Offline Handwritten Signature Identification and Verification Using Multi-Resolution Gabor Wavelet

Abstract

In this paper, we are proposing a new method for offline (static) handwritten signature<br><br>identification and verification based on Gabor wavelet transform. The whole idea is offering a<br><br>simple and robust method for extracting features based on Gabor Wavelet which the dependency<br><br>of the method to the nationality of signer has been reduced to its minimal. After pre-processing<br><br>stage, that contains noise reduction and signature image normalisation by size and rotation, a<br><br>virtual grid is placed on the signature image. Gabor wavelet coefficients with different frequencies<br><br>and directions are computed on each points of this grid and then fed into a classifier. The shortest<br><br>weighted distance has been used as the classifier. The weight that is used as the coefficient for<br><br>computing the shortest distance is based on the distribution of instances in each of signature<br><br>classes.<br><br>As it was pointed out earlier, one of the advantages of this system is its capability of signature<br><br>identification and verification of different nationalities; thus it has been tested on four signature<br><br>dataset with different nationalities including Iranian, Turkish, South African and Spanish<br><br>signatures. Experimental results and the comparison of the proposed system with other systems<br><br>are consistent with desirable outcomes. Despite the use of the simplest method of classification<br><br>i.e. the nearest neighbour, the proposed algorithm in comparison with other algorithms has very<br><br>good capabilities. Comparing the results of our system with the accuracy of human's identification<br><br>and verification, it shows that human identification is more accurate but our proposed system has<br><br>a lower error rate in verification.