Researchers in sign language recognition customized different sensors to capture hand signs. Gloves,digital cameras, depth cameras and Kinect were used alternatively in most systems. Due to signs closeness, input accuracy is a very essential constraint to reach a high recognition accuracy. Although previous systems accomplished high recognition accuracy, they suffer from stability in realistic environment due to variance in signing speed, lighting, etc… In this paper, a recognition system for ArSL has been developed based on a new digital sensor called “Leap Motion”. This sensor tackles the major issues in vision-based systems such as skin color, lighting etc… Leap motion captures hands and fingers movements in 3D digital format.
The sensor throws 3D digital information in each frame of movement. These temporal and spatial features are fed into a Multi-layer perceptron Neural Network (MLP). The system was tested on 50 different dynamic signs (distinguishable without non-manual features) and the recognition accuracy reached 88% for two different persons. Although Leap motion tracks both hands accurately, unfortunately Leap motion does not track non-manual features. This system can be enhanced by adding other sensors to track other non-manual features such as facial expressions and body poses. The proposed sensor can work simultaneously with leap motion to capture all sign’s features.