2026-06-26T06:05:28Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/4867 2021-12-13T02:34:36Z 1582963302567:1597824273898 user-ucsy

Printed Character Recognition based on Shape Descriptors with Competitive Neural Trees Thein, Yadana Htike, Theingi CompetitiveNeural Trees (CNeT) are widely used forclassification in pattern recognition. This paper appliesthis technique for recognizing of Myanmar printedcharacter s . This paper involves three important steps,typically preprocessing, feature extraction andclassification. The aim of prepro cessing is to improvethe quality of the images for further processing . Afterpre processing, features for each character image areextracted based on Shape description. Shape is animportant visual feature and it is one of the basicfeatures used to descr ibe image content. These featurevector from word images which are used in CompetitiveNeural Trees (CNeT) for recognition purpose. Thispaper introduces a global search method for the CNeT,which is utilized for training. 2013-02-26 http://hdl.handle.net/20.500.12678/0000004867 https://meral.edu.mm/records/4867