Image Processing Based Signature Verification Technique to Reduce Fraud in Financial Institutions

摘要: Handwritten signature is broadly utilized as personal verification in financial institutions ensures the necessity for a robust automatic tool. This tool aims to reduce fraud all related transactions’ sectors. paper proposes an online, robust, and technique using recent advances image processing machine learning. Once of handwritten customer captured, several pre-processing steps are performed on it including filtration detection edges. Afterwards, feature extraction process applied extract Speeded up Robust Features (SURF) Scale-Invariant Feature Transform (SIFT) features. Finally, developed compare extracted features with those stored database specified customer. Results indicate high accuracy, simplicity, rapidity technique, which main criteria judge banking other institutions.