Abstract
This paper describes a vision-based algorithm for recognizing the vehicle model type from time-sequence road images. Many types of vehicle models are offered commercially, and some of them are resemble in shape. This prevents us to discriminate their model types from the others easily. To solve these problems, we proposes a hierarchical recognition method with training process, in which the resemble model groups are firstly generated and the effective features to discriminate the models in the each group are then selected using the subspace method in training. In the recognition process, a front area is firstly detected from each frame of the input time-sequence images, then a hierarchical recognition which consists of a group and a category discrimination is performed. Finally, the results of frame recognition are integrated to realize stable recognition. The experimental results using time-sequence road images show the proposed method is effective: the recognition rate for the registered model types is more than 99%, and the rejection rate for unregistered vehicle type is more than 92%.
