期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2016
卷号：89
期号：2
出版社：Journal of Theoretical and Applied
摘要：The focus of this paper is to compare two artificial intelligent paradigms for gesture recognition from videos of Indian sign language with artless backgrounds. Hand and Head segmentation giving rise to shape features for the entire video sequence is inputted to Artificial Neural Networks (ANN) and Fuzzy Inference Engine (FIS). Chan Vase active contour extracts shape models for the classifier. Two classifiers are inputted with same set of training and testing samples. The classifiers are compared based on their data handling measured using recognition rate and execution times measured by training and testing times. The results indicate a tradeoff between training data size and execution times of ANN and FIS. The classifiers were tested on 86 gestures from 10 different signers.
