Offline Handwritten Signature Identification using Grid Gabor Features and Support Vector Machine

In this paper, a new method for signature identification based on wavelet transform is proposed. This method uses Gabor Wavelet Transform (GWT) as feature extractor and Support Vector Machine (SVM) as classifier. In proposed method, first signature image is normalized by size and then image is enhanced to remove noise. After pre-processing, a virtual grid is placed on signature image and Gabor coefficients are computed on each point of grid. Next, all Gabor coefficients are fed to a layer of SVM classifiers as feature vector. The number of SVM classifiers is equal to number of classes. Each SVM classifier determines that does the input image belong to corresponding class or not. The main characteristic of proposed method is independency to nation of signers. Two experiments on two signature sets were done. The first is on a Persian signature set and other is on a Turkish signature set. Based on these experiments, identification rate have achieved 96% and more than 93% on Persian and Turkish signature set respectively.

Keywords:

Signature Identification , Gabor Wavelet , Grid Features , Support Vector Machine