Deformed trademark retrieval based on 2D pseudo-bidden Markov model

A new deformed trademark retrieval method based on two-dimensional pseudo-h idden Markov model(2D PHMM) is proposed in this paper. Most trademark retri eval systems focus on color features, shape silhouettes, or the combination of color and shape. However these approaches adopted individual silhouette s as shape features, leading to the following two crucial problems. First, most trademarks have various numbers of decomposed components, while the si lhouette-based approaches cannot handle the variety correctly. Second, the infringement eases in which trademarks are changed by non-rigid deformation , in particular nonlinear deformation, may escape detection. Thus, our meth od focuses on the overall appearance of trademarks and incorporates color a nd shape features into 2D PHMM to tackle the above two problems. The reason to involve 2D PHMM is that it has high tolerance to noise and distortion, moreover, contextual information can be incorporated into it in a natural a nd elegant way. However, 2D PHMM is computation intensive and sensitive to rotation, scale and translation variations. Thus, it is the main originalit y of this paper to include the advantages of 2D PHMM but to exclude its dis advantages. As a result, similar trademarks can be retrieved effectively, e ven those with different numbers of components or. non-rigid deformation. V arious experiments have been conducted on a trademark database to prove the effectiveness and practicability of the proposed method. (C) 2001 Pattern Recognition Society. published by Elsevier Science Ltd. All rights reserved .