An Efficient Object Recognition Method Based on Association Rule Mining

A method for classifying objects into categories and indexing is proposed to implement object recognition. The relational measurements such as the distance between two points, color comparison is encoded by the attributed relational graph (ARG) representation to provide one-to-one correspondence between models and object features. If the contour is traversed counterclockwise, a sequence can be formed to roughly represent the shape of the object. Second, for a set of similar objects, an association rule mining approach is used to extract the feature pattern, which is the common subsequences of those objects, to represent their common shapes. Third, for a testing image containing an object, its sequence can be computed according to the first phase. Then, the testing sequence can be used to match feature patterns generated in the second phase by a dynamic programming approach. If the testing sequence approximately contains a feature pattern, the object is identified as the object represented by the feature pattern. The experiment results show that the proposed method can use a limited number of training data and the objects with slight variations also can be recognized since feature patterns can mainly represent the shapes of these objects.