Contents 1. Introduction
Recognition of handwritten characters has been a popular research area for many years because of its various application potentials. Some of its potential application areas are postal automation, bank cheque processing, automatic data entry, etc. Various approaches have been proposed by the researchers towards handwritten character recognition and many recognition systems for isolated handwritten numerals/characters in languages like English, Chinese, Japanese, Indian etc. are available in the literature [1]–[4]. Although high accuracy is obtained from some of the systems, it may be noted that most of the errors are due to similar shaped handwritten characters. Recognition of these similar shaped characters is one of the difficult problems and in this paper we proposed a novel feature extraction technique to improve the recognition results of two similar shaped characters. The technique is based on F-ratio, a statistical measure that is defined by the ratio to the between-class variance and within-class variance. F-ratio is calculated from feature vectors belong to the similar shaped character classes and enhanced the feature vector for better recognition. F-ratio modifies the feature vector of two similar shape characters by enhancing the feature elements that belongs to the distinguishable portions of the similar shaped characters and reducing the feature elements of the common portion of the characters, so that these similar shaped characters can be identified easily. This is done by weighting the feature elements. To get the idea of some similar shaped characters of different scripts considered, we provided some of their similar shape printed characters in Fig. 1. Although from these similar shape printed characters we can find some small differences but sometimes it is very difficult to get any difference because of writing style of different individuals.
