Fuzzy Features Extraction from Bangla Handwriten Character

Abstract

Character is the fundamental attribute for writing and reading a language. Character recognition is the process to classify the input character according to the predefined character class. Fuzzy logic has proved to be a powerful tool to represent imprecise and irregular patterns. The most tedious job associated with fuzzy logic is the extraction of meaningful features. This paper describes a method to efficiently detect the meaningful fuzzy features including global features, geometric features and positional features from handwritten Bangla character respectively. We have tested our design for different types of Bangla handwritten alphabets in various style and we got successful features extraction results for most of the test cases.