Thermal image processing using neural network

This paper proposes a thermal image processing system that consists of a small, lightweight noncooled thermal image sensor, a thermal image pre-processor, and a structured neural network. This system can measure precisely the position and posture of occupants from feature data using a low-resolution thermal image captured in the polar coordinate system. This thermal image processing system is expected to develop into an image processing system that will be built in various household electric appliances that utilize the information on the occupants of a room.