Deep Learning and Handcrafted Features for Virus Image Classification
In this work, we present an ensemble of descriptors for the classification of virus images acquired using transmission electron microscopy. We trained multiple support vector machines on different sets of features extracted from the data. We used both handcrafted algorithms and a pretrained deep neu...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-12-01
|
Series: | Journal of Imaging |
Subjects: | |
Online Access: | https://www.mdpi.com/2313-433X/6/12/143 |
id |
doaj-09b2771b13534ab5adfa28c8202d406f |
---|---|
record_format |
Article |
spelling |
doaj-09b2771b13534ab5adfa28c8202d406f2020-12-22T00:00:44ZengMDPI AGJournal of Imaging2313-433X2020-12-01614314310.3390/jimaging6120143Deep Learning and Handcrafted Features for Virus Image ClassificationLoris Nanni0Eugenio De Luca1Marco Ludovico Facin2Gianluca Maguolo3Dipartimento di Ingegneria Dell’informazione, University of Padova, via Gradenigo 6, 35131 Padova, ItalyDipartimento di Ingegneria Dell’informazione, University of Padova, via Gradenigo 6, 35131 Padova, ItalyDipartimento di Ingegneria Dell’informazione, University of Padova, via Gradenigo 6, 35131 Padova, ItalyDipartimento di Ingegneria Dell’informazione, University of Padova, via Gradenigo 6, 35131 Padova, ItalyIn this work, we present an ensemble of descriptors for the classification of virus images acquired using transmission electron microscopy. We trained multiple support vector machines on different sets of features extracted from the data. We used both handcrafted algorithms and a pretrained deep neural network as feature extractors. The proposed fusion strongly boosts the performance obtained by each stand-alone approach, obtaining state of the art performance.https://www.mdpi.com/2313-433X/6/12/143virus classificationtexture descriptorsdeep learninglocal binary patternsensemble of descriptors |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Loris Nanni Eugenio De Luca Marco Ludovico Facin Gianluca Maguolo |
spellingShingle |
Loris Nanni Eugenio De Luca Marco Ludovico Facin Gianluca Maguolo Deep Learning and Handcrafted Features for Virus Image Classification Journal of Imaging virus classification texture descriptors deep learning local binary patterns ensemble of descriptors |
author_facet |
Loris Nanni Eugenio De Luca Marco Ludovico Facin Gianluca Maguolo |
author_sort |
Loris Nanni |
title |
Deep Learning and Handcrafted Features for Virus Image Classification |
title_short |
Deep Learning and Handcrafted Features for Virus Image Classification |
title_full |
Deep Learning and Handcrafted Features for Virus Image Classification |
title_fullStr |
Deep Learning and Handcrafted Features for Virus Image Classification |
title_full_unstemmed |
Deep Learning and Handcrafted Features for Virus Image Classification |
title_sort |
deep learning and handcrafted features for virus image classification |
publisher |
MDPI AG |
series |
Journal of Imaging |
issn |
2313-433X |
publishDate |
2020-12-01 |
description |
In this work, we present an ensemble of descriptors for the classification of virus images acquired using transmission electron microscopy. We trained multiple support vector machines on different sets of features extracted from the data. We used both handcrafted algorithms and a pretrained deep neural network as feature extractors. The proposed fusion strongly boosts the performance obtained by each stand-alone approach, obtaining state of the art performance. |
topic |
virus classification texture descriptors deep learning local binary patterns ensemble of descriptors |
url |
https://www.mdpi.com/2313-433X/6/12/143 |
work_keys_str_mv |
AT lorisnanni deeplearningandhandcraftedfeaturesforvirusimageclassification AT eugeniodeluca deeplearningandhandcraftedfeaturesforvirusimageclassification AT marcoludovicofacin deeplearningandhandcraftedfeaturesforvirusimageclassification AT gianlucamaguolo deeplearningandhandcraftedfeaturesforvirusimageclassification |
_version_ |
1724374650595573760 |