Off-line signature verification using elementary combinations of directional codes from boundary pixels

Abstract Verifying the genuineness of official documents, such as bank checks, certificates, contract forms, bonds, etc., remains a challenging task when it comes to accuracy and robustness. Here, is related degree match signature contained in documents relating original signatures authorized person. Signatures persons are considered known advance. In this paper, novel feature set introduced based on quasi-straightness boundary pixel runs for verification. We extract quasi-straight line segments using elementary combinations directional codes from pixels subsequently we obtain various classes. The provide blending straightness small curvatures resulting robust verification signatures. have used Support Vector Machine (SVM) classification shown results standard datasets like CEDAR (Center Excellence Document Analysis Recognition) GPDS-100 (Grupo de Procesado Digital la Senal). establish how proposed method outperforms existing state art.