A Method for Calligraphy Writer Identification by Integrating Gabor Filter and Gaussian Markov Random Field

摘要

Extracting effective features to describe texture is always a key problem in writer identification. This paper proposes a novel method for texture feature extraction by integrating Gabor filter and Gauss Markov Random Field (GMRF). That is to say, the handwriting images are firstly filtered by a bank of Gabor filters in which global features such as directional information can be detected; then GMRF models are developed for every filtered image that is flexible enough to capture the local spatial structure and the model parameters of all GMRFs are concatenated as texture features for writer identification, and finally, Support Vector Machine (SVM) is applied to evaluate the performance of calligraphic artist identification. The experimental results show that this method can achieve a classification rate of 98%, which has outperformed the traditional Gabor filter method.