This study presents the development of an embedded night-vision system designed to enhance pedestrian detection in low-light environments. As urban areas continue to grow and nighttime activities increase, ensuring pedestrian safety becomes paramount. Traditional vision-based systems often struggle under poor lighting conditions, leading to a higher risk of accidents. The proposed system leverages advanced infrared imaging technology combined with machine learning algorithms to detect pedestrians accurately in real-time. Utilizing a compact embedded platform, the system processes video feeds captured in low-light settings, applying convolutional neural networks (CNNs) for effective object recognition. The performance of the system is evaluated against various datasets, demonstrating significant improvements in detection accuracy and response time compared to conventional methods. Results indicate that the embedded night-vision system can reliably identify pedestrians, enhancing safety measures for both pedestrians and drivers in urban environments. This research contributes to the ongoing efforts to develop intelligent transportation systems and improve road safety through innovative technological solutions.

Thermal and IR Night vision, OpenCV, VideoProcessing, Object and Human Detection, Automobile Safty.