Artificial neural network based identification and classification of images of Bharatanatya gestures
Chattopadhyay Soumya, Muzameel Ahmed · 2017
The main aim of recognising gestures is to build a system that can identify human gestures that are specific and then to use them to put forth desired information to the device. By using mathematical algorithms, human gestures can be interpreted. This is referred to as Gesture Recognition. Mudra is an expressive form of gesture that is mainly used in Indian classical dance form where the gesture is in visual form so as to connect with the audience. Different positions of body parts make an expressive and meaningful mudra which can be both static and dynamic. Here the project makes an attempt to identify the hand gesture/mudra using Image-processing and also Pattern Recognition methods. An attempt in computer aided recognition of Bharatnatya Mudras is made using Image classification and processing techniques using Artificial Neural Network. The entry that gives the least difference to the feature of a mudra is the match for the input image considered. Finally, the system also provides the health benefit of the identified mudra.