TK707 : Offline handwriting image verification baxsed on morphology methods
Thesis > Central Library of Shahrood University > Electrical Engineering > MSc > 2018
Authors:
Abstarct: In this research we to propose an mathematical approch for identification and verifying writer from offline handwriting sample. In handwriting identification system, in first step we need provide a handwriting databaxse, then we should devide this databse in two group, train and test group. In this research training group inclusive 80% of whol databaxse.Writers of training databaxse are known.
In the next srep we should specify features for charactrize likelihood bitween handwritings. In this research we use morphology features such as grayscale morphology.
We use ANN, RBF, SVM and KNN classifier for identification .We provide a databaxse that contain 110 handwriting samples from 100 writer that 99 sample are fake and 11 samples are origin.
In this research we could got 99% success to identification.
Keywords:
#Handwritting #pattern recognition #artificial network #support vector machine #fiction Link
Keeping place: Central Library of Shahrood University
Visitor:
Keeping place: Central Library of Shahrood University
Visitor: