Offline Signature Verification using Grid Indices Based Template Matching

Signature verification is a standout amongst the most broadly utilized biometrics for confirmation. This paper shows a novel methodology for offline signature verification. The proposed strategy depends on the grid features extraction from the image. For confirmation, the separated highlights of test signature are contrasted and the effectively prepared highlights of the reference signature. The procedure of verification then follows by calculating white indices in each grid cell for template matching purpose. This procedure is reasonable for different applications for example, bank exchanges, travel papers and so forth. The limit utilized in the proposed procedure can be progressively changed as indicated by the objective application. Essentially, the grid of matrix is taken for the feature extraction from the image. The extracted features are then compared with grid-based template matching which is based on counting the numbers of cumulative erros in both signature elements. The proposed system manages skilled imitations and has been tried on two types of signatures database. The proposed method grid- based template matching gives better result of FAR and FRR for the given set of signature database. The results of proposed method for given set of database includes simple& skilled signatures gives 0.55 % and 1.25 % FAR respectively. The system gives 5.12 % FRR which is superior to anything many existing confirmation procedures. This examination demonstrated that combination of white indices in the cell of the grid matrix and template matching may prompt better outcomes for offline signature acknowledgment and check. At last, creators outline the paper with some vital ends and call attention to the future potential research headings.