Sign language number recognition

• Main Author: Sandjaja, Iwan Njoto
• Format: text
• Language: English
• Published: Animo Repository 2008
• Subjects: Human-computer interaction; Computer vision; Neural networks (Computer science); Hidden Markov models
• Online Access: https://animorepository.dlsu.edu.ph/etd_masteral/3764; https://animorepository.dlsu.edu.ph/cgi/viewcontent.cgi?article=10602&context=etd_masteral
• Institution: De La Salle University

Sign language number recognition system lays down foundation for handshape recognition which addresses real and current problems in signing in the deaf community and leads to practical applications. The input for the sign language number recognition system is Filipino Sign Language number video files. The study is limited to include only 1000 numbers in Filipino Sign Language from number 1 to 1000. Each number is recorded 5 times using web camera. The frame size of the video is 640 x 480 and the speed is 15 frames per second. A student from School of Deaf Education and Applied Studies (SDEAS) De La Salle-College of Saint Benilde (DLS-CSB) does the Filipino Sign Language numbers with color-coded glove for dominant hand. The color coded gloves uses less color compared with other color-coded gloves in the existing research. The system extracts important features from the video using multi-color tracking algorithm which is faster than existing color tracking algorithm because it did not use recursive technique. The feature vectors contain the position of dominant-hands thumb in x and y coordinates and the x and y coordinates of other fingers relatively to the thumb position. Next, the system learns the Filipino Sign Language number in training phase and recognizes the Filipino Sign Language number in testing phase by transcribing Filipino Sign Language number into text. The system uses Hidden Markov Model (HMM) for training and testing phase. The system was evaluated in terms of training time and accuracy. The feature extraction could track 92.3% of all objects. The recognizer also could recognize Filipino sign language number with 85.52% average accuracy using the features from feature extraction module. Keywords Computer vision, Human Computer Interaction (HCI), Sign Language Recognition (SLR), Hidden Markov Model (HMM), hand tracking, multi-color tracking.