Comparing local descriptors and bags of visualwords to deep convolutional neural networks for plant recognition

Pornntiwa Pawara; Emmanuel Okafor; Olarik Surinta; Lambert Schomaker; Marco Wiering

doi:10.5220/0006196204790486

Comparing local descriptors and bags of visualwords to deep convolutional neural networks for plant recognition

Pornntiwa Pawara
, Emmanuel Okafor
, Olarik Surinta
, Lambert Schomaker
, Marco Wiering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

102 Scopus citations

Abstract

The use of machine learning and computer vision methods for recognizing different plants from images has attracted lots of attention from the community. This paper aims at comparing local feature descriptors and bags of visual words with different classifiers to deep convolutional neural networks (CNNs) on three plant datasets; AgrilPlant, LeafSnap, and Folio. To achieve this, we study the use of both scratch and fine-tuned versions of the GoogleNet and the AlexNet architectures and compare them to a local feature descriptor with k-nearest neighbors and the bag of visual words with the histogram of oriented gradients combined with either support vector machines and multi-layer perceptrons. The results shows that the deep CNN methods outperform the hand-crafted features. The CNN techniques can also learn well on a relatively small dataset, Folio.

Original language	English
Title of host publication	ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods
Editors	Maria De De Marsico, Gabriella Sanniti di Baja, Ana Fred
Publisher	SciTePress
Pages	479-486
Number of pages	8
ISBN (Electronic)	9789897582226
DOIs	https://doi.org/10.5220/0006196204790486
State	Published - 2017
Externally published	Yes
Event	6th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2017 - Porto, Portugal Duration: 24 Feb 2017 → 26 Feb 2017

Publication series

Name	ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods
Volume	2017-January

Conference

Conference	6th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2017
Country/Territory	Portugal
City	Porto
Period	24/02/17 → 26/02/17

Bibliographical note

Publisher Copyright:
Copyright © 2017 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.

Keywords

Bags of visual words
Convolutional neural network
Deep learning
Local descriptor
Plant classification

ASJC Scopus subject areas

Computer Vision and Pattern Recognition

Access to Document

10.5220/0006196204790486

Cite this

Pawara, P., Okafor, E., Surinta, O., Schomaker, L., & Wiering, M. (2017). Comparing local descriptors and bags of visualwords to deep convolutional neural networks for plant recognition. In M. D. De Marsico, G. S. di Baja, & A. Fred (Eds.), ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods (pp. 479-486). (ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods; Vol. 2017-January). SciTePress. https://doi.org/10.5220/0006196204790486

Pawara, Pornntiwa ; Okafor, Emmanuel ; Surinta, Olarik et al. / Comparing local descriptors and bags of visualwords to deep convolutional neural networks for plant recognition. ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods. editor / Maria De De Marsico ; Gabriella Sanniti di Baja ; Ana Fred. SciTePress, 2017. pp. 479-486 (ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods).

@inproceedings{9a87d9842645403d89d361c343a49dcc,

title = "Comparing local descriptors and bags of visualwords to deep convolutional neural networks for plant recognition",

abstract = "The use of machine learning and computer vision methods for recognizing different plants from images has attracted lots of attention from the community. This paper aims at comparing local feature descriptors and bags of visual words with different classifiers to deep convolutional neural networks (CNNs) on three plant datasets; AgrilPlant, LeafSnap, and Folio. To achieve this, we study the use of both scratch and fine-tuned versions of the GoogleNet and the AlexNet architectures and compare them to a local feature descriptor with k-nearest neighbors and the bag of visual words with the histogram of oriented gradients combined with either support vector machines and multi-layer perceptrons. The results shows that the deep CNN methods outperform the hand-crafted features. The CNN techniques can also learn well on a relatively small dataset, Folio.",

keywords = "Bags of visual words, Convolutional neural network, Deep learning, Local descriptor, Plant classification",

author = "Pornntiwa Pawara and Emmanuel Okafor and Olarik Surinta and Lambert Schomaker and Marco Wiering",

