الفهرس 
ABSTRACTOnly 14 pages are availabe for public view
 |  | from | 151 |  |  |
| | | from | 151 | | |
Abstract
This thesis introduces a real time system for automatic Arabic sign language recognition system based on dynamic time warping matching algorithm and kinect sensor. The communication between human and machines or between people could be facilitated using gestures called sign language. The aim of the sign language recognition is to provide an accurate and convenient mechanism to transcribe sign gestures into meaningful text or speech so that communication between deaf and others can be made easily. In this thesis a translator based on dynamic time warping algorithm is developed. In this algorithm each signed word is coordinated and matched among database, then display the text and the corresponding pronunciation of the input sign. To catch the sign the Microsoft Kinect XBOX 360TM is used. The kinect is capable of giving in-depth vision image and color vision image of everything in front of it. The kinect produces data about the body joints, the skeleton body action which can be tracked more precise and easier. To test our system we built our data using a large set of samples for a dictionary of 30 isolated Arabic words homemade signs from the Standard Arabic sign language. The system operates in different modes including online, signer-dependent, and signer-independent modes. The presented system allows the signers to perform signs freely and naturally. Experimental results demonstrate that the presented system has higher recognition rate compared with other systems for all modes. For signer-dependent online case, the proposed system achieved recognition rate of 97.58%, and for signer-independent online case, we achieved a Recognation rate of 95.25%.
Keywords: Arabic sign language recognition, Microsoft Visual Studio, Kinect, Dynamic time warping algorithm.