CNNFET: Convolutional neural network feature Extraction Tools

Neither machines nor even human can learn something not represented well enough. Therefore, feature extraction is one of the most important topics in machine learning. Deep convolutional neural networks are able to catch distinguishing features that can represent images or other digital signals. Thi...

Full description

Saved in:
Bibliographic Details
Main Authors: Huseyin Atasoy, Yakup Kutlu
Format: Article
Language:English
Published: Elsevier 2025-05-01
Series:SoftwareX
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S235271102500055X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850109340819652608
author Huseyin Atasoy
Yakup Kutlu
author_facet Huseyin Atasoy
Yakup Kutlu
author_sort Huseyin Atasoy
collection DOAJ
description Neither machines nor even human can learn something not represented well enough. Therefore, feature extraction is one of the most important topics in machine learning. Deep convolutional neural networks are able to catch distinguishing features that can represent images or other digital signals. This makes them very popular in signal processing and especially in image processing community. Despite the proven success of these networks, training processes of them are often expensive in terms of time and required hardware capabilities. In this paper, a user-friendly standalone Windows application titled “Convolutional Neural Network Feature Extraction Tools” (CNNFET) is presented. The application consists of tools that extract features from image sets using certain layers of pre-trained CNNs, process them, perform classifications on them and export features for further processing in Matlab or the popular machine learning software Weka.
format Article
id doaj-art-1c6ef38adb09482d99fb6f08dd7103fc
institution OA Journals
issn 2352-7110
language English
publishDate 2025-05-01
publisher Elsevier
record_format Article
series SoftwareX
spelling doaj-art-1c6ef38adb09482d99fb6f08dd7103fc2025-08-20T02:38:06ZengElsevierSoftwareX2352-71102025-05-013010208810.1016/j.softx.2025.102088CNNFET: Convolutional neural network feature Extraction ToolsHuseyin Atasoy0Yakup Kutlu1Information Technologies Office, Iskenderun Technical University, Hatay, Turkey; Corresponding author.Computer Engineering Department, Iskenderun Technical University, Hatay, TurkeyNeither machines nor even human can learn something not represented well enough. Therefore, feature extraction is one of the most important topics in machine learning. Deep convolutional neural networks are able to catch distinguishing features that can represent images or other digital signals. This makes them very popular in signal processing and especially in image processing community. Despite the proven success of these networks, training processes of them are often expensive in terms of time and required hardware capabilities. In this paper, a user-friendly standalone Windows application titled “Convolutional Neural Network Feature Extraction Tools” (CNNFET) is presented. The application consists of tools that extract features from image sets using certain layers of pre-trained CNNs, process them, perform classifications on them and export features for further processing in Matlab or the popular machine learning software Weka.http://www.sciencedirect.com/science/article/pii/S235271102500055XConvolutional neural networksFeature extractionTransfer learning
spellingShingle Huseyin Atasoy
Yakup Kutlu
CNNFET: Convolutional neural network feature Extraction Tools
SoftwareX
Convolutional neural networks
Feature extraction
Transfer learning
title CNNFET: Convolutional neural network feature Extraction Tools
title_full CNNFET: Convolutional neural network feature Extraction Tools
title_fullStr CNNFET: Convolutional neural network feature Extraction Tools
title_full_unstemmed CNNFET: Convolutional neural network feature Extraction Tools
title_short CNNFET: Convolutional neural network feature Extraction Tools
title_sort cnnfet convolutional neural network feature extraction tools
topic Convolutional neural networks
Feature extraction
Transfer learning
url http://www.sciencedirect.com/science/article/pii/S235271102500055X
work_keys_str_mv AT huseyinatasoy cnnfetconvolutionalneuralnetworkfeatureextractiontools
AT yakupkutlu cnnfetconvolutionalneuralnetworkfeatureextractiontools