Work place: Department of Studies in Computer Science, Manasagangothri, University of Mysore, Mysore-570006. INDIA.
Research Interests: Network Security, Network Architecture, Pattern Recognition, Neural Networks
D S Guru currently is working as a Professor in the Department of Studies in Computer Science, University of Mysore. He obtained Doctorate in Computer Science from the University of Mysore. He has received the prestigious BOYSCAST fellowship from the DST, Government of India to take up advanced research at Michigan State University, USA. He has been a reviewer for many international journals including IEEE Transaction on Neural Network, ACM transaction, Pattern Recognition, Patter Recognition Letters. HE has published about 235 research papers out of which about 50 are in the international journals including IEEE Transaction, Pattern Recognition, and Pattern Recognition Letters.
DOI: https://doi.org/10.5815/ijigsp.2015.09.07, Pub. Date: 8 Aug. 2015
In this paper, we propose a model for recognition of sign language being used by communication impaired people in the society. A novel method of extracting features from a video sequence of signs is proposed. Key frames are selected from a given video shots of signs to reduce the computational complexity yet retaining the significant information for recognition. A set of features is extracted from each key frame to capture the trajectory of hand movements made by the signer. The same sign made by different signers and by the same signers at different instances may have variations. The concept of symbolic data particularly interval type data is used to capture such variations and to efficiently represent signs in the knowledgebase. A suitable similarity measure is explored for the purpose of matching and recognition of signs. A database of signs made by communication impaired people of Mysore region is created and extensive experiments are conducted on this database to demonstrate the performance of the proposed approach.[...] Read more.
Subscribe to receive issue release notifications and newsletters from MECS Press journals