Fast Classification of Large Germinated Fields Via High-Resolution UAV Imagery

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Crop breeding consists of the process of editing crop genetic profile for increasing many crop qualities. In order to achieve optimal results, crop breeders have to plant thousands of plants and keep a track of their growth almost daily. This process requires increased man-hour inspection over large fields, which results in poor accuracy due to human fatigue and a time-inefficient strategy. In this letter, two machine vision approaches were compared for classifying three crop germination classes (good, average, and bad). A naive approach using a classical segmentation and an unsupervised learning approach using k-means segmentation were compared within a high-resolution unmanned aerial vehicles imagery dataset. Experimental results demonstrate the classification of germinated patches up to 0.05 m2/patch of resolution with a minimum F1-score of 76% and 80%, and AUC of 95% and 91% for high and low spatial image resolutions, respectively.

LanguageEnglish
Article number8755475
Pages3216-3223
JournalIEEE Robotics and Automation Letters
Volume4
Issue number4
DOIs
Publication statusPublished - Oct 2019

Fingerprint

Unmanned aerial vehicles (UAV)
Crops
Patch
High Resolution
Segmentation
Machine Vision
Unsupervised Learning
Germination
K-means
Fatigue
Inspection
Experimental Results
Unsupervised learning
Demonstrate
Image resolution
Computer vision
Imagery
Fatigue of materials
Strategy
Class

Keywords

  • agro-food robotics
  • crop emergence
  • field assessment
  • machine vision
  • plants breeding
  • Unmanned aerial vehicles

Cite this

@article{39c9a993145c478e9325ea9c917389e9,
title = "Fast Classification of Large Germinated Fields Via High-Resolution UAV Imagery",
abstract = "Crop breeding consists of the process of editing crop genetic profile for increasing many crop qualities. In order to achieve optimal results, crop breeders have to plant thousands of plants and keep a track of their growth almost daily. This process requires increased man-hour inspection over large fields, which results in poor accuracy due to human fatigue and a time-inefficient strategy. In this letter, two machine vision approaches were compared for classifying three crop germination classes (good, average, and bad). A naive approach using a classical segmentation and an unsupervised learning approach using k-means segmentation were compared within a high-resolution unmanned aerial vehicles imagery dataset. Experimental results demonstrate the classification of germinated patches up to 0.05 m2/patch of resolution with a minimum F1-score of 76{\%} and 80{\%}, and AUC of 95{\%} and 91{\%} for high and low spatial image resolutions, respectively.",
keywords = "agro-food robotics, crop emergence, field assessment, machine vision, plants breeding, Unmanned aerial vehicles",
author = "Jo{\~a}o Valente and Lammert Kooistra and Sander M{\"u}cher",
year = "2019",
month = "10",
doi = "10.1109/LRA.2019.2926957",
language = "English",
volume = "4",
pages = "3216--3223",
journal = "IEEE Robotics and Automation Letters",
issn = "2377-3766",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

Fast Classification of Large Germinated Fields Via High-Resolution UAV Imagery. / Valente, João; Kooistra, Lammert; Mücher, Sander.

In: IEEE Robotics and Automation Letters, Vol. 4, No. 4, 8755475, 10.2019, p. 3216-3223.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Fast Classification of Large Germinated Fields Via High-Resolution UAV Imagery

AU - Valente, João

AU - Kooistra, Lammert

AU - Mücher, Sander

PY - 2019/10

Y1 - 2019/10

N2 - Crop breeding consists of the process of editing crop genetic profile for increasing many crop qualities. In order to achieve optimal results, crop breeders have to plant thousands of plants and keep a track of their growth almost daily. This process requires increased man-hour inspection over large fields, which results in poor accuracy due to human fatigue and a time-inefficient strategy. In this letter, two machine vision approaches were compared for classifying three crop germination classes (good, average, and bad). A naive approach using a classical segmentation and an unsupervised learning approach using k-means segmentation were compared within a high-resolution unmanned aerial vehicles imagery dataset. Experimental results demonstrate the classification of germinated patches up to 0.05 m2/patch of resolution with a minimum F1-score of 76% and 80%, and AUC of 95% and 91% for high and low spatial image resolutions, respectively.

AB - Crop breeding consists of the process of editing crop genetic profile for increasing many crop qualities. In order to achieve optimal results, crop breeders have to plant thousands of plants and keep a track of their growth almost daily. This process requires increased man-hour inspection over large fields, which results in poor accuracy due to human fatigue and a time-inefficient strategy. In this letter, two machine vision approaches were compared for classifying three crop germination classes (good, average, and bad). A naive approach using a classical segmentation and an unsupervised learning approach using k-means segmentation were compared within a high-resolution unmanned aerial vehicles imagery dataset. Experimental results demonstrate the classification of germinated patches up to 0.05 m2/patch of resolution with a minimum F1-score of 76% and 80%, and AUC of 95% and 91% for high and low spatial image resolutions, respectively.

KW - agro-food robotics

KW - crop emergence

KW - field assessment

KW - machine vision

KW - plants breeding

KW - Unmanned aerial vehicles

U2 - 10.1109/LRA.2019.2926957

DO - 10.1109/LRA.2019.2926957

M3 - Article

VL - 4

SP - 3216

EP - 3223

JO - IEEE Robotics and Automation Letters

T2 - IEEE Robotics and Automation Letters

JF - IEEE Robotics and Automation Letters

SN - 2377-3766

IS - 4

M1 - 8755475

ER -