Beta
ManageSearchChatDiscover
311989

Video Analysis For Human Action Recognition Using Deep Convolutional Neural Networks

Article

Last updated: 28 Dec 2024

OverviewSimilar Items

Subjects

-

Tags

-

Abstract

 In the last few years, human action recognition potential applications have been studied in many fields such as
robotics, human computer interaction, and video surveillance systems and it has been evaluated as an active research area. This paper presents a recognition system using deep learning to recognize and identify human actions from video input. The proposed system has been fine-tuned by partial training and dropout of the classification layer of Alexnet and
replacing it by another one that use SVM. The performance of the network is boosted by using key frames that were
extracted via applying Kalman filter during dataset augmentation. The proposed system resulted in promising performance compared to the state of the art approaches. The classification accuracy reached 92.35%. 

DOI

10.21608/mjcis.2018.311989

Keywords

Human action recognition, Deep learning, Machine Learning, Video analysis, Image Segmentation, Feature Extraction, Kalman filter, human action, Convolutional neural network

Authors

First Name

Nehal

Last Name

N. Mostafa

MiddleName

-

Affiliation

Faculty of computers and information systems , C.S dep. Mansoura University, Egypt

Email

-

City

-

Orcid

-

First Name

Mohammed

Last Name

F. Alrahmawy

MiddleName

-

Affiliation

Faculty of computers and information systems , C.S dep. Mansoura University, Egypt

Email

-

City

-

Orcid

-

First Name

Omaima

Last Name

Nomair

MiddleName

-

Affiliation

Faculty of computers and information systems , C.S dep. Mansoura University, Egypt

Email

-

City

-

Orcid

-

Volume

14

Article Issue

1

Related Issue

42819

Issue Date

2018-06-01

Receive Date

2023-08-10

Publish Date

2018-06-01

Page Start

1

Page End

7

Print ISSN

2090-1666

Online ISSN

2090-1674

Article File

MJCIS_Volume 14_Issue 1_Pages 1-7.pdf

PDF . 780.1kB

Link

https://mjcis.journals.ekb.eg/article_311989.html

Detail API

https://mjcis.journals.ekb.eg/service?article_code=311989

Order

311,989

Type

Original Research Articles.

Type Code

1,784

Publication Type

Journal

Publication Title

Mansoura Journal for Computer and Information Sciences

Publication Link

https://mjcis.journals.ekb.eg/

MainTitle

Video Analysis For Human Action Recognition Using Deep Convolutional Neural Networks

Details

Type

Article

Created At

28 Dec 2024