note = "Publisher Copyright: Copyright {\textcopyright} 2017 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.; 6th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2017 ; Conference date: 24-02-2017 Through 26-02-2017",

year = "2017",

doi = "10.5220/0006196204790486",

language = "English",

series = "ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods",

publisher = "SciTePress",

pages = "479--486",

editor = "\{De Marsico\}, \{Maria De\} and \{di Baja\}, \{Gabriella Sanniti\} and Ana Fred",

booktitle = "ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods",

}

Pawara, P, Okafor, E, Surinta, O, Schomaker, L & Wiering, M 2017, Comparing local descriptors and bags of visualwords to deep convolutional neural networks for plant recognition. in MD De Marsico, GS di Baja & A Fred (eds), ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods. ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods, vol. 2017-January, SciTePress, pp. 479-486, 6th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2017, Porto, Portugal, 24/02/17. https://doi.org/10.5220/0006196204790486

Comparing local descriptors and bags of visualwords to deep convolutional neural networks for plant recognition. / Pawara, Pornntiwa; Okafor, Emmanuel; Surinta, Olarik et al.
ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods. ed. / Maria De De Marsico; Gabriella Sanniti di Baja; Ana Fred. SciTePress, 2017. p. 479-486 (ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods; Vol. 2017-January).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Comparing local descriptors and bags of visualwords to deep convolutional neural networks for plant recognition

AU - Pawara, Pornntiwa

AU - Okafor, Emmanuel

AU - Surinta, Olarik

AU - Schomaker, Lambert

AU - Wiering, Marco

PY - 2017

Y1 - 2017

N2 - The use of machine learning and computer vision methods for recognizing different plants from images has attracted lots of attention from the community. This paper aims at comparing local feature descriptors and bags of visual words with different classifiers to deep convolutional neural networks (CNNs) on three plant datasets; AgrilPlant, LeafSnap, and Folio. To achieve this, we study the use of both scratch and fine-tuned versions of the GoogleNet and the AlexNet architectures and compare them to a local feature descriptor with k-nearest neighbors and the bag of visual words with the histogram of oriented gradients combined with either support vector machines and multi-layer perceptrons. The results shows that the deep CNN methods outperform the hand-crafted features. The CNN techniques can also learn well on a relatively small dataset, Folio.

AB - The use of machine learning and computer vision methods for recognizing different plants from images has attracted lots of attention from the community. This paper aims at comparing local feature descriptors and bags of visual words with different classifiers to deep convolutional neural networks (CNNs) on three plant datasets; AgrilPlant, LeafSnap, and Folio. To achieve this, we study the use of both scratch and fine-tuned versions of the GoogleNet and the AlexNet architectures and compare them to a local feature descriptor with k-nearest neighbors and the bag of visual words with the histogram of oriented gradients combined with either support vector machines and multi-layer perceptrons. The results shows that the deep CNN methods outperform the hand-crafted features. The CNN techniques can also learn well on a relatively small dataset, Folio.

KW - Bags of visual words

KW - Convolutional neural network

KW - Deep learning

KW - Local descriptor

KW - Plant classification

UR - https://www.scopus.com/pages/publications/85046347005

U2 - 10.5220/0006196204790486

DO - 10.5220/0006196204790486

M3 - Conference contribution

AN - SCOPUS:85046347005

T3 - ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods

SP - 479

EP - 486

BT - ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods

A2 - De Marsico, Maria De

A2 - di Baja, Gabriella Sanniti

A2 - Fred, Ana

PB - SciTePress

T2 - 6th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2017

Y2 - 24 February 2017 through 26 February 2017

ER -

Pawara P, Okafor E, Surinta O, Schomaker L, Wiering M. Comparing local descriptors and bags of visualwords to deep convolutional neural networks for plant recognition. In De Marsico MD, di Baja GS, Fred A, editors, ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods. SciTePress. 2017. p. 479-486. (ICPRAM 2017 - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods). doi: 10.5220/0006196204790486

Comparing local descriptors and bags of visualwords to deep convolutional neural networks for plant recognition

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this