Technical Methods and Algorithms for Developing Efficient Optical Character Recognition System: An Overview
Journal Title: Annals. Computer Science Series - Year 2018, Vol 16, Issue 2
Abstract
Machine recognition problem of printed documents in Optical Character Recognition has been the target of research in the area of pattern recognition. Study in this aspect has been controlled by a need to join the natural process of image input with the data processing abilities of computer system. In this regard, engineers and scientists are not able to make judicious use of OCR systems for their technical work because they lack effective algorithms for interpretation of complex expressions even though it is important part of a Human Computer Interaction system. However, OCR in a nut shell is the electronic change of printed or handwritten text into images using machine encoding scheme representation like ASCII or Unicode. Hence, various methods and algorithms have been demonstrated in the study to increase the efficiency of system to effectively solve OCR problems. The study shows the concept of various OCR algorithms such as support vector machine, decision tree classifier, statistical, structural, artificial neural networks and template matching algorithms.
Authors and Affiliations
Yakubu Aliyu IBRAHIM, Tunji Samuel IBIYEMI
Keywords
Related Articles
- EP ID EP550041
- DOI -
- Views 100
- Downloads 0
Handling Multicollinearity; A Comparative Study Of The Prediction Performance Of Some Methods Based On Some Probabiltiy Distributions
This study used some probability distribution (Gamma, Beta and Chi-square distributions) to assess the performance of partial least square regression (PLSR), ridge regression (RR) and LASSO regression (LR) methods. Ordin...
A perspective of metacognition in solving math problems in Vietnam secondary schools
In an effort to innovate teaching and learning process to prepare for a new generation for the demands of the new era, many educators have discovered the value of metacognition. Students will need both mathematical skill...
Development of an Enhanced AODV Energy Management model and Link Stability in MANET
A mobile ad hoc network (MANET) nodes move arbitrarily and as a result the networks experience a rapid and unpredictable topology changes. The mobile nodes can receive and forward packets as router which leads to superfl...
Developing Predictive Models using Typical Machine Learning and Computational Techniques
This study investigates the accuracy of developing predictive models using machine learning techniques. The machine learning techniques considered in this study include artificial neural network (ANN) and Kalman filter a...
Modeling a geometric locus problem with GeoGebra
În rezolvarea unor probleme de loc geometric analiza, modelarea, studiul cazurilor particulare ce ajută la intuirea soluţiei poate atrage din partea rezolvitorului un efort de lungă durata. În acest sens, folosirea unor...