Human action recognition in a wide and complex environment

Sanoj Kumar; Sanjeev Kumar; Balasubramanian Raman; Nagarajan Sukavanam

doi:10.1117/12.872316

2 February 2011 Human action recognition in a wide and complex environment

Sanoj Kumar, Sanjeev Kumar, Balasubramanian Raman, Nagarajan Sukavanam

Proceedings Volume 7871, Real-Time Image and Video Processing 2011; 78710I (2011) https://doi.org/10.1117/12.872316
Event: IS&T/SPIE Electronic Imaging, 2011, San Francisco Airport, California, United States

Abstract

In this paper, a linear discriminant analysis (LDA) based classifier employed in a tree structure is presented to recognize the human actions in a wide and complex environment. In particular, the proposed classifier is based on a supervised learning process and achieves the required classification in a multi-step process. This multi-step process is performed simply by adopting a tree structured which is built during the training phase. Hence, there is no need of any priori information like in other classifiers such as the number of hidden neurons or hidden layers in a multilayer neural network based classifier or an exhaustive search as used in training algorithms for decision trees. A skeleton based strategy is adopted to extract the features from a given video sequence representing any human action. A Pan-Tilt-Zoom (PTZ) camera is used to monitor the wide and complex test environment. A background mosaic image is built offline and used to compute the background images in real time. A background subtraction strategy has been adopted for detecting the object in various frames and to extract their corresponding silhouette. A skeleton based process is used to extract attributes of a feature vector corresponding to a human action. Finally, the proposed framework is tested on various indoor and outdoor scenarios and encouraging results are achieved in terms of classification accuracy.

Citation Download Citation

Sanoj Kumar, Sanjeev Kumar, Balasubramanian Raman, and Nagarajan Sukavanam "Human action recognition in a wide and complex environment", Proc. SPIE 7871, Real-Time Image and Video Processing 2011, 78710I (2 February 2011); https://doi.org/10.1117/12.872316

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