Grid structured morphological pattern spectrum for off-line signature verification

In this paper, we present a grid structured morphological pattern spectrum based approach for off-line signature verification. The proposed approach has three major phases: preprocessing, feature extraction and verification. In the feature extraction phase, the signature image is partitioned into eight equally sized vertical grids and grid structured morphological pattern spectra for each grid is obtained. The grid structured morphological spectrum is represented in the form of 10-bin histogram and normalised to overcome the problem of scaling. The eighty dimensional feature vector is obtained by concatenating all the eight vertical morphological spectrum based normalised histogram. For verification purpose, we have considered two well known classifiers, namely SVM and MLP and conducted experiments on standard signature datasets namely CEDAR, GPDS-160 and MUKOS, a regional language (Kannada) dataset. The comparative study is also provided with the well known approaches to exhibit the performance of the proposed approach.