key area of research, particularly for enhancing road safety and aiding self-driving vehicles. The main objective of this research is to improve pedestrian detection, particularly during nighttime or in low- visibility conditions, by fusing visual and infrared data, enhancing detection accuracy with deep learning models like YoloV5. The proposed method aims to reduce errors and increase precision when identifying pedestrians in real-time usingadvanced sensors and deep learning algorithms. "Fusion of Visual and Infrared Information for Nightmare Pedestrian Detection" refers to combining visual (camera-based) data with infrared (heat-sensing) data to detect pedestrians, particularlyin challenging conditions such as nighttimedriving. "Nightmare" metaphorically highlights the difficulty of detecting pedestrians in low-visibility or extreme conditions. Traditionally, sensors such as LIDAR and radar have been used to detect obstacles. Before AI-bas