DESIGN OF AN AUTOMATIC CHARACTER RECOGNITION SYSTEM FOR PITMAN SHORTHAND

ABSTRACT

This study was set out to design an automatic character recognition system for pitman shorthand. The experimental research design was used and the population used was a total of 97 students of the Information Management and communication department of Higher Teachers Technical Training College Kumba and using the Kycee and Morgan table the sample size of the work was 28. Data was collected mainly through primary sources of data using an interview guide and the Document review guide. This study follows the qualitative approach meaning the data collected would not require numerical analysis. The system modeling approach used in this study is Unified Modelling Language (UML). The method of system implementation adopted for this study is the Object-Oriented Programming (OOP) for the programming of the application to generate Pitman Shorthand characters and English characters during drilling and transcription respectively. The results of the design is a web application built in PHP with a database in MySQL. The system is generally made to transcribe Pitman Shorthand characters into English characters as well as drilling English characters into Pitman Shorthand characters. Based on the result of the study, it is concluded that there is significant feasibility in implementing a centralized system for the processing of Pitman Shorthand outlines into English or English letters to Pitman Shorthand outlines. It is recommended to secretaries that they should adopt this new technique of rapid note-taking to enable them to take accurate notes and make rapid transcriptions and drilling.

CHAPTER ONE

GENERAL INTRODUCTION

Introduction

In this chapter the work would be introduced by the background to the study, the statement of the problem, the objectives and research questions, the significance of the study, scope and operational definition of terms

 Background of the Study

The  earliest  identified  examples  of  shorthand,  or  speed  writing,  date  back  about  2,500  years  to  the  Greek  and  Roman civilizations and the Dynasties of Imperial China. The evolution of modern shorthand for English scripts began in the 16th Century.  An  early  example  is  Characterie;  An  Arte  of  Shorte,  Swifte  and  Secrete  Writing  by  Character published by Timothy Bright in 1588 which introduced a system with 500 arbitrary signs resembling  words;  John  Willis’s  Art  of  Stenography  appeared  in  1602,  and  Thomas  Shelton’s  Short  Writing  appeared  in  1626 (a  system  which,  incidentally,  was  used  at  times  by  Sir  Isaac  Newton).  Later,  in  1720,  John  Byrom  introduced  the  New  Universal  Shorthand,  which  was  based  on  geometric  shapes,  followed  soon  afterwards,  in  1786, by a similar system proposed by Samuel Taylor, which became  widely used at the time. Taylor’s system  was  subsequently  super ceded  by  Pitman  shorthand  which  was  introduced  in  1837  by  Isaac  Pitman. 

Pitman  shorthand, which is based on simple geometric symbols representing the consonant and vowel sounds of speech (Isaac, 2003), became the predominant shorthand system in the English speaking world (with  the exception of the USA) right up to the present day. In the USA the Gregg system, introduced by John Gregg in 1888 and based on cursive symbols (Greeg, 1955), first proposed by Franz Gabelsberger in 1834, has predominated. 

By the 1960’s shorthand was considered an essential part of secretarial training as well as being useful for journalists.  Many colleges offered training courses and speed testing for intending secretaries.  In  the  1980’s  it  was conservatively  estimated  that  there were  more  than  one  million  proficient  users  of Pitman shorthand and a  similar number of Gregg shorthand writers. However, since the 1970’s office technology has radically changed.

The  introduction  of  dictation  machines  reduced  the  importance  of  shorthand  skills  and  subsequently,  the  widespread  adoption  of  word  processing  applications  on  mobile  computing  devices  such  as  Personal Digital Assistants (PDAs) or Palm tops and  laptop  computers  has  also  reduced  the  usage  of  dictation  devices.  Only  in  specific  areas  such  as  court  and  news reporting, or the Personal Assistant (human version) to Senior Executives are handwritten shorthand skills still in  active  use.  While  shorthand  was  originally  intended  as  a  means  of  fast  note-taking,  and  the  reader  of  the  shorthand would usually be the original writer, it has been shown that Pitman shorthand has many features of a  machinography – a machine compatible handwritten script, whilst Gregg shorthand is based on curves which are less compatible with current computer recognition techniques (Brooks, 1985).   Pen-computing has been the unrealized dream since the 1950’s when the earliest (non-portable) computers were beginning to appear.  With  the  recent  rapid  advances  in  handheld  devices  such  as  PDAs,  Palm Organizers  and  the  Tablet  PC,  entering  text  information  into  such  devices  remains  a  serious  bottleneck.

