AN EFFICIENT ALGORITHM FOR HAND WRITTEN SIGNATURE RECOGNITION USING TRANSFORM BASEDAPPROACH WITH IMAGE STATISTICS
Kesana Mohana Lakshmi, Tummala Ranga Babu · Journal of Critical Reviews · 2020
In present scenario world forgery of signature is very much increased. Novel techniques are there to identify a correct signature. Hand written or bio metric signatures are widely used for authentication and identification of a person because every person has a own style of signature with its specific structures, so it is needy and necessary to recognize the authenticity of person hand written signature. There are limitations and complexities in recognizing a hand written signature image due to lack of abundant technical literature and sophisticated optimal methodologies. , we introduced an efficient approach for hand written signature recognition and retrieval using transformation-based approach Non-subsampled contourlet transform used for texture classification and the transform consists of Non-subsampled pyramid filter bank and Nonsubsampled directional filter bank Also gray level matrix is used to extract the texture features. In addition, image statistics is computed to enhance the recognition performance further. Finally, computed hamming distance. It measures the similarity between trained and test images, it is an effective distance metric over conventional Euclidean distance.