A Comparative Study of Certain Classifiers for Bharatanatyam Mudra Images' Classification using Hu-Moments

A Comparative Study of Certain Classifiers for Bharatanatyam Mudra Images' Classification using Hu-Moments

Basavaraj S. Anami (KLEIT Hubballi, Karnataka, India) and Venkatesh Arjunasa Bhandage (Tontadarya College of Engineering, Gadag, India)
Copyright: © 2019 |Pages: 14
DOI: 10.4018/IJACDT.2019070104

Abstract

India is rich in culture and heritage where various traditional dances are practiced. Bharatanatyam is an Indian classical dance, which is composed of various body postures and hand gestures. This ancient art of dance has to be studied under guidance of dance teachers. In present days there is a scarcity of Bharatanatyam dance teachers. There is a need to adopt technology to popularize this dance form. This article presents a 3-stage methodology for the classification of Bharatanatyam mudras. In the first stage, acquired images of Bharatanatyam mudras are preprocessed to obtain contours of mudras using canny edge detector. In the second stage, Hu-moments are extracted as features. In the third stage, rule-based classifiers, artificial neural networks, and k-nearest neighbor classifiers are used for the classification of unknown mudras. The comparative study of classification accuracies of classifiers is provided at the end. The work finds application in e-learning of ‘Bharatanatyam' dance in particular and dances in general and automation of commentary during concerts.
Article Preview
Top

Introduction

Recent advancements in digital image processing have led to developments of various applications such as hand gesture recognition (Liu, Yin, & Zhang, 2012; Wang, Chen, & Li, 2016), sign language identification (Karishma & Jalal, 2013; Raut, Mali, Thepade, & Sanas, 2014; Zadghorban & Nahvi, 2016; Fagiani, Principi, Squartini, & Piazza, 2015; Zaki & Shaheen, 2011), body posture recognition (Dantone, 2013), and human action recognition (Kishore et al., 2018). One such significant application is seen in Bharatanatyam dance and mudra classification. Bharatanatyam dance is an ancient Indian classical dance form which is composed of various body postures, facial expressions, and coordinated movement of various hand gestures which are performed to the accompaniment of dance syllables. These dance movements convey some useful meaning to the audience. The semantic meaning and feelings of these dance syllables can only be properly understood by the connoisseur. The novice learners and general audience may find it difficult to understand the inner as well as outer feelings conveyed by the dancers. The Natyashastra, an earlier repository of Indian classical dance, has details of dance syllables used in various dance forms. A mudra is the most striking feature of this Indian classical dance, which uses hand gestures. The Natyashastra mentions a set of 108 dance postures called Karanas, 28 Asamyukta mudras, 23 Samyukta mudrasi, and 13 Nritya mudras or decorative hand gestures. The 24th mudra Avahitta was recently added to the list of Samyukta mudras. The mudra is also called hasta in the literature. Karana means doing in the Sanskrit language. Natyashastra states that karanas are the dance frameworks which are supposed to spiritually enlighten the spectators, as opposed to other dance forms such as folk or pop dance, which can only entertain the spectators. Karanas are brief movement phrases describing specific leg, hip, body, and arm movements accompanied by mudras. It is also stated that one who performs well this karana dance will become free from all the sins.

The dance gestures, a way of non-verbal communication, are used to visually convey outer events or things, as well as inner feelings. Hand sign is a ritual gesture in Hinduism and Buddhism. There are two types of hand gestures used in Bharatanatyam dance. Asamyukta mudras are performed using single hand, and Samyukta mudras are performed using both hands. The decorative hand gestures are used to increase the beauty of dance during concerts. These karanas and mudras are prominent parts of the dancer's vocabulary. Figure 1 shows typical postures of some of Bharatanatyam dance along with some mudras.

Figure 1.

Typical postures in Bharatanatyam dance with mudras

IJACDT.2019070104.f01

The Asamyukta mudras are the key elements of Bharatanatyam dance and play a significant role in conveying the feelings of the dancer to the audience. Hence, the current work proposes developing a system for classification of Asamyukta mudras. Some single hand mudras are shown in Figure 2. Some of the mudras look similar and hence conflict, leading to misclassification of mudras. In this work, the most conflicting mudras are taken out and a separate list of conflicting Asamyukta mudras is formed. Non-conflicting mudras look totally dissimilar, leading to effective classification of mudras. The experiment on these separate groups gives an improvement in the classification accuracies of mudras. Hence, in this work, Asamyukta mudras are broadly classified into conflicting mudras and non-conflicting mudras. Figure 2 shows some of the non-conflicting mudras. Figure 3 shows that Mrugashirsha and Hamsapakshika mudras look similar to each other and hence are conflicting.

Figure 2.

Single hand mudra images of Bharatanatyam dance

IJACDT.2019070104.f02

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 9: 2 Issues (2020): Forthcoming, Available for Pre-Order
Volume 8: 2 Issues (2019)
Volume 7: 2 Issues (2018)
Volume 6: 2 Issues (2017)
Volume 5: 2 Issues (2016)
Volume 4: 2 Issues (2014)
Volume 3: 2 Issues (2013)
Volume 2: 2 Issues (2012)
Volume 1: 2 Issues (2011)
View Complete Journal Contents Listing