Single Classifier-based Passive System for Source Printer Classification using Local Texture Features
There are many methods available for printer identification of questioned documents, however most of them need identical contents of the training and testing documents. There is no effective method yet when the contents of the training and testing documents are different. To overcome this obstacle a method based on synthetic texture analysis is proposed in this paper. The inner texture of printed characters is reconstructed in order to generate an artificial textural image similar to the real texture while independent of document content. Fast Fourier Transform (FFT) and Gray Level Co-occurrence Matrix (GLCM) methods are then used for feature extraction. Experimental results demonstrate that the proposed method can achieve a high recognition rate and provide a new platform for printer identification independent on the content of documents.