Machine Learning Approaches for Crop Identification from Remote Sensing Imagery: A Review

Falin Wu; Reza Maleki; Amel Oubara; Dimaris Gómez; Akram Eftekhari; Gongliu Yang

doi:10.1007/978-3-031-27524-1_31

Machine Learning Approaches for Crop Identification from Remote Sensing Imagery: A Review

Falin Wu
, Reza Maleki^*
, Amel Oubara
, Dimaris Gómez
, Akram Eftekhari
, Gongliu Yang

^*Corresponding author for this work

School of Instrumentation and Optoelectronic Engineering

Beihang University
University of Tehran

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

4 Scopus citations

Abstract

Crop area mapping as well as other remote sensing - based agricultural applications are vital for food security and production. This review investigates several studies published in two last years which focus on crop identification and classification using machine learning methods and remote sensing data. The popular conventional machine learning algorithms in crop classification are RF, SVM and LDA, and for deep learning algorithms are MLP, CNN and several different architectures of CNN include AlexNet, ResNet, Google-net, ResNet and U-net. RNN variations such as LSTM model are well-known deep learning architectures for processing time-series remote sensing data. Transfer learning builds a local classification algorithm to a new destination with a limited number of training samples. According to our investigation, the most widely used TL methods in recent researches are U-net (97% OA) and RF (90% OA) in agriculture scopes which have the best overall accuracies among the other methods. Also, more than 40 types of crop have been extracted in various researches, the main of which are corn, rice, wheat, soybeans, accounting for nearly 50% of all researches.

Original language	English
Title of host publication	Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022
Editors	Ajith Abraham, Ajith Abraham, Niketa Gandhi, Thomas Hanne, Pooja Manghirmalani Mishra, Anu Bajaj, Patrick Siarry
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	325-336
Number of pages	12
ISBN (Print)	9783031275234
DOIs	https://doi.org/10.1007/978-3-031-27524-1_31
State	Published - 2023
Event	14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022, and the 14th World Congress on Nature and Biologically Inspired Computing, NaBIC 2022 - Virtual, Online Duration: 14 Dec 2022 → 16 Dec 2022

Publication series

Name	Lecture Notes in Networks and Systems
Volume	648 LNNS
ISSN (Print)	2367-3370
ISSN (Electronic)	2367-3389

Conference

Conference	14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022, and the 14th World Congress on Nature and Biologically Inspired Computing, NaBIC 2022
City	Virtual, Online
Period	14/12/22 → 16/12/22

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 2 Zero Hunger

Keywords

Classification
Crop identification
Deep learning
Food security
Machine learning
Remote sensing
Transfer learning

Access to Document

10.1007/978-3-031-27524-1_31

Cite this

Wu, F., Maleki, R., Oubara, A., Gómez, D., Eftekhari, A., & Yang, G. (2023). Machine Learning Approaches for Crop Identification from Remote Sensing Imagery: A Review. In A. Abraham, A. Abraham, N. Gandhi, T. Hanne, P. Manghirmalani Mishra, A. Bajaj, & P. Siarry (Eds.), Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022 (pp. 325-336). (Lecture Notes in Networks and Systems; Vol. 648 LNNS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-27524-1_31

Wu, Falin ; Maleki, Reza ; Oubara, Amel et al. / Machine Learning Approaches for Crop Identification from Remote Sensing Imagery : A Review. Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022. editor / Ajith Abraham ; Ajith Abraham ; Niketa Gandhi ; Thomas Hanne ; Pooja Manghirmalani Mishra ; Anu Bajaj ; Patrick Siarry. Springer Science and Business Media Deutschland GmbH, 2023. pp. 325-336 (Lecture Notes in Networks and Systems).

@inproceedings{1a1ca9d538b74e55ac710e1e0a4d9619,

title = "Machine Learning Approaches for Crop Identification from Remote Sensing Imagery: A Review",

abstract = "Crop area mapping as well as other remote sensing - based agricultural applications are vital for food security and production. This review investigates several studies published in two last years which focus on crop identification and classification using machine learning methods and remote sensing data. The popular conventional machine learning algorithms in crop classification are RF, SVM and LDA, and for deep learning algorithms are MLP, CNN and several different architectures of CNN include AlexNet, ResNet, Google-net, ResNet and U-net. RNN variations such as LSTM model are well-known deep learning architectures for processing time-series remote sensing data. Transfer learning builds a local classification algorithm to a new destination with a limited number of training samples. According to our investigation, the most widely used TL methods in recent researches are U-net (97\% OA) and RF (90\% OA) in agriculture scopes which have the best overall accuracies among the other methods. Also, more than 40 types of crop have been extracted in various researches, the main of which are corn, rice, wheat, soybeans, accounting for nearly 50\% of all researches.",

keywords = "Classification, Crop identification, Deep learning, Food security, Machine learning, Remote sensing, Transfer learning",

author = "Falin Wu and Reza Maleki and Amel Oubara and Dimaris G{\'o}mez and Akram Eftekhari and Gongliu Yang",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022, and the 14th World Congress on Nature and Biologically Inspired Computing, NaBIC 2022 ; Conference date: 14-12-2022 Through 16-12-2022",