Compared with other slow speed approaches such as the QWERTY keyboard (50-60 wpm), printed script (20-25 wpm) and cursive script (35 wpm with limited recognition performance), Pitman shorthand writers can readily achieve a recording speed of over 120 wpm. Currently, Pitman shorthand has been widely taught and used in 75 countries throughout the world.  Over  two  decades  ago,  intensive  research  was  carried  out  on  the  potential  of Pitman shorthand as a means of rapid text entry to computer (Leedham N. , 1989).  At  that  time,  there  was  no  suitable  pen-based  device  on  which  to  exploit  the approach. In the last ten  years, a research group at Mysore University (Nagabhushan & Anami, A knowledge based approach for composing English text from phonetic text documented through Pitman shorthand languag, 1999) has  been  working  on  automation  text  production  from  offline  Pitman  notes.  From  their  point  of  view,  Pitman shorthand  has  advantages  in  being  integrated  into  the  present  speech  processing  systems  as  it  is  the  only universally accepted medium known for enabling real-time speech to text production. With  the  recent  widespread  availability  of  inexpensive  portable  hand-held  devices,  it  is  now  time  to  re-assess  the  potential  of  Pitman  shorthand  as  a  viable  means  of  rapid  handwritten  text  entry  in  mobile  computer applications.  Some  promising  algorithms  have  been  arising  recently  for  the  recognition  and  transcription  of Pitman shorthand (Htwe, Higgins, & Leedham, 2004). Quite significantly, the Pitman shorthand structure is also applicable to Mandarin Chinese where the phonetic sounds of Chinese are mapped to the geometric strokes of  the  Pitman-style  shorthand (Ma & Leedham, 2007)  in  a  shorthand  system  called  Renqun (Liao, 1985).   If  the  accurate  recognition  of  Pitman  shorthand  is  achievable,  the  techniques  are  equally  applicable  to Mandarin Chinese (with a different transcription system) and thus the recognition engine is potentially usable by two of the world’s most widely spoken languages – English and Mandarin Chinese. The potential users of such a system would  be  anyone  who needs to  make  notes  or record  verbatim  speech  in  real-time as it  is  spoken.  This would include news reporters, students and secretaries.  It  is  estimated  that  about  750  million  people  speak English  and  1000  million  speak  Mandarin.  If  only  1  in  10,000  or  0.01%  of  these  people  found  such  an application useful there would be over 1.5 million users.

Character recognition system has received considerable attention in recent years due to the tremendous need for digitization of printed documents. The textual representations of images convey information relating to what is actually depicted in the image as well as what the image is about. Manual assignment of text data from images is time consuming and costly. Hence automation of text extraction from images is a challenging area in image processing due to its potential applications (Sadasivan, 2012).

Optical Character Recognition (OCR) traces its roots back to telegraphy. On the eve of the First World War, physicist Emanuel Goldberg invented a machine that could read characters and convert them into telegraph code. In the 1920s, he went a step further and created the first electronic document retrieval system.

At this time, businesses were microfilming financial records – great in principle, but quickly retrieving specific records from spools of film was nigh on impossible. To overcome this, Goldberg used a photoelectric cell to do pattern recognition with the help of a movie projector. By repurposing existing technologies, he took the first steps towards the automation of record keeping. The US patent for his “Statistical Machine” was later acquired by IBM.

Since then, OCR technology has proliferated, with businesses all over the world relying on it to help reduce overheads when it comes to converting extracting data from paper documents (Steve, 2019).

This Thesis would be focused on the designing an Automatic Character recognition System for the Pitman Shorthand.

 Statement of the Problem                                              

Shorthand in the corporate and educational world today is basically English. Employees and employers, teachers and even students in the country have had to endure lots of problems in the writing of shorthand due largely to a number of problems which can be noted and seen especially in schools such as the Higher Technical Teachers Training College in Kumba. Some of these include fear of strokes and even most a times poor materials while trying to learn the Pitman shorthand. Also with advancement in technology with devices such as the dictating machine and Google translate, pitman shorthand is gradually losing its essence and use in modern day world. In a bid to match up with technology and not to make extinct this special form of writing, the researcher saw the need to design an automatic character recognition system for pitman shorthand. The problem of shorthand is not hinged on the subject proper. Findings from pre-literature investigation revealed that most people and students come in with already ill-conceived belief that shorthand is a difficult subject; every subject is difficult, and there are experts in all subject. This has been identified as contributory toward a lot of people shying away from the subject Shorthand: Shorthand like mathematics is expected to have its peculiar methods and problems among others. The investigation revealed that people’s ill-conceived notion that shorthand is a difficult field inflicted a lot of psychological damage to their understanding of shorthand. It again revealed that it gave rise to many needs and imagined problems which hinder the much desired progress in shorthand. Also it revealed that some persons are interested in the art but when they encounter minor problems, their interest fizzles out, the high rate of shying away from shorthand is also due largely to the fact that People treat shorthand like any other subject.

It is because of the above challenges that the main focus of this dissertation is to design a computer based system or application that would be able to convert pitman shorthand strokes into readable English language and it is hoped that upon completion, this would go a long way to solve the fear people encounter in trying to learn shorthand or even using it in the cooperate world as freely as they would like to and match this special speed way of taking notes with modern technology.

Objectives of the Study

The objectives of the study are divided into General and Specific objectives

 General Objectives

The main objective of this study is to design an automatic character recognition system for Pitman Shorthand.

 Specific objectives

1. To develop a system that embodies the outlines of the Pitman Shorthand.

2. To design the architecture of the Automatic Character Recognition System for Pitman Shorthand.

3. To design the Automatic Character Recognition System for outline conversion in Pitman Shorthand

Research Questions

 General Research Question

How can an automatic character recognition system for Pitman Shorthand be designed?

 Specific Research Questions

1. How can a System that embodies the outlines of pitman shorthand be developed?
2. What is the architecture of the Automatic Character Recognition for Pitman Shorthand?
3. How would the Automatic Character Recognition System convert the Pitman Shorthand Strokes?