Target recognition and classification system based on probability fusion of camera-radar and method thereof

摘要

The present invention relates to a target recognition and classification system and method based on probability fusion of a camera and a radar. According to the present invention, N probability values are acquired by accumulating N class classification probabilities for a given target determined for every frame of the radar reception signal up to the current frame, and the first class judgment probability distribution including the obtained N probability values The radar signal processing unit outputting a, and the classification probability of N classes for the target determined for each frame of the camera image signal is accumulated up to the current frame to obtain N probability values, and a second class including the obtained N probability values A camera signal processing unit outputting a judgment probability distribution, and the first and second class judgment probability distributions for the target are grouped among the same class to obtain N result values, and then derive a maximum value among the N result values It provides a target recognition and classification system including a fusion determination unit to finally classify a class into the type of the target. According to the present invention, it is possible to increase target recognition and classification performance using a camera-radar fusion technique based on probability distribution, and has an advantage that can be utilized in various object surveillance systems as well as intelligent vehicles.