Estimating melon yield for breeding processes by machine-vision processing of UAV images

A. Kalantar; A. Dashuta; Y. Edan; A. Dafna; A. Gur; I. Klapp

doi:10.3920/978-90-8686-888-9_47

Estimating melon yield for breeding processes by machine-vision processing of UAV images

A. Kalantar, A. Dashuta, Y. Edan, A. Dafna, A. Gur, I. Klapp

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

4 Scopus citations

Abstract

Monitoring plants, for yield estimation in melon breeding, is a highly labor-intensive task. An algorithmic pipeline for detection and yield estimation of melons from top-view images of a melon's field is presented. The pipeline developed at the individual melon level includes three main stages: melon recognition, feature extraction, and yield estimation. For each region of interest classified as a melon, the melon features were extracted by fitting an ellipse to the melon contour. A regression model that ties the ellipse features to the melon's weight is presented. The modified R² value of the regression model was 0.94. Comparing yield estimation to ground truth, the average estimation error was 16%. The yield accuracy is highly dependent on the ellipse estimation accuracy, with promising results of only 4% error for the best ellipse-fitted melons.

Original language	English
Title of host publication	Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019
Editors	John V. Stafford
Publisher	Wageningen Academic Publishers
Pages	381-387
Number of pages	7
ISBN (Electronic)	9789086863372
DOIs	https://doi.org/10.3920/978-90-8686-888-9_47
State	Published - 2019
Externally published	Yes
Event	12th European Conference on Precision Agriculture, ECPA 2019 - Montpellier, France Duration: 8 Jul 2019 → 11 Jul 2019

Publication series

Name	Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019

Conference

Conference	12th European Conference on Precision Agriculture, ECPA 2019
Country/Territory	France
City	Montpellier
Period	8/07/19 → 11/07/19

Bibliographical note

Publisher Copyright:
© Wageningen Academic Publishers 2019

Funding

This work was partially supported by BARD Program number IS-4911-16 and by the Rabbi W. Gunther Plaut Chair in Manufacturing Engineering at Ben-Gurion University of the Negev.

Funders	Funder number
Rabbi W. Gunther Plaut Chair
United States - Israel Binational Agricultural Research and Development Fund	IS-4911-16
Ben-Gurion University of the Negev

Keywords

Active contour
Breeding
CNN
Machine learning
Melon
Phenotyping
Yield estimation

Access to Document

10.3920/978-90-8686-888-9_47

Cite this

Kalantar, A., Dashuta, A., Edan, Y., Dafna, A., Gur, A., & Klapp, I. (2019). Estimating melon yield for breeding processes by machine-vision processing of UAV images. In J. V. Stafford (Ed.), Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019 (pp. 381-387). (Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019). Wageningen Academic Publishers. https://doi.org/10.3920/978-90-8686-888-9_47

Kalantar, A. ; Dashuta, A. ; Edan, Y. et al. / Estimating melon yield for breeding processes by machine-vision processing of UAV images. Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019. editor / John V. Stafford. Wageningen Academic Publishers, 2019. pp. 381-387 (Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019).

@inproceedings{106380863ad24cbd9cbc64eddbbb0048,

title = "Estimating melon yield for breeding processes by machine-vision processing of UAV images",

abstract = "Monitoring plants, for yield estimation in melon breeding, is a highly labor-intensive task. An algorithmic pipeline for detection and yield estimation of melons from top-view images of a melon's field is presented. The pipeline developed at the individual melon level includes three main stages: melon recognition, feature extraction, and yield estimation. For each region of interest classified as a melon, the melon features were extracted by fitting an ellipse to the melon contour. A regression model that ties the ellipse features to the melon's weight is presented. The modified R2 value of the regression model was 0.94. Comparing yield estimation to ground truth, the average estimation error was 16%. The yield accuracy is highly dependent on the ellipse estimation accuracy, with promising results of only 4% error for the best ellipse-fitted melons.",

keywords = "Active contour, Breeding, CNN, Machine learning, Melon, Phenotyping, Yield estimation",

author = "A. Kalantar and A. Dashuta and Y. Edan and A. Dafna and A. Gur and I. Klapp",

