Signature Verification Using Support Vector Machine and Convolution Neural Network

Signature is used for recognition of an individual. considered as a mark that individual write on paper his/her identity or proof. It unique feature identifying highly in social and business functions which gives rise to verification signature. There are chances signature getting forged. Hence, the need identify genuine forged utmost important. In this paper, identification done using two approaches. First approach SVM second CNN. For SVM, pre-processing image extraction performed. Features extracted histogram gradient, shape, aspect ratio, bounding area, contour area convex hull area. Further, applied classify accuracy determined. approach, pre-processed, CNN Dataset here ICDAR Dutch dataset along with 80 signatures taken from 4 people.Dutch consists 362 imagesand images people 10 sums 442 images. The proposed system provides 86.39% around 83.78%