IT PhD Thesis Title – Graphic Character Recognition (GCR)

Information Technology PhD Thesis Topic: This is one of the PhD topic on IT that gives an idea of how to synthesise the information on IT related phD Thesis. One of the best IT PhD Thesis Title example of writing an overview.

PhD Title:   Graphic Character Recognition (GCR)

Document: This document provides a summary of the idea behind the project (GCR) and specifies the approach that will be take for this project.

Project Summary:

1.    The Optical Character Reader (OCR) software has become very prevalent in recent years for converting character images into a textual form. Examples of its implementation would include:
a.    Book scanning and converting into eBooks
b.    Writing recognition
c.    Automatic number plate recognition etc.
For most of its history, OCR software relied on algorithms that did something called matrix matching and used odd parity rules for character recognition, which had drawbacks like limited font support, complex font sizing and problems with matching ambiguous & characters with missing or distorted parts.

2.    The Project “Graphic Character Recognition” will consist of two sections:

a.    Project Report:
i.    To start with the project report will study the initial versions of OCR’s and sketch the flow of algorithms used. This will also bring out the possible problems with the initial models.
ii.    Then we will take a look at the current advancements on the OCR technology and its implications. We would also cover the scope and usability of the OCR software thus bringing out the importance of research in this field.
iii.    The most critical phase of the project will be to research upon and create a new innovative algorithm for character recognition named “Graphic Character Recognition (GCR)”.
•    This algorithm would aim towards providing a faster and efficient processing and would not only be font independent but also font size independent.
•    This algorithm will also cater for the ambiguous or characters with missing or distorted parts.
•    This algorithm will be an intelligent learning system which would allow the user to train it for reading new fonts or recognizing characters of other languages.

b.    GCR Algorithm – POC (proof of concept):

The second section of the project will consist of a working Proof of Concept for the new GCR algorithm.
This software implementation will aim towards demonstrating each step of processing of the developed GCR algorithm:
1.    Reading incoming stream of image characters.
2.    Preprocessing and executing the new GCR algorithm to recognize the characters.
3.    Performing intelligent learning in case of new fonts and using the new knowledge base for recognizing similar characters thereafter.
4.    Post processing and providing the output stream in form of text characters.
Please note that this software implementation of the GCR system will not be a final implementation, but will only be a simplified version aiming to display the proof of concept of the GCR algorithm. As we would have to work with closely with graphic, we would be using C++ language to build the working POC.
However later if someone wants to develop a better efficient complete implementation of the system, they can take this POC as a starting point and develop a product utilizing the GCR algorithm.