Pedestrian fatalities and injuries most likely occur in vehicle-pedestrian crashes. Meanwhile, engineers have tried to reduce the problems by developing a pedestrian detection function in Advanced Driver-Assistance Systems (ADAS) and autonomous vehicles. However, the system is still not perfect. A remaining problem in pedestrian detection is the performance reduction at nighttime, although pedestrian detection should work well regardless of lighting conditions. This study presents an evaluation of pedestrian detection performance in different lighting conditions, then proposes to adopt multispectral image and deep neural network to improve the detection accuracy. In the evaluation, different image sources including RGB, thermal, and multispectral format are compared for the performance of the pedestrian detection. In addition, the optimizations of the architecture of the deep neural network are performed to achieve high accuracy and short processing time in the pedestrian detection task. The result implies that using multispectral images is the best solution for pedestrian detection at different lighting conditions. The proposed deep neural network accomplishes a 6.9% improvement in pedestrian detection accuracy compared to the baseline method. Moreover, the optimization for processing time indicates that it is possible to reduce 22.76% processing time by only sacrificing 2% detection accuracy.

Autonomous vehicles keep evolving in years. Pedestrian detection is a significant function for autonomous vehicles. However, there is an issue that affects the performance of pedestrian detection, which is the lighting condition. Pedestrian detection performance usually decreases at nighttime. This research utilized Deep Neural Network as a tool to accomplish a pedestrian detection task in different lighting conditions. RGB, thermal infrared, and multispectral images were inputted to the DNN for pedestrian detection and it was evaluated afterward. The results imply that the multispectral image-based pedestrian detection can achieve better performance than the RGB image-based or thermal image-based pedestrian detections.