00000ctm a22000004a 4500 UP-99796217603116304 Buklod 20230503092542.0 m |o d | ta 070713s2001 xx r |||| u|eng d (iLib)UPD-00000179746 DENGII eng LG 993.5 2001 C65 C35 Math bilang Ma. Irene I. Callanta ... [et al.]. 2001. 49 leaves ill. Thesis (B.S. Computer Science)--University of the Philippines Diliman. Several systems for gesture recognition have been developed using different strategies and approaches. This paper describes a system for real time gesture recognition, which can distinguish a small vocabulary of about 14 different, static hand gestures. The fourteen gestures are representatives for actions of Addition, Multiplication, Division, Subtraction and the numbers One to Ten. A voice translation of the mathematical expression and its corresponding answer is also part of this system. Our approach uses two-dimensional images of hand postures. The pattern recognition is employed using the back propagation algorithm. All the computation occurs on a SGI workstation. Experiments with our system achieved recognition accuracy up to 90%. Our system uses input from a monocular color video camera. Human-computer interaction. Gesture Data processing. Image processing Digital techniques Mathematical models. Optical pattern recognition Mathematical models. Neural networks (Computer science) Hand gesture recognition. Callanta, Ma. Irene I. del Castillo, Marion S. Gaerlan, Ma. Joan. Molato, Maricris M. Palomique, Norleen O. Thesis FI UP UPD DENG-II LG 993.5 2001 C65 C35 Thesis