Characters Recognition based on Geometrical Features

Abstract

Character recognition is one of the important subjects in the field of Document Analysis and Recognition (DAR). The general objective of DAR research is to fully automate the process of entering and understanding printed or handwritten data into the computer. The proposed pattern recognition system consists of two-stage process. The first stage is feature extraction and the second stage is classification. Feature extraction is the measurement on a population of entities that will be used in recognition process. This assists the recognition stage by looking for features that allows fairly easy to distinguish between the different classes. Several different features have been used for recognition process. The set of proposed features that are used makes up a feature vector. These set of features are: the first feature is represented the number of character pixels (the summation of pixels), the second features is represented the width of each character in pixels, and the third feature represented the height of each character in pixels. Finally, Pattern recognition system classifies each member of the population on the basis of information contained in the feature vector. The results show that the suggested features gives higher accuracy in text and character recognition.