An Efficient Framework for 2-Dimensional Gesture Based Telugu Character Recognition

Gesture identification plays a vital role in today's human-computer interaction. In this paper, we proposed a sensor based gesture recognition system which makes the teacher to write in Telugu language on digital board from anywhere within the class room. Various classification algorithms k-Nearest Neighbor (KNN), Support Vector Machine (SVM) and Decision tree are individually used for hand gesture based Telugu character recognition. Here, we assess the performance of three classification algorithms which are compared with 16 different Telugu character vowel gestures. Each gesture is collected by using an inertial sensor based embedded device. The dataset contains 16 gestures, each gesture repeated for eleven times from three different people. The gesture identification accuracy for k-Nearest Neighbor classification is 97.2%, SVM is 92.8% and Decision tree is 86.5%.

[1]  William T. Freeman,et al.  Television control by hand gestures , 1994 .

[2]  Thomas H. Speeter.  Transforming Human Hand Motion for Telemanipulation , 1992, Presence: Teleoperators & Virtual Environments.

[3]  N. Balaji,et al.  INTERACTIVE CHARACTER RECOGNITION TECHNIQUE FOR MOBILE BASED SYSTEMS: AN APPLICATION FOR AUTISTIC PEOPLE , 2013 .

[4]  Deyou Xu.  A Neural Network Approach for Hand Gesture Recognition in Virtual Reality Driving Training System of SPG , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[5]  Jing Yang,et al.  Inertial sensor based recognition of 3-D character gestures with an ensemble classifiers , 2004, Ninth International Workshop on Frontiers in Handwriting Recognition.

[6]  Kongqiao Wang,et al.  Hand gesture recognition and virtual game control based on 3D accelerometer and EMG sensors , 2009, IUI.

[7]  Seungjin Choi,et al.  Gesture-recognizing hand-held interface with vibrotactile feedback for 3D interaction , 2009, IEEE Transactions on Consumer Electronics.

[8]  Delbert Dueck,et al.  Clustering by Passing Messages Between Data Points , 2007, Science.

[9]  N. Balaji,et al.  Real Time Implementation of Telugu Character Recognition using Raspberry Pi , 2014 .

[10]  Allen Y. Yang,et al.  Distributed segmentation and classification of human actions using a wearable motion sensor network , 2008, 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.

[11]  Ashok Kumar Sahoo,et al.  INDIAN SIGN LANGUAGE RECOGNITION USING NEURAL NETWORKS AND KNN CLASSIFIERS , 2014 .

[12]  Timo Pylvänäinen,et al.  Accelerometer Based Gesture Recognition Using Continuous HMMs , 2005, IbPRIA.

[13]  Lars Bretzner,et al.  Use Your Hand as a 3-D Mouse, or, Relative Orientation from Extended Sequences of Sparse Point and Line Correspondences Using the Affine Trifocal Tensor , 1998, ECCV.

[14]  Christos H. Papadimitriou,et al.  A Worst-Case Analysis of Nearest Neighbor Searching by Projection , 1980, ICALP.

[15]  Robert F. Sproull,et al.  Refinements to nearest-neighbor searching ink-dimensional trees , 1991, Algorithmica.

[16]  Shengli Zhou,et al.  Hand-written character recognition using MEMS motion sensing technology , 2008, 2008 IEEE/ASME International Conference on Advanced Intelligent Mechatronics.

[17]  Jani Mäntyjärvi,et al.  Enabling fast and effortless customisation in accelerometer based gesture interaction , 2004, MUM '04.