• P-ISSN 0974-6846 E-ISSN 0974-5645

Indian Journal of Science and Technology


Indian Journal of Science and Technology

Year: 2016, Volume: 9, Issue: Special Issue 1, Pages: 1-8

Original Article

Performance Analysis of KNN, SVM and ANN Techniques for Gesture Recognition System


The Gesture identification or recognition system is an emerging area of a current dynamic research about in PC vision and machine learning concepts. The primary design of the projected method is to develop a framework; it can understand the American Sign Language, which can recognize exact human gestures and practice them to express the information. Gesture signals are captured utilizing Kinect cameras along with a static back ground. The proposed strategy is isolated into 2 sections, significant features extraction stage, and Classification stage. Patterns are calculated using KNN, SVM and finally the ANN; it is a classification phase.It is then subjected to feature extraction and classification. In present work, SVM, KNN, and ANN tools are used in recognizing the gestures of ASL. The gesture recognition system is capable of recognizing only numerical ASL static signs with 97.10% accurateness. The obtained results proved that the SVM had a very good performance compared to other techniques.
Keywords: American Sign Language, ANN, Gesture recognition, KNN, SVM


Subscribe now for latest articles and news.