Vision-Based Apple Counting and Yield Estimation

Pravakar Roy; Volkan Isler

doi:10.1007/978-3-319-50115-4_42

Vision-Based Apple Counting and Yield Estimation

Pravakar Roy, Volkan Isler

Computer Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Chapter

10 Scopus citations

Abstract

We present a novel method for yield estimation in apple orchards. Our method takes segmented and registered images of apple clusters as input. It outputs number and location of individual apples in each cluster. Our primary technical contributions are a representation based on a mixture of Gaussians, and a novel selection criterion to choose the number of components in the mixture. The method is experimentally verified on four different datasets using images acquired by a vision platform mounted on an aerial robot, a ground vehicle and a hand-held device. The accuracy of the counting algorithm itself is 91 %. It achieves 81–85% accuracy coupled with segmentation and registration which is significantly higher than existing image based methods.

Original language	English (US)
Title of host publication	Springer Proceedings in Advanced Robotics
Publisher	Springer Science and Business Media B.V.
Pages	478-487
Number of pages	10
DOIs	https://doi.org/10.1007/978-3-319-50115-4_42
State	Published - 2017

Publication series

Name	Springer Proceedings in Advanced Robotics
Volume	1
ISSN (Print)	2511-1256
ISSN (Electronic)	2511-1264

Bibliographical note

Funding Information:
Acknowledgments. This work is supported in part by NSF grant # 1317788, USDA NIFA MIN-98-G02 and the MnDrive initiative.

Publisher Copyright:
© 2017, Springer International Publishing AG.

Keywords

Apple Orchard
Expectation Maximization
Gaussian Mixture Model
Greedy Method
Minimum Description Length

Publisher link

10.1007/978-3-319-50115-4_42

Find It

Find It at U of M Libraries

Cite this

@inbook{93d2dc9d9a7b49a2b0421063bfe6219d,

title = "Vision-Based Apple Counting and Yield Estimation",

abstract = "We present a novel method for yield estimation in apple orchards. Our method takes segmented and registered images of apple clusters as input. It outputs number and location of individual apples in each cluster. Our primary technical contributions are a representation based on a mixture of Gaussians, and a novel selection criterion to choose the number of components in the mixture. The method is experimentally verified on four different datasets using images acquired by a vision platform mounted on an aerial robot, a ground vehicle and a hand-held device. The accuracy of the counting algorithm itself is 91 %. It achieves 81–85% accuracy coupled with segmentation and registration which is significantly higher than existing image based methods.",

keywords = "Apple Orchard, Expectation Maximization, Gaussian Mixture Model, Greedy Method, Minimum Description Length",

author = "Pravakar Roy and Volkan Isler",

note = "Funding Information: Acknowledgments. This work is supported in part by NSF grant # 1317788, USDA NIFA MIN-98-G02 and the MnDrive initiative. Publisher Copyright: {\textcopyright} 2017, Springer International Publishing AG.",

year = "2017",

doi = "10.1007/978-3-319-50115-4_42",

language = "English (US)",

series = "Springer Proceedings in Advanced Robotics",

publisher = "Springer Science and Business Media B.V.",

pages = "478--487",

booktitle = "Springer Proceedings in Advanced Robotics",

}

TY - CHAP

T1 - Vision-Based Apple Counting and Yield Estimation

AU - Roy, Pravakar

AU - Isler, Volkan

N1 - Funding Information: Acknowledgments. This work is supported in part by NSF grant # 1317788, USDA NIFA MIN-98-G02 and the MnDrive initiative. Publisher Copyright: © 2017, Springer International Publishing AG.

PY - 2017

Y1 - 2017

N2 - We present a novel method for yield estimation in apple orchards. Our method takes segmented and registered images of apple clusters as input. It outputs number and location of individual apples in each cluster. Our primary technical contributions are a representation based on a mixture of Gaussians, and a novel selection criterion to choose the number of components in the mixture. The method is experimentally verified on four different datasets using images acquired by a vision platform mounted on an aerial robot, a ground vehicle and a hand-held device. The accuracy of the counting algorithm itself is 91 %. It achieves 81–85% accuracy coupled with segmentation and registration which is significantly higher than existing image based methods.

AB - We present a novel method for yield estimation in apple orchards. Our method takes segmented and registered images of apple clusters as input. It outputs number and location of individual apples in each cluster. Our primary technical contributions are a representation based on a mixture of Gaussians, and a novel selection criterion to choose the number of components in the mixture. The method is experimentally verified on four different datasets using images acquired by a vision platform mounted on an aerial robot, a ground vehicle and a hand-held device. The accuracy of the counting algorithm itself is 91 %. It achieves 81–85% accuracy coupled with segmentation and registration which is significantly higher than existing image based methods.

KW - Apple Orchard

KW - Expectation Maximization

KW - Gaussian Mixture Model

KW - Greedy Method

KW - Minimum Description Length

UR - http://www.scopus.com/inward/record.url?scp=85107040674&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85107040674&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-50115-4_42

DO - 10.1007/978-3-319-50115-4_42

M3 - Chapter

AN - SCOPUS:85107040674

T3 - Springer Proceedings in Advanced Robotics

SP - 478

EP - 487

BT - Springer Proceedings in Advanced Robotics

PB - Springer Science and Business Media B.V.

ER -

Vision-Based Apple Counting and Yield Estimation

Abstract

Publication series

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this