HANDWRITTEN ENGLISH WORD RECOGNITION USING HMM, BAUM-WELCH AND GENETIC ALGORITHM
Journal: International Journal of Computer Engineering and Technology (IJCET) (Vol.9, No. 4)Publication Date: 2018-12-27
Authors : ATMA PRAKASH SINGH RAVINDRA NATH SANTOSH KUMAR;
Page : 176-186
Keywords : HMM (Hidden Markov Model); GA (Genetic Algorithm); Baum Welch Method (BWM); Handwritten Character Recognition (HC).;
Abstract
We face one problem in in the field of image processing and pattern recognition of computer science, a challenge of correct detection or recognition of the handwritten word. Word recognition talk about the identification of the word which is written by a human being. Most of the researchers are trying to solve recognition problems. We also try to give one solution in this paper to handwritten word recognition, we consider English alphabets of both type and also consider numeric numbers as (A-Z, a-z or 0-9). In this paper we use three approaches Hidden Markov Model (HMM), Baum-Welch and Genetic Algorithm (GA) to identify features of each character and compare with its testing set of characters. we also use stages of handwritten word recognition system that are: read a scanned image of hand written word as “HELLO CAN”. We take “CAN” word image, now we converting this CAN word image into binary matrix form (0 and 1), resizing each character of word into the character binary matrix into size of (n x m where n and m may be same), and thinning of an image to get a clear skeleton of each character. Then in this paper identify each character using three algorithms namely: Forward Algorithm, Baum Welch and Genetic Algorithm. The results obtained from each of the algorithm are compared separately and at the end the accuracy of these algorithms are compared separately.
Other Latest Articles
- V-BIBE: A NEW IDENTITY BASED ENCRYPTION SCHEME
- EVALUATION OF NTH BIT REPLACEMENT IN AUDIO STEGANOGRAPHY
- AN EFFICIENT APPROACH OF SOFTWARE COST ESTIMATION USING SOFT COMPUTING TECHNIQUES
- AUTOMATIC IDENTIFICATION OF NEWS TWEETS ON TWITTER
- COMPARATIVE ANALYSIS TO CONQUER SEMANTIC INTERPRETABILITY IN SEMANTIC WEB MINING
Last modified: 2018-12-08 16:25:44