note = "Publisher Copyright: {\textcopyright} Wageningen Academic Publishers 2019; 12th European Conference on Precision Agriculture, ECPA 2019 ; Conference date: 08-07-2019 Through 11-07-2019",

year = "2019",

doi = "10.3920/978-90-8686-888-9_47",

language = "אנגלית",

series = "Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019",

publisher = "Wageningen Academic Publishers",

pages = "381--387",

editor = "Stafford, {John V.}",

booktitle = "Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019",

address = "הולנד",

}

Kalantar, A, Dashuta, A, Edan, Y, Dafna, A, Gur, A & Klapp, I 2019, Estimating melon yield for breeding processes by machine-vision processing of UAV images. in JV Stafford (ed.), Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019. Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019, Wageningen Academic Publishers, pp. 381-387, 12th European Conference on Precision Agriculture, ECPA 2019, Montpellier, France, 8/07/19. https://doi.org/10.3920/978-90-8686-888-9_47

Estimating melon yield for breeding processes by machine-vision processing of UAV images. / Kalantar, A.; Dashuta, A.; Edan, Y. et al.
Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019. ed. / John V. Stafford. Wageningen Academic Publishers, 2019. p. 381-387 (Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019).

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

TY - GEN

T1 - Estimating melon yield for breeding processes by machine-vision processing of UAV images

AU - Kalantar, A.

AU - Dashuta, A.

AU - Edan, Y.

AU - Dafna, A.

AU - Gur, A.

AU - Klapp, I.

N1 - Publisher Copyright: © Wageningen Academic Publishers 2019

PY - 2019

Y1 - 2019

N2 - Monitoring plants, for yield estimation in melon breeding, is a highly labor-intensive task. An algorithmic pipeline for detection and yield estimation of melons from top-view images of a melon's field is presented. The pipeline developed at the individual melon level includes three main stages: melon recognition, feature extraction, and yield estimation. For each region of interest classified as a melon, the melon features were extracted by fitting an ellipse to the melon contour. A regression model that ties the ellipse features to the melon's weight is presented. The modified R2 value of the regression model was 0.94. Comparing yield estimation to ground truth, the average estimation error was 16%. The yield accuracy is highly dependent on the ellipse estimation accuracy, with promising results of only 4% error for the best ellipse-fitted melons.

AB - Monitoring plants, for yield estimation in melon breeding, is a highly labor-intensive task. An algorithmic pipeline for detection and yield estimation of melons from top-view images of a melon's field is presented. The pipeline developed at the individual melon level includes three main stages: melon recognition, feature extraction, and yield estimation. For each region of interest classified as a melon, the melon features were extracted by fitting an ellipse to the melon contour. A regression model that ties the ellipse features to the melon's weight is presented. The modified R2 value of the regression model was 0.94. Comparing yield estimation to ground truth, the average estimation error was 16%. The yield accuracy is highly dependent on the ellipse estimation accuracy, with promising results of only 4% error for the best ellipse-fitted melons.

KW - Active contour

KW - Breeding

KW - CNN

KW - Machine learning

KW - Melon

KW - Phenotyping

KW - Yield estimation

UR - http://www.scopus.com/inward/record.url?scp=85066492655&partnerID=8YFLogxK

U2 - 10.3920/978-90-8686-888-9_47

DO - 10.3920/978-90-8686-888-9_47

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85066492655

T3 - Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019

SP - 381

EP - 387

BT - Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019

A2 - Stafford, John V.

PB - Wageningen Academic Publishers

T2 - 12th European Conference on Precision Agriculture, ECPA 2019

Y2 - 8 July 2019 through 11 July 2019

ER -

Kalantar A, Dashuta A, Edan Y, Dafna A, Gur A, Klapp I. Estimating melon yield for breeding processes by machine-vision processing of UAV images. In Stafford JV, editor, Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019. Wageningen Academic Publishers. 2019. p. 381-387. (Precision Agriculture 2019 - Papers Presented at the 12th European Conference on Precision Agriculture, ECPA 2019). doi: 10.3920/978-90-8686-888-9_47

Estimating melon yield for breeding processes by machine-vision processing of UAV images

Abstract

Publication series

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this