Open AccessTwo-Dimensional Optical Character Recognition of Mouse Drawn in Turkish Capital Letters Using Multi-Layer Perceptron Classification
Author(s) -
Ehsan Ali Al-Zubaidi,
Maad M. Mijwil,
Aysar Sh. Alsaadi
Publication year - 2019
Publication title -
xi'nan jiaotong daxue xuebao
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.308
H-Index - 21
ISSN - 0258-2724
DOI - 10.35741/issn.0258-2724.54.4.4
Subject(s) - optical character recognition , computer science , character (mathematics) , artificial intelligence , turkish , perceptron , reading (process) , speech recognition , software , intelligent character recognition , natural language processing , pattern recognition (psychology) , character encoding , transformation (genetics) , character recognition , image (mathematics) , artificial neural network , programming language , linguistics , philosophy , geometry , mathematics , biochemistry , chemistry , gene
The Optical Character Recognition (OCR) is software for text recognition that takes an image containing text, to transform it into strings, then save them into a format that make it able to use in text editing programs. The OCR plays a significant role in the transformation of printed materials into digital text files. These digital files can be very useful for children and adults who have awkward reading. This is because a digital text can be used with computer programs that allow people to read them in different ways. In this paper, we developed system for Turkish character recognition under visual studio (C#) development environment, where machine learning is used to accurately predict optical characters, the reason why it has a high precision and high recognition speed through deep learning, which is one of the machine learning methods for OCR when drawing letters by mouse on the screen, then recognize by using back propagation algorithm.