Intelligent human action recognition: A framework of optimal features selection based on Euclidean Distance and Strong Correlation

Atiqa Sharif; Muhammad Attique Khan; Kashif Javed; Hafiz Gulfam Umer; Tassawar Iqbal; Tanzila Saba; Hashim Ali; Wasif Nisar

Intelligent human action recognition: A framework of optimal features selection based on Euclidean Distance and Strong Correlation

Atiqa Sharif
, Muhammad Attique Khan
, Kashif Javed
, Hafiz Gulfam Umer
, Tassawar Iqbal
, Tanzila Saba
, Hashim Ali
, Wasif Nisar

COMSATS University Islamabad
HITEC University
National University of Sciences and Technology Pakistan
Ghazi University
Prince Sultan University (PSU)

Research output: Contribution to journal › Article › peer-review

24 Citations (Scopus)

Abstract

Extracting salient and most prominent features from a given video sequence is a critical step in Human Action Recognition (HAR). The work presented within this article proposes a new method for HAR, which efficiently addresses the issue of robust feature selection. The proposed method initially fuses three different feature categories based on their highest values, and later selects most optimal features using a novel Euclidean distance (ED) and strong correlation (SC) methods. Finally, it classifies the selected features using multi-class classifier. For experimentation, four publically available datasets including Weizmann, KTH, UCF YouTube, and HMDB51 are used and results with improved classification accuracy, on average more than 94%, are obtained. Experimental results validate that the proposed approach outperforms the existing techniques.

Original language	English
Pages (from-to)	3-11
Number of pages	9
Journal	Control Engineering and Applied Informatics
Volume	21
Issue number	3
Publication status	Published - 2019
Externally published	Yes

Funding

Prince Sultan University Riyadh KSA, Grant/Award Number: 11-02-2019

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 9 Industry, Innovation, and Infrastructure
SDG 11 Sustainable Cities and Communities

Keywords

Features computation
Features fusion
Intelligent surveillance
Selection

ASJC Scopus subject areas

General Computer Science
Electrical and Electronic Engineering

Cite this

@article{3683dc44ee4f4d04ad3feda1853e81ad,

title = "Intelligent human action recognition: A framework of optimal features selection based on Euclidean Distance and Strong Correlation",

abstract = "Extracting salient and most prominent features from a given video sequence is a critical step in Human Action Recognition (HAR). The work presented within this article proposes a new method for HAR, which efficiently addresses the issue of robust feature selection. The proposed method initially fuses three different feature categories based on their highest values, and later selects most optimal features using a novel Euclidean distance (ED) and strong correlation (SC) methods. Finally, it classifies the selected features using multi-class classifier. For experimentation, four publically available datasets including Weizmann, KTH, UCF YouTube, and HMDB51 are used and results with improved classification accuracy, on average more than 94\%, are obtained. Experimental results validate that the proposed approach outperforms the existing techniques.",

keywords = "Features computation, Features fusion, Intelligent surveillance, Selection",

author = "Atiqa Sharif and Khan, \{Muhammad Attique\} and Kashif Javed and Umer, \{Hafiz Gulfam\} and Tassawar Iqbal and Tanzila Saba and Hashim Ali and Wasif Nisar",

note = "Funding Information: Prince Sultan University Riyadh KSA, Grant/Award Number: 11-02-2019 Publisher Copyright: {\textcopyright} 2019, Control Engineering and Applied Informatics Journal.",

year = "2019",

language = "English",

volume = "21",

pages = "3--11",

journal = "Control Engineering and Applied Informatics",

issn = "1454-8658",

publisher = "Control Engineering and Applied Informatics Journal",

number = "3",

}

TY - JOUR

T1 - Intelligent human action recognition

T2 - A framework of optimal features selection based on Euclidean Distance and Strong Correlation

AU - Sharif, Atiqa

AU - Khan, Muhammad Attique

AU - Javed, Kashif

AU - Umer, Hafiz Gulfam

AU - Iqbal, Tassawar

AU - Saba, Tanzila

AU - Ali, Hashim

AU - Nisar, Wasif

PY - 2019

Y1 - 2019

N2 - Extracting salient and most prominent features from a given video sequence is a critical step in Human Action Recognition (HAR). The work presented within this article proposes a new method for HAR, which efficiently addresses the issue of robust feature selection. The proposed method initially fuses three different feature categories based on their highest values, and later selects most optimal features using a novel Euclidean distance (ED) and strong correlation (SC) methods. Finally, it classifies the selected features using multi-class classifier. For experimentation, four publically available datasets including Weizmann, KTH, UCF YouTube, and HMDB51 are used and results with improved classification accuracy, on average more than 94%, are obtained. Experimental results validate that the proposed approach outperforms the existing techniques.

AB - Extracting salient and most prominent features from a given video sequence is a critical step in Human Action Recognition (HAR). The work presented within this article proposes a new method for HAR, which efficiently addresses the issue of robust feature selection. The proposed method initially fuses three different feature categories based on their highest values, and later selects most optimal features using a novel Euclidean distance (ED) and strong correlation (SC) methods. Finally, it classifies the selected features using multi-class classifier. For experimentation, four publically available datasets including Weizmann, KTH, UCF YouTube, and HMDB51 are used and results with improved classification accuracy, on average more than 94%, are obtained. Experimental results validate that the proposed approach outperforms the existing techniques.

KW - Features computation

KW - Features fusion

KW - Intelligent surveillance

KW - Selection

UR - https://www.scopus.com/pages/publications/85074419852

UR - https://www.scopus.com/pages/publications/85074419852#tab=citedBy

M3 - Article

AN - SCOPUS:85074419852

SN - 1454-8658

VL - 21

SP - 3

EP - 11

JO - Control Engineering and Applied Informatics

JF - Control Engineering and Applied Informatics

IS - 3

ER -

Intelligent human action recognition: A framework of optimal features selection based on Euclidean Distance and Strong Correlation

Abstract

Funding

UN SDGs

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this