Offline handwritten signature verification using various Machine Learning Algorithms

In today’s world it is necessary to protect one’s authenticity in order ensure the protection of personal information that only authenticate credentials a person can have access to. Nowadays there an increase number malpractices like signature forgery important person. To encounter verification problem, been advances verifying using various techniques including Machine Learning and Deep Learning. This paper introduces novel approach verify signatures difference gaussian filtering technique, gray level co-occurrence matrix feature extraction principle component analysis kernel principal associated with machine learning algorithms. The publicly available Kaggle offline handwritten dataset used for training. article compares accuracy on After training datasets lowest achieved 56.66% Naive Bayes algorithm. highest 82% K-Nearest Neighbour (KNN) 81.66% Random Forest components dataset.