Spectral domain cross correlation function and generalized Learning Vector Quantization for recognizing and classifying Indonesian Sign Language

This paper shows the first part of the automatic Indonesian Sign Language (SIBI) into text translation system. The focus of this project is on translation of the alphabet (A to Z) and numbers 1 to 10. Using a combination of a Kinect camera, Discrete Cosine Transform (DCT), Cross Correlation Function and classifying algorithm Generalized Learning Vector Quantization (GLVQ) can create a simple system to recognize alphabet A to Z and number 1 to 10 in Indonesian Sign Language. The skeleton extraction function and depth sensor from the Kinect camera are used to capture and transfer of hand gesture movements into frames of images. DCT is used to transform spatial data of each frame of image into its spectral domain. Collection of Cross Correlation values between same rows or columns of data from two consecutive frames can be used as a signature of a character. Each signature is unique and needs a small amount of data. GLVQ is used as the classifying algorithm to recognize the character. From our experiments, the system we proposed has obtained a high degree of accuracy in the recognition of alphabet and numbers in Indonesian Sign Language.

[1]  Jianqin Zhou,et al.  On discrete cosine transform , 2011, ArXiv.

[2]  Kishor S. Trivedi.  Probability and Statistics with Reliability, Queuing, and Computer Science Applications , 1984 .

[3]  Dipak Kumar Ghosh,et al.  A static hand gesture recognition algorithm using k-mean based radial basis function neural network , 2011, 2011 8th International Conference on Information, Communications & Signal Processing.

[4]  Narendra Ahuja,et al.  Extraction of 2D Motion Trajectories and Its Application to Hand Gesture Recognition , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Atsushi Sato,et al.  Generalized Learning Vector Quantization , 1995, NIPS.

[6]  B. Lekhashri,et al.  Use of motion-print in sign language recognition , 2011, 2011 National Conference on Innovations in Emerging Technology.

[7]  Ying Gao,et al.  Real-time sign language recognition based on neural network architecture , 2011, 2011 IEEE 43rd Southeastern Symposium on System Theory.

[8]  Christos Faloutsos,et al.  Compressed-domain video indexing techniques using DCT and motion vector information in MPEG video , 1997, Electronic Imaging.

[9]  G.I.S. Ruas,et al.  Real-time video based finger spelling recognition system using low computational complexity Artificial Neural Networks , 2006, 2006 International Telecommunications Symposium.

[10]  Hafiz Imtiaz,et al.  A face recognition scheme based on spectral domain cross-correlation function , 2011, TENCON 2011 - 2011 IEEE Region 10 Conference.