Essay On Sign Language

Sign language is a language for the people who are challenging in hearing and listening. Despite common misconceptions, sign languages are complete natural languages, with their own syntax and grammar.

1.1 Problem Statement:

There is need of a system in which they can communicate directly with everyone. In such a system, a signer can make a sign and the other person can understand. In that system the input will be sign and corresponding output will be speech. It will use sign to text then text to speech conversions. Signs are made of units referred to as cheremes.

Now the problem exists is about first phase that is making signs and working on images. We can use simple backgrounds and gloves. But what if it’s about real time environment? So our focus will be recognizing signs without using gloves and without making specifications about environment. Another important thing is about head movements, facial expressions and body pose.

There is standard sign language in America and Europe. But in India there is no standardization of sign language. But in recent past Ramakrishna missions Vivekanand University, Coimbatore made a dictionary on Indian sign language. There are 2037 signs that are available in sign language.

1.2 Literature Survey:

Linguistic work on ISL began during late 1970s. Before that, the existence of ISL was not acknowledged. In 1977 a survey was conducted and it was revealed that ISL is a complete natural language instigated at the Indian subcontinent

There are systems from text to sign language.

There is system called INGIT, Hindi-To-Indian Sign Language machine translation system that is developed for railway reservation domain. The system takes input from clerk in the form of Hindi string...

... middle of paper ...

...zontal edges in blurred image. Then one stage is non maximum suppression, it is an edge thinning technique. Then canny operator trace edges through threshold. Differential edge detection can also be used to obtain edges. The result of it is shown in fig2.4.

Fig2.4 Binary image as a result of canny edge detector.

2.3.2 Zone based approach:

The output of canny edge detection is used to get features. Two [10 10] zones are created to store features. z= [10 10] s= [10 10]

i.e. there are two zones and each having capacity to store 100 features. So we can store 200 features of each image.

Image is having size [250 250] so it is divided into zones and their mean is stored in z zone and standard deviation in s zone. The output of this method is stored in a text file.

We can see the output in following figure Fig2.5 :

Fig2.5 Training sample