Parametric vehicle recognition using knowledge acquisition system

This research focuses on a technique to recognize vehicles using the knowledge acquisition system. The knowledge acquisition system consists of object recognition modules which represent a car image viewed form the rear, such as a window, tail lights, and so on. This knowledge system recognizes a car by detecting each object. The proposed system has an ability to modify itself in order to adapt to the shot condition of the input image. Sharpness is used as the parameter of the shot condition. The recognition modules are trained by merged data of the object image. As the sharpness is introduced, the knowledge system recognizes the vehicle in various condition images.