Open AccessHANDWRITING IDENTIFICATION USING ANFIS
Author(s) -
T. Meera Devi
Publication year - 2014
Publication title -
journal of social sciences research
Language(s) - English
Resource type - Journals
ISSN - 2321-1091
DOI - 10.24297/jssr.v3i2.3234
Subject(s) - handwriting , adaptive neuro fuzzy inference system , column (typography) , artificial neural network , character (mathematics) , computer science , identification (biology) , pattern recognition (psychology) , artificial intelligence , function (biology) , image (mathematics) , paragraph , word (group theory) , fuzzy logic , mathematics , fuzzy control system , frame (networking) , telecommunications , botany , geometry , evolutionary biology , world wide web , biology
A new method for handwriting identification was presented.Individual characters was separated from a word choosed from a paragraph of handwritten text image which is given as input to the system. Then each of the separated characters are converted into column vectors of 625 values that are later fed into the adaptive neural fuzzy inference system(ANFIS), which was calculate membership function(MF) and normalized firing strength.In our paper we were used triangular membership function and compare with others MF.The networks has been designed with single layered neural network corresponding to a character from a-z, the outputs of all the column vector is fed into network the which has been developed using the concepts of correlation, with the help of this the overall network is optimized with the help of column vector thus providing us with recognized outputs with great efficiency.