year = "2023",

doi = "10.1007/978-3-031-27524-1\_31",

language = "英语",

isbn = "9783031275234",

series = "Lecture Notes in Networks and Systems",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "325--336",

editor = "Ajith Abraham and Ajith Abraham and Niketa Gandhi and Thomas Hanne and \{Manghirmalani Mishra\}, Pooja and Anu Bajaj and Patrick Siarry",

booktitle = "Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022",

address = "德国",

}

Wu, F, Maleki, R, Oubara, A, Gómez, D, Eftekhari, A & Yang, G 2023, Machine Learning Approaches for Crop Identification from Remote Sensing Imagery: A Review. in A Abraham, A Abraham, N Gandhi, T Hanne, P Manghirmalani Mishra, A Bajaj & P Siarry (eds), Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022. Lecture Notes in Networks and Systems, vol. 648 LNNS, Springer Science and Business Media Deutschland GmbH, pp. 325-336, 14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022, and the 14th World Congress on Nature and Biologically Inspired Computing, NaBIC 2022, Virtual, Online, 14/12/22. https://doi.org/10.1007/978-3-031-27524-1_31

Machine Learning Approaches for Crop Identification from Remote Sensing Imagery: A Review. / Wu, Falin; Maleki, Reza; Oubara, Amel et al.
Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022. ed. / Ajith Abraham; Ajith Abraham; Niketa Gandhi; Thomas Hanne; Pooja Manghirmalani Mishra; Anu Bajaj; Patrick Siarry. Springer Science and Business Media Deutschland GmbH, 2023. p. 325-336 (Lecture Notes in Networks and Systems; Vol. 648 LNNS).

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

TY - GEN

T1 - Machine Learning Approaches for Crop Identification from Remote Sensing Imagery

T2 - 14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022, and the 14th World Congress on Nature and Biologically Inspired Computing, NaBIC 2022

AU - Wu, Falin

AU - Maleki, Reza

AU - Oubara, Amel

AU - Gómez, Dimaris

AU - Eftekhari, Akram

AU - Yang, Gongliu

PY - 2023

Y1 - 2023

N2 - Crop area mapping as well as other remote sensing - based agricultural applications are vital for food security and production. This review investigates several studies published in two last years which focus on crop identification and classification using machine learning methods and remote sensing data. The popular conventional machine learning algorithms in crop classification are RF, SVM and LDA, and for deep learning algorithms are MLP, CNN and several different architectures of CNN include AlexNet, ResNet, Google-net, ResNet and U-net. RNN variations such as LSTM model are well-known deep learning architectures for processing time-series remote sensing data. Transfer learning builds a local classification algorithm to a new destination with a limited number of training samples. According to our investigation, the most widely used TL methods in recent researches are U-net (97% OA) and RF (90% OA) in agriculture scopes which have the best overall accuracies among the other methods. Also, more than 40 types of crop have been extracted in various researches, the main of which are corn, rice, wheat, soybeans, accounting for nearly 50% of all researches.

AB - Crop area mapping as well as other remote sensing - based agricultural applications are vital for food security and production. This review investigates several studies published in two last years which focus on crop identification and classification using machine learning methods and remote sensing data. The popular conventional machine learning algorithms in crop classification are RF, SVM and LDA, and for deep learning algorithms are MLP, CNN and several different architectures of CNN include AlexNet, ResNet, Google-net, ResNet and U-net. RNN variations such as LSTM model are well-known deep learning architectures for processing time-series remote sensing data. Transfer learning builds a local classification algorithm to a new destination with a limited number of training samples. According to our investigation, the most widely used TL methods in recent researches are U-net (97% OA) and RF (90% OA) in agriculture scopes which have the best overall accuracies among the other methods. Also, more than 40 types of crop have been extracted in various researches, the main of which are corn, rice, wheat, soybeans, accounting for nearly 50% of all researches.

KW - Classification

KW - Crop identification

KW - Deep learning

KW - Food security

KW - Machine learning

KW - Remote sensing

KW - Transfer learning

UR - https://www.scopus.com/pages/publications/85152552911

U2 - 10.1007/978-3-031-27524-1_31

DO - 10.1007/978-3-031-27524-1_31

M3 - 会议稿件

AN - SCOPUS:85152552911

SN - 9783031275234

T3 - Lecture Notes in Networks and Systems

SP - 325

EP - 336

BT - Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022

A2 - Abraham, Ajith

A2 - Gandhi, Niketa

A2 - Hanne, Thomas

A2 - Manghirmalani Mishra, Pooja

A2 - Bajaj, Anu

A2 - Siarry, Patrick

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 14 December 2022 through 16 December 2022

ER -

Wu F, Maleki R, Oubara A, Gómez D, Eftekhari A, Yang G. Machine Learning Approaches for Crop Identification from Remote Sensing Imagery: A Review. In Abraham A, Abraham A, Gandhi N, Hanne T, Manghirmalani Mishra P, Bajaj A, Siarry P, editors, Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2022. Springer Science and Business Media Deutschland GmbH. 2023. p. 325-336. (Lecture Notes in Networks and Systems). doi: 10.1007/978-3-031-27524-1_31