Beta
ManageSearchChatDiscover
198376

Product Based Classification of Bulk Food Grains using Bag of Visual Words and Deep Features

Article

Last updated: 22 Jan 2023

OverviewSimilar Items

Subjects

-

Tags

AI & Expert Systems
Algorithms & Applications

Abstract

The goal of this research is to compare between the performance of the traditional machine learning classification algorithm using Bag of Visual Words (BoVW) method and off-the-shelf deep features extracted by VGG-19, and Inception-V3 models and trained SVMs using the extracted features. By comparing the AUC, sensitivity, and specificity of SVM with VGG-19 and Inception-V3, we can conclude that off-the-shelf deep features has an important impact on food grains image classification. 

DOI

10.21608/kjis.2021.198376

Keywords

Image classification, Bag of visual words, Transfer Learning, Convolutional Neural Networks, Deep learning

Authors

First Name

Abdelmgeid

Last Name

Ali

MiddleName

-

Affiliation

Computer Science Department, Faculty of Science, Minia University, Al Minia 61519, Egypt

Email

abdelmgeid@yahoo.com

City

-

Orcid

-

First Name

Usama

Last Name

Mohammed

MiddleName

Sayed

Affiliation

Electrical Engineering Department, Faculty of Engineering, Assiut University, Assiut 71516, Egypt

Email

usama@aun.edu.eg

City

-

Orcid

-

First Name

Rehab

Last Name

Nour

MiddleName

Ragaa

Affiliation

Computer Science Department, Faculty of Science, Minia University, Al Minia 61519, Egypt

Email

rehab.ibrahim@mu.edu.eg

City

-

Orcid

-

Volume

2

Article Issue

2

Related Issue

28022

Issue Date

2021-10-01

Receive Date

2021-10-01

Publish Date

2021-10-07

Page Start

1

Page End

6

Print ISSN

2537-0677

Online ISSN

2535-1478

Article File

KJIS_Volume 2_Issue 2_Pages 1-6.pdf

PDF . 406.5kB

Link

https://kjis.journals.ekb.eg/article_198376.html

Detail API

https://kjis.journals.ekb.eg/service?article_code=198376

Order

8

Type

Original Article

Type Code

462

Publication Type

Journal

Publication Title

Kafrelsheikh Journal of Information Sciences

Publication Link

https://kjis.journals.ekb.eg/

MainTitle

-

Details

Type

Article

Created At

22 Jan 2023