International Journal of Scientific and Research Publications, Volume 2, Issue 6, June 2012 1 ISSN 2250-3153 www.ijsrp.org Comparative Study of Sign Language Recognition Systems Ms. Rashmi D. Kyatanavar, Prof. P. R. Futane Department of Computer Engineering, Sinhgad College of Engineering, Pune, India Abstract- Sign Language (SL) is the communication method for deaf people. But these sign languages are not standard & universal. The grammar differs from country to country. As the sign language is the only method for deaf community to communicate with normal people, we need Sign Language Translators (SLT). The basic automatic SLT uses two approaches. First one is using the electronic data gloves. These gloves have inbuilt sensors & it is worn by the signer to detect hand posture. But these gloves are having high cost. So, visual approach is most suitable & widely used. Here a camera is used to capture images of signer & then image processing is carried out to perform recognition. For the vision based Sign Language Recognition (SLR), various methods are available. Three such methods are discussed & compared in this paper. Index Terms- Sign Language Recognition (SLR) I. INTRODUCTION Deaf people interfacing is a very challenging issue due to its cardinal social interest and its inherent complexity like, (1) Nature of information is multi-channel. (2) Signing vary from person to person. (3) Presence of disturbances (surrounding furniture, cloths) (4) Sign languages from different region differ significantly. There are two main directions in sign language recognition. One is using data gloves & other is visual approach. Instrumented glove approach simplifies the recognition but complicates the hardware. Also it is expensive & less user friendly. On the other hand, vision-based approach is most suitable, user-friendly & affordable. So, it is widely used. II. VISION BASED METHODS FOR SIGN LANGUAGE RECOGNITION A. SLR Based on Skin Color [2] This is an intelligent & simple system for converting sign language into voice signal by tracking head & hand gestures. This system proposes a simple gesture extraction algorithm for extracting features from the images of a video stream. Figure 1: System block diagram for SLR using skin color [2] This system is very simple & subject is not required to wear any glove. But the subject must wear a dark color, long sleeve shirt. Then the gesture signs are recorded. Each image frame is segmented into three regions- head, left hand and right hand. Then these segmented images are converted into binary images. In feature extraction stage, for each frame the area of objects in segmented binary image is calculated. So, each frame has 3 segmented areas- head area, left hand area and right hand area. There is a different segmented area for each gesture type. Each segmented area is treated as a discrete event & DCT is applied to it. First 15 DCT coefficients are considered as features. They correspond to each segmented area. Combination of DCT coefficients from 3 segmented image areas are used as feature vector for Neural Network (NN). Here, a simple NN model is developed for sign recognition. The features computed from video stream are given as an input to this NN. To classify the