A simplified shock graph for human posture classification using the adaptive neuro fuzzy inference system

Aini Hussain, S. Shahbudin, Hafizah Husain, Salina Abdul Samad, N. M. Tahir

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

This paper affords the use of neuro-fuzzy technique called the Adaptive Network-based Fuzzy Inference System (ANFIS) to highlight its ability to perform human posture classification tasks using the new Simplified Shock Graph (SSG) representation. The ANFIS model with a five-layered architecture was trained using the SSG features to classify image data comprising various human postures. SSG was derived from a shock graph in which a shock graph is a shape abstraction that decomposes a shape into a set of hierarchically organized primitive parts. It is a powerful shape representation technique in which the silhouette of an object is represented in terms of a set of qualitatively defined parts, organized in a hierarchical and directed acyclic graph. The SSG feature provides a compact, unique and simple way of representing human shape. To test the effectiveness of the SSG features, the ANFIS was identified as the appropriate classifier to perform the task of human posture classification. Results obtained showed that the ANFIS model is very suitable and can generate excellent classification results provided that the right type and number of Membership Functions (MFs) are used in the classification task.

Original languageEnglish
Pages (from-to)2035-2048
Number of pages14
JournalJournal of Information and Computational Science
Volume9
Issue number8
Publication statusPublished - Aug 2012

Fingerprint

Fuzzy inference
system model
Membership functions
abstraction
Classifiers
ability

Keywords

  • Adaptive network-based fuzzy inference system
  • Human posture classification
  • Hybrid learning algorithm
  • Membership functions
  • Posture positions
  • Simplified shock graph

ASJC Scopus subject areas

  • Information Systems
  • Computer Graphics and Computer-Aided Design
  • Computational Theory and Mathematics
  • Library and Information Sciences

Cite this

A simplified shock graph for human posture classification using the adaptive neuro fuzzy inference system. / Hussain, Aini; Shahbudin, S.; Husain, Hafizah; Abdul Samad, Salina; Tahir, N. M.

In: Journal of Information and Computational Science, Vol. 9, No. 8, 08.2012, p. 2035-2048.

Research output: Contribution to journalArticle

@article{ebc0a0f0ce664b86b499d1118ec49562,
title = "A simplified shock graph for human posture classification using the adaptive neuro fuzzy inference system",
abstract = "This paper affords the use of neuro-fuzzy technique called the Adaptive Network-based Fuzzy Inference System (ANFIS) to highlight its ability to perform human posture classification tasks using the new Simplified Shock Graph (SSG) representation. The ANFIS model with a five-layered architecture was trained using the SSG features to classify image data comprising various human postures. SSG was derived from a shock graph in which a shock graph is a shape abstraction that decomposes a shape into a set of hierarchically organized primitive parts. It is a powerful shape representation technique in which the silhouette of an object is represented in terms of a set of qualitatively defined parts, organized in a hierarchical and directed acyclic graph. The SSG feature provides a compact, unique and simple way of representing human shape. To test the effectiveness of the SSG features, the ANFIS was identified as the appropriate classifier to perform the task of human posture classification. Results obtained showed that the ANFIS model is very suitable and can generate excellent classification results provided that the right type and number of Membership Functions (MFs) are used in the classification task.",
keywords = "Adaptive network-based fuzzy inference system, Human posture classification, Hybrid learning algorithm, Membership functions, Posture positions, Simplified shock graph",
author = "Aini Hussain and S. Shahbudin and Hafizah Husain and {Abdul Samad}, Salina and Tahir, {N. M.}",
year = "2012",
month = "8",
language = "English",
volume = "9",
pages = "2035--2048",
journal = "Journal of Information and Computational Science",
issn = "1548-7741",
publisher = "Binary Information Press",
number = "8",

}

TY - JOUR

T1 - A simplified shock graph for human posture classification using the adaptive neuro fuzzy inference system

AU - Hussain, Aini

AU - Shahbudin, S.

AU - Husain, Hafizah

AU - Abdul Samad, Salina

AU - Tahir, N. M.

PY - 2012/8

Y1 - 2012/8

N2 - This paper affords the use of neuro-fuzzy technique called the Adaptive Network-based Fuzzy Inference System (ANFIS) to highlight its ability to perform human posture classification tasks using the new Simplified Shock Graph (SSG) representation. The ANFIS model with a five-layered architecture was trained using the SSG features to classify image data comprising various human postures. SSG was derived from a shock graph in which a shock graph is a shape abstraction that decomposes a shape into a set of hierarchically organized primitive parts. It is a powerful shape representation technique in which the silhouette of an object is represented in terms of a set of qualitatively defined parts, organized in a hierarchical and directed acyclic graph. The SSG feature provides a compact, unique and simple way of representing human shape. To test the effectiveness of the SSG features, the ANFIS was identified as the appropriate classifier to perform the task of human posture classification. Results obtained showed that the ANFIS model is very suitable and can generate excellent classification results provided that the right type and number of Membership Functions (MFs) are used in the classification task.

AB - This paper affords the use of neuro-fuzzy technique called the Adaptive Network-based Fuzzy Inference System (ANFIS) to highlight its ability to perform human posture classification tasks using the new Simplified Shock Graph (SSG) representation. The ANFIS model with a five-layered architecture was trained using the SSG features to classify image data comprising various human postures. SSG was derived from a shock graph in which a shock graph is a shape abstraction that decomposes a shape into a set of hierarchically organized primitive parts. It is a powerful shape representation technique in which the silhouette of an object is represented in terms of a set of qualitatively defined parts, organized in a hierarchical and directed acyclic graph. The SSG feature provides a compact, unique and simple way of representing human shape. To test the effectiveness of the SSG features, the ANFIS was identified as the appropriate classifier to perform the task of human posture classification. Results obtained showed that the ANFIS model is very suitable and can generate excellent classification results provided that the right type and number of Membership Functions (MFs) are used in the classification task.

KW - Adaptive network-based fuzzy inference system

KW - Human posture classification

KW - Hybrid learning algorithm

KW - Membership functions

KW - Posture positions

KW - Simplified shock graph

UR - http://www.scopus.com/inward/record.url?scp=84865116619&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865116619&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84865116619

VL - 9

SP - 2035

EP - 2048

JO - Journal of Information and Computational Science

JF - Journal of Information and Computational Science

SN - 1548-7741

IS - 8

ER -