Offline text-independent writer identification using codebook and efficient code extraction methods
Image and Vision Computing
| Hi-index | 0.00 |
Most studies about writer identification are based on English documents and to our knowledge no research has been reported on Farsi or Arabic documents. In this paper, we have proposed a method for off-line writer identification and verification based on Farsi handwriting, which is text-dependent. Based on the idea that has been presented in the previous studies, here we assume handwriting as texture image and after normalization step, the Gabor filters are applied to image and then new features are extracted. Substantially, the property of proposed method is using of the bank of Gabor filters which is appropriate for the structure of Farsi handwritten texts and vision system. Also, a new method for feature extraction from output of Gabor filters is proposed which is based on moments and nonlinear transform. In this paper, with definition a confidence criterion, a new method for writer verification is proposed. Evaluation of other methods and proposed method demonstrates that proposed method achieves better performance on Farsi handwritten from 40 peoples.