Trademark Image Retrieval Using Weighted Region Information Entropy

摘要

This paper describes a new trademark image retrieval method based on the weighted region information entropy. In the first stage, image is rotated according to principal orientation, and the object region in the rotated image is extracted. Then, the object region is partitioned into a lot of sub-blocks along its circle orientation. In the third stage, the weighted information entropy of each partitioned region is computed, which construct a feature vector for describing the shape of the image. Finally, the Euclidean distance is adopted to measure the similarity between the images based on the feature vector of each image obtained. Experiment results show that this method can keep good invariance under translation, rotation, and scale, and the retrieved results match human visual perception very well.