Multi-Script Handwriting Recognition with FOHDEL*
نویسندگان
چکیده
Until now handwritten character recognition systems used script specific methodologies. In this paper we present a unified syntactic approach for multi-script recognition. The fuzzy pattern description language -FOHDELis used to store fuzzy features in the form of fuzzy rules. First we briefly describe the proposed recognition methodology for Latin, Devanagari and Kanji scripts by analyzing their characteristic properties. Further we present the main features of FOHDEL by a comparative rule generation of three scripts under experiment. Finally the system integration of the proposed multi-script recognition scheme in the existing Latin recognition system is presented.
منابع مشابه
Fuzzy handwriting description language: : FOHDEL
In this paper we present a new fuzzy language } FOHDEL } for the syntactic description of handwritten symbols. The presented language incorporates the fuzzy logic techniques to describe the syntactic relations of the semantic features extracted from a symbol pattern. A FOHDEL rule-base represents the compact feature information extracted from a small number of character prototypes and covers va...
متن کاملA New Fuzzy Language for Online Handwriting Recognition
In this paper we present a new fuzzy language for the syntactic description of online handwritten symbols. The proposed language FOHDEL can be extended to describe other image patterns as well. The presented language incorporates the fuzzy logic techniques to describe the syntactic relations of the semantic features extracted from a symbol pattern. The basic idea behind FOHDEL is to have a lang...
متن کاملA Fuzzy Syntactic Method for On-Line Handwriting Recognition
Fuzzy logic has proved to be a powerful tool to represent imprecise and irregular patterns. Thus the representation of handwriting parameters is accomplished with fuzzy linguistic modeling. The development of a multi-layered rule based handwriting recognition system is described. The selection of the representative features, which describe the shapes and location of segments, is the core of the...
متن کاملOff-line Arabic Handwritten Recognition Using a Novel Hybrid HMM-DNN Model
In order to facilitate the entry of data into the computer and its digitalization, automatic recognition of printed texts and manuscripts is one of the considerable aid to many applications. Research on automatic document recognition started decades ago with the recognition of isolated digits and letters, and today, due to advancements in machine learning methods, efforts are being made to iden...
متن کاملA Fuzzy Online Handwriting REcognition System : FOHRES
In this paper we present a new fuzzy online recognition system (FOHRES) for handwritten characters. Our new approach uses the advantages of fuzzy set theory for the classification of handwritten characters at different processing stages. The proposed fuzzy online handwriting description language (FOHDEL) provide a flexible multiwriter environment which is extendable to various tasks like recogn...
متن کامل