Design And Implementation Of Gabor Filter Based Yoruba Handwriting Recognition System

Introduction

1.1 Background of the Study

Character is the basic building block of any language which is used to develop different language structures. Characters are alphabets and the structures developed are the words, strings, sentences, paragraphs and so on (Le Cunet al., 1990). Character recognition also known as optical character recognition is the recognition of optically processed characters. The purpose of character recognition is to interpret input as a sequence of characters from an already existing set of characters (Kader and Deb, 2012).

Handwritten character recognition is the process of converting handwritten text into a form that can be read by the computer, the major problem in handwritten character recognition system is the variation of the handwriting styles of individuals, which can be completely different for different writers (Patel and Thakkar, 2015).

Handwritten character recognition system can be divided into two categories namely the online character recognition and the offline character recognition.

Online character recognition is the conversion of text written on a digitizer or PDA automatically where the sensor picks up the pen – tip movements and the pen-up/pen-down switching. The signal obtained from the pen – tip movements is converted into letter codes that can be used by the system and text processing applications. In offline character recognition, the image of the written text is scanned and sensed offline by optical scanning (optical character recognition) or intelligent character recognition (Tawde and Kundargi, 2013).

Yorùbá (natively èdè Yorùbá) is a Niger–Congo language spoken in West Africa. The number of speakers of Yoruba was estimated at around 20 million in the 1990s.The native tongue of the Yoruba people is spoken principally in Nigeria and Benin, with communities in other parts of Africa, Europe and the Americas. A variety of the language, Lucumi, is the liturgical language of religion of the Caribbean. Yoruba the Santería is most closely related to the Owo and Itsekiri languages (spoken in the Niger Delta) and to Igala (spoken in central Nigeria).

There has been extensive work in the literature regarding features extraction approaches in the off-line Arabic handwriting recognition. Many of these methods require high quality binarization of the document images which is difficult due to varying characteristics of noisy artifacts common in such documents. In addition, large amount of gray-level information is lost during binarization. Therefore, features that are extracted from the original gray-level images should be useful to discriminate handwritten character shapes (Jin Chen et al, 2017).

Gabor filters, which operate directly on gray-level images, have several advantages. First, Gabor features have been used for capturing local information in both spatial and frequency domains from images, as opposed to other global techniques such as Fourier Transforms. Second, Gabor filters are orientation specific. This property allows us to analyze stroke directions in the handwriting. Third, the filtering output is robust to various noises since Gabor filters use information from all pixels in the kernel (Jin C. et al, 2017).This research works tends to use Gabor features extraction techniques on Yoruba characters.

1.2 Statement of Problem

Many feature extraction approaches for offline handwriting recognition (OHR) rely on accurate binarization of gray level Images, However, high-quality binarization of most real-world documents is extremely difficult due to varying characteristics of noises artifacts common in such documents, hence the research work used consider Gabor features for off-line Yorùbá handwritten images.

1.3 Aim and Objectives

The aim of the project is to design and implementation of Gabor Filter based offline Yorùbá handwritten recognition system.

The objectives of this project are to:

Review the existing literature on Yoruba handwriting recognition system

Design a Gabor filter based Yoruba Handwriting Recognition system Implementing the designed system in “2”.

1.4 Proposed Research Methodology

Gathering of books and online PDF books on related work of handwriting recognition system.

Designing the system using Object Oriented Approach.

Implementing the designed system using C++ programming Language

1.5 Justification

The Yoruba handwriting recognition system is one of most important software in university for both students and instructors because of the development in technology.

1.6 Scope of the Study

This research work covers the performance of sub word recognition for off-line Yorùbá handwritten images. We will also compare the recognition performance with other binarization based features which have been proven to be effective in capturing shape characteristics of handwritten Arabic sub words, such as GSC (a set of gradient, structure, and concavity features) and skeleton based Graph features.

1.7 Chapter Layout

The chapter one consists of the introduction that is background of the study, statement of the problem, aim and objective of the study, proposed methodology, research justification and finally the chapter layout. Chapter two consists of review of related concept and related works. Chapter three consists of research methodology. Chapter four consists of result discussion and finally, chapter five consists of summary, conclusion and recommendation.