Chapter 1
Chapter 1
Chapter 1
INTRODUCTION
1.1 PROBLEM DEFINITION & RELEVENT THEORY:Optical character recognition, usually abbreviated to OCR, is the mechanical or electronic translation of images of handwritten or typewritten text (usually captured by a scanner) into machine-editable text. OCR is a field of research in pattern recognition, artificial intelligence and machine vision. hough academic research in the field continues, the focus on OCR has shifted to implementation of proven techni!ues. Optical character recognition (using optical techni!ues such as mirrors and lenses) and digital character recognition (using scanners and computer algorithms) were originally considered separate fields. "ecause very few applications survive that use true optical techni!ues, the OCR term has now been broadened to include digital image processing as well. #arly systems re!uired training (the provision of $nown samples of each character) to read a specific font. %&ntelligent% systems with a high degree of recognition accuracy for most fonts are now common. 'ome systems are even capable of reproducing formatted output that closely approximates the original scanned page including images, columns and other non-textual components. (owever such techni!ues are limited to static scanned images, and not real time ac!uisi)ed handwritten characters. *ith advent of digital input devices li$e 'tylus, etc. the user can enter handwritten text as input to the OCR system. he system will have the capability of recogni)ing and updating the entered characters on-the-fly. his enables the user to rectify the errors at the time of entering the text and not when he+s done entering all the characters. he system will also host a universal language recognition capability. his can be achieved using pattern rec !n"t" n and #eat$re e%tract" n. ,attern recognition is a subtopic of machine learning. &t can be defined as %the act of ta$ing in raw data and ta$ing an action based on the category of the data%. -ost research in pattern recognition is about methods for supervised learning and unsupervised learning. ,attern recognition aims to classify data (patterns) based on either a priori $nowledge or on statistical information extracted from the patterns. he patterns to be classified are usually groups of measurements or observations, defining points in an appropriate multidimensional space.
VACOE, AHMEDNAGAR
Feat$re E%tract" n &n pattern recognition and in image processing, /eature extraction is a special form of dimensionality reduction. *hen the input data to an algorithm is too large to be processed and it is suspected to be notoriously redundant (much data, but not much information) then the input data will be transformed into a reduced representation set of features (also named features vector). ransforming the input data into the set of features is called features extraction. &f the features extracted are carefully chosen it is expected that the features set will extract the relevant information from the input data in order to perform the desired tas$ using this reduced representation instead of the full si)e input. he system also employs some of the best image processing algorithms for the input to be suitable for feature extraction and pattern recognition. /ollowing are some of the image processing algorithms that will be used0 o o o o o 1rayscale Conversion hresholding #rosion 2 3ilation 'harpen 2 "lur &mage 'egmentation 4
VACOE, AHMEDNAGAR
VACOE, AHMEDNAGAR
1.& 'COPE:he proposed OCR system covers the range of topics, many shared with mainstream image processing. 'ome of the most important are0o 6nderstanding image user+s needs and information see$ing behavior. o &dentification of suitable ways of describing image content. o #fficient any number of feature extraction techni!ues. o ,roviding usable human interface to OCR within the vicinity of system.
1.( OB)ECTIVE:o o o o o o support more than one languages. o reduce total manpower re!uired for manual data entry wor$. o improve the !uality of the degraded data. o understand the different users handwritten scripts. o introduce the dynamic OCR system to the users.