Multilevel Classifier in Recognition of Handwritten Arabic Characters

Abstract

Problem statement: Arabic offline handwriting recognition is considered one of the most challenging topics. This is probably caused by the fact that Arabic recognition system faced many problems during the development stage. It faces the usual problems of character recognition in general, in addition to the problems that are specific to Arabic language only. The aim of this study was to build a classifier to solve Arabic text ambiguity; to be used in text recognition applications. Approach: A multilevel classifier based on pattern recognition techniques, is proposed. The proposed classifier was implemented using MATLAB and also tested with a large sample of handwritten datasets. Results: Pattern recognition techniques are used to identify Arabic handwritten text. After testing, the recognition rates reached {93, 84, 89 and 85%} for the isolated letters, letters at the beginning, at the middle and at the end of the word respectively. Conclusion: Even that the Arabic letters change their shape depending on their position in a word, the proposed classifier, using the powerful set of features, is proved to be effective in the recognition of Arabic letters.

A Survey on Arabic Handwritten Character Recognition

“…The technique of thinning will convert the binary image of character into one pixel thick image used by [4]. Thresholding in this technique includes converting gray input images into document images of binary type for getting good features from the image [5], skew detection and correction [6], noise removal; the basic noise objective removal which is to eliminate all bit-patterns which are unwanted those did not have any value in the result. The technique of median filtering utilized in [3] for the purpose of remove noise, horizontal projection profile in [7], filterization in [8] and smoothing in [9][10][11].…”

Arabic Character Recognition based on Statistical Features - A Comparative Study

“…Character recognition is an attractive subject in the field of pattern recognition. Unlike English language, there has been only a few works on Arabic characters recognition (Abandah et al, 2009); (Abdul Sattar and Shah, 2012); (Imran et al, 2012); (Jenabzadeh et al, 2011); (Zaghloul et al, 2011). However, Arabic texts have main specifications which make them difficult to segment and to recognize.…”

Offline Recognition of Handwritten Urdu Characters using B Spline Curves: A Survey

“…The thinning technique reduces thickness of the binary image of the character to one pixel as applied by Hanan et al [3] in their research.…”

“…Thresholding: This technique cited by [2,3] is used to transform an input image from gray scale to binary form to make it suitable for extracting most appropriate features.…”