Sign Language Recognition for Bengali Characters

Sign Language Recognition for Bengali Characters

Tanzila Ferdous Ayshee, Sadia Afrin Raka, Quazi Ridwan Hasib, Rashedur M. Rahman, Md. Hossain
Copyright: © 2015 |Pages: 14
DOI: 10.4018/IJFSA.2015100101
(Individual Articles)
No Current Special Offers


Sign language is the primary means of communication for people having speaking and hearing impairment. This language uses a system of manual, facial, and other body movements as the means of communication, as opposed to acoustically conveyed sound patterns. This paper uses image processing and fuzzy logic to develop an intelligent system to recognize Bengali Sign Language. The proposed system works in two phases. In the first phase, the fuzzification methods are defined. Then in the next phase, the raw images are processed to identify the fuzzy rules. A detailed implementation procedure of the proposed system is demonstrated by describing the recognition process of four Bengali characters.
Article Preview

1. Introduction

People having speaking and hearing impairment are usually deprived of normal communication with other people in the society. Since they cannot talk or hear like normal people, they have to depend on some sort of visual communication in most of the time. Sign Language is the primary means of communication among the people having speaking and hearing impairment. It is a visual language consisting of various signs, gestures, finger spelling and facial expressions. A common misconception about sign language is that all sign languages are the same worldwide (Sign Language, Wikipedia, 2015). Hundreds of sign languages are actually used worldwide and each country generally has one or more native sign languages (Sign Language, Wikipedia, 2015). The most widely used sign language is probably American Sign Language (ASL), which is used in the USA, Canada, parts of Mexico and, with modifications, in quite a few other countries in Central America, Asia and Africa (American Sign Language, 2015). There are many other sign languages such as:

  • British Sign Language (BSL)

  • Brazilian Sign Language (LIBRAS)

  • Australian Sign Language (Auslan)

  • French Sign Language (LSF)

  • Bengali Sign Language (BaSL)

As like any other language, sign language has also got grammar and vocabulary but uses visual modality for exchanging information (Armstrong et al., 1995) These languages are as rich and complex as any spoken language, despite the common misconception that they are not “real languages” (Sign Language, Wikipedia, 2015). But normal people are usually unaware of these grammars. So when mute or hard of hearing people try to express themselves to other people with the help of these sign language grammars, most normal people do not understand those. As a result it has been seen that communication of a person having speaking or hearing impairment are only limited within his/her family.

At this age of technology, the demand for a computer based system targeting this problem is highly demanding. However, researchers have been attacking the problem for quite some time now and the results are quite promising.

In Bangladesh, about 16 million people are living with a disability which includes speaking and hearing impairment (Center for Disability Management, 2015). Though there are some interesting technologies for speech recognition, however, not a large number of researches have been carried out in sign language recognition, especially in Bengali Sign Language recognition. Hand gestures for some Bengali characters are given in Fig. 1. Our objective is to develop an intelligent system using image processing and fuzzy feature based rule which can act as an interpreter between the Bengali sign language and the spoken language dynamically.

Figure 1.

Some BaSL single hand gestures


Complete Article List

Search this Journal:
Volume 13: 1 Issue (2024)
Volume 12: 1 Issue (2023)
Volume 11: 4 Issues (2022)
Volume 10: 4 Issues (2021)
Volume 9: 4 Issues (2020)
Volume 8: 4 Issues (2019)
Volume 7: 4 Issues (2018)
Volume 6: 4 Issues (2017)
Volume 5: 4 Issues (2016)
Volume 4: 4 Issues (2015)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing