Detection of citrus canker and Huanglongbing using fluorescence imaging spectroscopy and support vector machine technique

Caio Bruno Wetterich; Ruan Felipe de Oliveira Neves; José Belasque; Luis Gustavo Marcassa

doi:10.1364/AO.55.000400

Applied Optics
Vol. 55,
Issue 2,
pp. 400-407
(2016)
•https://doi.org/10.1364/AO.55.000400

Detection of citrus canker and Huanglongbing using fluorescence imaging spectroscopy and support vector machine technique

Caio Bruno Wetterich, Ruan Felipe de Oliveira Neves, José Belasque, and Luis Gustavo Marcassa

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Caio Bruno Wetterich, Ruan Felipe de Oliveira Neves, José Belasque, and Luis Gustavo Marcassa, "Detection of citrus canker and Huanglongbing using fluorescence imaging spectroscopy and support vector machine technique," Appl. Opt. 55, 400-407 (2016)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Related Topics
Table of Contents Category
- Spectroscopy
Optics & Photonics Topics
?

The topics in this list come from the Optics and Photonics Topics applied to this article.

About this Article
History
- Original Manuscript: August 14, 2015
- Revised Manuscript: November 12, 2015
- Manuscript Accepted: December 1, 2015
- Published: January 8, 2016
Virtual Issues
Virtual Journal for Biomedical Optics Vol. 11, Iss. 3

Abstract

Citrus canker and Huanglongbing (HLB) are citrus diseases that represent a serious threat to the citrus production worldwide and may cause large economic losses. In this work, we combined fluorescence imaging spectroscopy (FIS) and a machine learning technique to discriminate between these diseases and other ordinary citrus conditions that may be present at citrus orchards, such as citrus scab and zinc deficiency. Our classification results are highly accurate when discriminating citrus canker from citrus scab (97.8%), and HLB from zinc deficiency (95%). These results show that it is possible to accurately identify citrus diseases that present similar symptoms.

Full Article | PDF Article

More Like This

Detection of Huanglongbing in Florida using fluorescence imaging spectroscopy and machine-learning methods

Caio Bruno Wetterich, Ruan Felipe de Oliveira Neves, José Belasque, Reza Ehsani, and Luis Gustavo Marcassa
Appl. Opt. 56(1) 15-23 (2017)

Optical fiber laser induced fluorescence spectroscopy as a citrus canker diagnostic

E. C. Lins, J. Belasque, Jr., and L. G. Marcassa
Appl. Opt. 49(4) 663-667 (2010)

Detection of mechanical and disease stresses in citrus plants by fluorescence spectroscopy

J. Belasque, Jr., M. C. G. Gasparoto, and L. G. Marcassa
Appl. Opt. 47(11) 1922-1926 (2008)

Previous Article Next Article

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (3)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (7)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (9)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Disease	Excitation 405 nm	Excitation 470 nm
Citrus canker and citrus scab (Group 1)	Filters at 550, 580, and 690 nm	Filters at 550, 570, and 690 nm
HLB and zinc deficiency (Group 2)	Filters at 560, 580, and 690 nm	Filters at 530, 550, and 690 nm

			Condition Determined by Assay Value
			Condition Positive	Condition Negative
Excitation 405 nm	F550	Test outcome positive	$TP = 90$	$FP = 34$
	F550	Test outcome negative	$FN = 0$	$TN = 56$
	F580	Test outcome positive	$TP = 90$	$FP = 35$
	F580	Test outcome negative	$FN = 0$	$TN = 55$
	F690	Test outcome positive	$TP = 72$	$FP = 67$
	F690	Test outcome negative	$FN = 18$	$TN = 23$
Excitation 470 nm	F550	Test outcome positive	$TP = 86$	$FP = 41$
	F550	Test outcome negative	$FN = 4$	$TN = 49$
	F570	Test outcome positive	$TP = 90$	$FP = 79$
	F570	Test outcome negative	$FN = 0$	$TN = 11$
	F690	Test outcome positive	$TP = 61$	$FP = 11$
	F690	Test outcome negative	$FN = 29$	$TN = 79$

	Excitation 405 nm			Excitation 470 nm
	F550	F580	F690	F550	F570	F690
Sensitivity	100%	100%	80%	95.6%	100%	67.8%
Specificity	62.2%	61.1%	25.6%	54.4%	12.2%	87.8%
Accuracy	81.1%	80.6%	52.8%	75%	56.1%	77.8%

			Condition Determined by Assay Value
			Condition Positive	Condition Negative
Excitation 405 nm	F560	Test outcome positive	$TP = 81$	$FP = 93$
	F560	Test outcome negative	$FN = 0$	$TN = 6$
	F580	Test outcome positive	$TP = 0$	$FP = 0$
	F580	Test outcome negative	$FN = 81$	$TN = 99$
	F690	Test outcome positive	$TP = 6$	$FP = 0$
	F690	Test outcome negative	$FN = 75$	$TN = 99$
Excitation 470 nm	F530	Test outcome positive	$TP = 73$	$FP = 76$
	F530	Test outcome negative	$FN = 8$	$TN = 23$
	F550	Test outcome positive	$TP = 1$	$FP = 0$
	F550	Test outcome negative	$FN = 80$	$TN = 99$
	F690	Test outcome positive	$TP = 46$	$FP = 16$
	F690	Test outcome negative	$FN = 35$	$TN = 83$

	Excitation 405 nm			Excitation 470 nm
	F560	F580	F690	F530	F550	F690
Sensitivity	100%	0%	7.4%	90.1%	1.2%	56.8%
Specificity	6.1%	100%	100%	23.2%	100%	83.8%
Accuracy	48.3%	55%	58.3%	53.3%	55.6%	71.7%

	Excitation 405 nm		Excitation 470 nm		Combination of the Two Excitations
	F550 + F580	F550 + F580 + F690	F550 + F690	F550 + F570 + F690	All Fluorescence
Sensitivity	100%	100%	96.7%	100%	100%
Specificity	68.9%	76.7%	93.3%	93.3%	95.6%
Accuracy	84.4%	88.3%	95%	96.7%	97.8%

	Excitation 405 nm		Excitation 470 nm		Combination of the Two Excitations
	F580 + F690	F560 + F580 + F690	F550 + F690	F530 + F550 + F690	All Fluorescence
Sensitivity	7.4%	100%	59.7%	96.3%	100%
Specificity	100%	90.9%	100%	90.9%	90.9%
Accuracy	58.3%	95%	80.6%	93.3%	95%

Abstract

Cited By

Figures (3)

Tables (7)

Equations (9)

Applied Optics