Hand gesture recognition based on motion history images for a simple human-computer interaction system

Ivanna K Timotius; Iwan Setyawan

doi:10.1117/12.2011383

14 March 2013 Hand gesture recognition based on motion history images for a simple human-computer interaction system

Ivanna K Timotius, Iwan Setyawan

Proceedings Volume 8768, International Conference on Graphic and Image Processing (ICGIP 2012); 87684M (2013) https://doi.org/10.1117/12.2011383
Event: 2012 International Conference on Graphic and Image Processing, 2012, Singapore, Singapore

Abstract

A human-computer interaction can be developed using several kind of tools. One choice is using images captured using a camera. This paper proposed a simple human-computer interaction system based on hand movement captured by a web camera. The system aims to classify the captured movement into one of three classes. The first two classes contain hand movements to the left and right, respectively. The third class contains non-hand movements or hand movements to other directions. The method used in this paper is based on Motion History Images (MHIs) and nearest neighbor classifier. The resulting MHIs are processed in two manners, namely by summing the pixel values along the vertical axis and reshaping into vectors. We also use two distance criteria in this paper, respectively the Euclidian distance and cross correlation. This paper compared the performance of the combinations of different MHI data processing and distance criteria using 10 runs of 2-fold cross validation. Our experiments show that reshaping the MHI data into vectors combined with a Euclidean distance criterion gives the highest average accuracy, namely 55.67%.

Citation Download Citation

Ivanna K Timotius and Iwan Setyawan "Hand gesture recognition based on motion history images for a simple human-computer interaction system", Proc. SPIE 8768, International Conference on Graphic and Image Processing (ICGIP 2012), 87684M (14 March 2013); https://doi.org/10.1117/12.2011383

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available