Vision-based preharvest yield mapping for apple orchards

Pravakar Roy; Abhijeet Kislay; Patrick A. Plonski; James Luby; Volkan Isler

doi:10.1016/j.compag.2019.104897

Vision-based preharvest yield mapping for apple orchards

Pravakar Roy, Abhijeet Kislay, Patrick A. Plonski, James Luby, Volkan Isler

Research output: Contribution to journal › Article › peer-review

34 Scopus citations

Abstract

We present an end-to-end computer vision system for yield mapping in apple orchards. Our proposed system is platform independent and does not require any specific lighting conditions. Our main technical contributions are (1) a semi-supervised clustering algorithm that utilizes colors to identify apples and (2) an unsupervised clustering method that utilizes spatial properties to estimate fruit counts from apple clusters having arbitrarily complex geometry. Additionally, we utilize camera motion to merge the counts across multiple views. We verified the performance of our algorithms by conducting multiple field trials. Results indicate that the detection method achieves F₁-measure.95–.97 for multiple color varieties and lighting conditions. The counting method achieves an accuracy of 89–98%. Additionally, we report merged fruit counts from both sides of the tree rows. Our yield estimation method achieves an overall accuracy of 91.98–94.81% across different datasets.

Original language	English (US)
Article number	104897
Journal	Computers and Electronics in Agriculture
Volume	164
DOIs	https://doi.org/10.1016/j.compag.2019.104897
State	Published - Sep 2019

Bibliographical note

Funding Information:
The authors thank Joshua Anderson, Professors Emily Hoover, and Cindy Tong from the Department of Horticultural Science, University of Minnesota, for their expertise and help with the experiments. This work is supported in part by NSF grant # 1317788, USDA NIFA MIN-98-G02 and the MnDrive initiative.

Funding Information:
The authors thank Joshua Anderson, Professors Emily Hoover, and Cindy Tong from the Department of Horticultural Science, University of Minnesota, for their expertise and help with the experiments. This work is supported in part by NSF grant # 1317788 , USDA NIFA MIN-98-G02 and the MnDrive initiative.

Publisher Copyright:
© 2019 Elsevier B.V.

Keywords

Apple counting
Apple detection
Clustering
Machine vision
Semi-supervised image segmentation
Yield estimation

Access

10.1016/j.compag.2019.104897

OpenUrl availability

Full text

Cite this

@article{b198d5e796524d78b4596ff3481a5153,

title = "Vision-based preharvest yield mapping for apple orchards",

abstract = "We present an end-to-end computer vision system for yield mapping in apple orchards. Our proposed system is platform independent and does not require any specific lighting conditions. Our main technical contributions are (1) a semi-supervised clustering algorithm that utilizes colors to identify apples and (2) an unsupervised clustering method that utilizes spatial properties to estimate fruit counts from apple clusters having arbitrarily complex geometry. Additionally, we utilize camera motion to merge the counts across multiple views. We verified the performance of our algorithms by conducting multiple field trials. Results indicate that the detection method achieves F1-measure.95–.97 for multiple color varieties and lighting conditions. The counting method achieves an accuracy of 89–98%. Additionally, we report merged fruit counts from both sides of the tree rows. Our yield estimation method achieves an overall accuracy of 91.98–94.81% across different datasets.",

keywords = "Apple counting, Apple detection, Clustering, Machine vision, Semi-supervised image segmentation, Yield estimation",

author = "Pravakar Roy and Abhijeet Kislay and Plonski, {Patrick A.} and James Luby and Volkan Isler",

note = "Funding Information: The authors thank Joshua Anderson, Professors Emily Hoover, and Cindy Tong from the Department of Horticultural Science, University of Minnesota, for their expertise and help with the experiments. This work is supported in part by NSF grant # 1317788, USDA NIFA MIN-98-G02 and the MnDrive initiative. Funding Information: The authors thank Joshua Anderson, Professors Emily Hoover, and Cindy Tong from the Department of Horticultural Science, University of Minnesota, for their expertise and help with the experiments. This work is supported in part by NSF grant # 1317788 , USDA NIFA MIN-98-G02 and the MnDrive initiative. Publisher Copyright: {\textcopyright} 2019 Elsevier B.V.",

year = "2019",

month = sep,

doi = "10.1016/j.compag.2019.104897",

language = "English (US)",

volume = "164",

journal = "Computers and Electronics in Agriculture",

issn = "0168-1699",

publisher = "Elsevier",

}

TY - JOUR

T1 - Vision-based preharvest yield mapping for apple orchards

AU - Roy, Pravakar

AU - Kislay, Abhijeet

AU - Plonski, Patrick A.

AU - Luby, James

AU - Isler, Volkan

N1 - Funding Information: The authors thank Joshua Anderson, Professors Emily Hoover, and Cindy Tong from the Department of Horticultural Science, University of Minnesota, for their expertise and help with the experiments. This work is supported in part by NSF grant # 1317788, USDA NIFA MIN-98-G02 and the MnDrive initiative. Funding Information: The authors thank Joshua Anderson, Professors Emily Hoover, and Cindy Tong from the Department of Horticultural Science, University of Minnesota, for their expertise and help with the experiments. This work is supported in part by NSF grant # 1317788 , USDA NIFA MIN-98-G02 and the MnDrive initiative. Publisher Copyright: © 2019 Elsevier B.V.

PY - 2019/9

Y1 - 2019/9

N2 - We present an end-to-end computer vision system for yield mapping in apple orchards. Our proposed system is platform independent and does not require any specific lighting conditions. Our main technical contributions are (1) a semi-supervised clustering algorithm that utilizes colors to identify apples and (2) an unsupervised clustering method that utilizes spatial properties to estimate fruit counts from apple clusters having arbitrarily complex geometry. Additionally, we utilize camera motion to merge the counts across multiple views. We verified the performance of our algorithms by conducting multiple field trials. Results indicate that the detection method achieves F1-measure.95–.97 for multiple color varieties and lighting conditions. The counting method achieves an accuracy of 89–98%. Additionally, we report merged fruit counts from both sides of the tree rows. Our yield estimation method achieves an overall accuracy of 91.98–94.81% across different datasets.

AB - We present an end-to-end computer vision system for yield mapping in apple orchards. Our proposed system is platform independent and does not require any specific lighting conditions. Our main technical contributions are (1) a semi-supervised clustering algorithm that utilizes colors to identify apples and (2) an unsupervised clustering method that utilizes spatial properties to estimate fruit counts from apple clusters having arbitrarily complex geometry. Additionally, we utilize camera motion to merge the counts across multiple views. We verified the performance of our algorithms by conducting multiple field trials. Results indicate that the detection method achieves F1-measure.95–.97 for multiple color varieties and lighting conditions. The counting method achieves an accuracy of 89–98%. Additionally, we report merged fruit counts from both sides of the tree rows. Our yield estimation method achieves an overall accuracy of 91.98–94.81% across different datasets.

KW - Apple counting

KW - Apple detection

KW - Clustering

KW - Machine vision

KW - Semi-supervised image segmentation

KW - Yield estimation

UR - http://www.scopus.com/inward/record.url?scp=85069630985&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069630985&partnerID=8YFLogxK

U2 - 10.1016/j.compag.2019.104897

DO - 10.1016/j.compag.2019.104897

M3 - Article

AN - SCOPUS:85069630985

SN - 0168-1699

VL - 164

JO - Computers and Electronics in Agriculture

JF - Computers and Electronics in Agriculture

M1 - 104897

ER -

Vision-based preharvest yield mapping for apple orchards

Abstract

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this