A method for automatic segmentation and splitting of hyperspectral images of raspberry plants collected in field conditions

Dominic Williams, Avril Britten, Susan McCallum, Hamlyn Jones, Matt Aitkenhead, Alison Karley, Ken Loades, Ankush Prashar, Julie Graham

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Hyperspectral imaging is a technology that can be used to monitor plant responses to stress. Hyperspectral images have a full spectrum for each pixel in the image, 400-2500 nm in this case, giving detailed information about the spectral reflectance of the plant. Although this technology has been used in laboratory-based controlled lighting conditions for early detection of plant disease, the transfer of such technology to imaging plants in field conditions presents a number of challenges. These include problems caused by varying light levels and difficulties of separating the target plant from its background. Here we present an automated method that has been developed to segment raspberry plants from the background using a selected spectral ratio combined with edge detection. Graph theory was used to minimise a cost function to detect the continuous boundary between uninteresting plants and the area of interest. The method includes automatic detection of a known reflectance tile which was kept constantly within the field of view for all image scans. A method to split images containing rows of multiple raspberry plants into individual plants was also developed. Validation was carried out by comparison of plant height and density measurements with manually scored values. A reasonable correlation was found between these manual scores and measurements taken from the images (r2 = 0.75 for plant height). These preliminary steps are an essential requirement before detailed spectral analysis of the plants can be achieved.

Original languageEnglish
Article number74
JournalPlant Methods
Volume13
Issue number1
DOIs
Publication statusPublished - 1 Nov 2017

Fingerprint

hyperspectral imagery
raspberries
methodology
reflectance
Rubus
image analysis
automatic detection
technology transfer
Technology
Technology Transfer
Plant Diseases
tiles
plant diseases and disorders
spectral analysis
Lighting
plant response
lighting
stress response
Early Diagnosis

Cite this

Williams, Dominic ; Britten, Avril ; McCallum, Susan ; Jones, Hamlyn ; Aitkenhead, Matt ; Karley, Alison ; Loades, Ken ; Prashar, Ankush ; Graham, Julie. / A method for automatic segmentation and splitting of hyperspectral images of raspberry plants collected in field conditions. In: Plant Methods. 2017 ; Vol. 13, No. 1.
@article{dbcd7a4e132c4821b289c43bb2f375db,
title = "A method for automatic segmentation and splitting of hyperspectral images of raspberry plants collected in field conditions",
abstract = "Hyperspectral imaging is a technology that can be used to monitor plant responses to stress. Hyperspectral images have a full spectrum for each pixel in the image, 400-2500 nm in this case, giving detailed information about the spectral reflectance of the plant. Although this technology has been used in laboratory-based controlled lighting conditions for early detection of plant disease, the transfer of such technology to imaging plants in field conditions presents a number of challenges. These include problems caused by varying light levels and difficulties of separating the target plant from its background. Here we present an automated method that has been developed to segment raspberry plants from the background using a selected spectral ratio combined with edge detection. Graph theory was used to minimise a cost function to detect the continuous boundary between uninteresting plants and the area of interest. The method includes automatic detection of a known reflectance tile which was kept constantly within the field of view for all image scans. A method to split images containing rows of multiple raspberry plants into individual plants was also developed. Validation was carried out by comparison of plant height and density measurements with manually scored values. A reasonable correlation was found between these manual scores and measurements taken from the images (r2 = 0.75 for plant height). These preliminary steps are an essential requirement before detailed spectral analysis of the plants can be achieved.",
keywords = "Field imaging, Hyperspectral imaging, Image segmentation, Phenotyping, Raspberry",
author = "Dominic Williams and Avril Britten and Susan McCallum and Hamlyn Jones and Matt Aitkenhead and Alison Karley and Ken Loades and Ankush Prashar and Julie Graham",
year = "2017",
month = "11",
day = "1",
doi = "10.1186/s13007-017-0226-y",
language = "English",
volume = "13",
journal = "Plant Methods",
issn = "1746-4811",
publisher = "BioMed Central",
number = "1",

}

Williams, D, Britten, A, McCallum, S, Jones, H, Aitkenhead, M, Karley, A, Loades, K, Prashar, A & Graham, J 2017, 'A method for automatic segmentation and splitting of hyperspectral images of raspberry plants collected in field conditions' Plant Methods, vol. 13, no. 1, 74. https://doi.org/10.1186/s13007-017-0226-y

A method for automatic segmentation and splitting of hyperspectral images of raspberry plants collected in field conditions. / Williams, Dominic; Britten, Avril; McCallum, Susan; Jones, Hamlyn; Aitkenhead, Matt; Karley, Alison; Loades, Ken; Prashar, Ankush; Graham, Julie.

In: Plant Methods, Vol. 13, No. 1, 74, 01.11.2017.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A method for automatic segmentation and splitting of hyperspectral images of raspberry plants collected in field conditions

AU - Williams, Dominic

AU - Britten, Avril

AU - McCallum, Susan

AU - Jones, Hamlyn

AU - Aitkenhead, Matt

AU - Karley, Alison

AU - Loades, Ken

AU - Prashar, Ankush

AU - Graham, Julie

PY - 2017/11/1

Y1 - 2017/11/1

N2 - Hyperspectral imaging is a technology that can be used to monitor plant responses to stress. Hyperspectral images have a full spectrum for each pixel in the image, 400-2500 nm in this case, giving detailed information about the spectral reflectance of the plant. Although this technology has been used in laboratory-based controlled lighting conditions for early detection of plant disease, the transfer of such technology to imaging plants in field conditions presents a number of challenges. These include problems caused by varying light levels and difficulties of separating the target plant from its background. Here we present an automated method that has been developed to segment raspberry plants from the background using a selected spectral ratio combined with edge detection. Graph theory was used to minimise a cost function to detect the continuous boundary between uninteresting plants and the area of interest. The method includes automatic detection of a known reflectance tile which was kept constantly within the field of view for all image scans. A method to split images containing rows of multiple raspberry plants into individual plants was also developed. Validation was carried out by comparison of plant height and density measurements with manually scored values. A reasonable correlation was found between these manual scores and measurements taken from the images (r2 = 0.75 for plant height). These preliminary steps are an essential requirement before detailed spectral analysis of the plants can be achieved.

AB - Hyperspectral imaging is a technology that can be used to monitor plant responses to stress. Hyperspectral images have a full spectrum for each pixel in the image, 400-2500 nm in this case, giving detailed information about the spectral reflectance of the plant. Although this technology has been used in laboratory-based controlled lighting conditions for early detection of plant disease, the transfer of such technology to imaging plants in field conditions presents a number of challenges. These include problems caused by varying light levels and difficulties of separating the target plant from its background. Here we present an automated method that has been developed to segment raspberry plants from the background using a selected spectral ratio combined with edge detection. Graph theory was used to minimise a cost function to detect the continuous boundary between uninteresting plants and the area of interest. The method includes automatic detection of a known reflectance tile which was kept constantly within the field of view for all image scans. A method to split images containing rows of multiple raspberry plants into individual plants was also developed. Validation was carried out by comparison of plant height and density measurements with manually scored values. A reasonable correlation was found between these manual scores and measurements taken from the images (r2 = 0.75 for plant height). These preliminary steps are an essential requirement before detailed spectral analysis of the plants can be achieved.

KW - Field imaging

KW - Hyperspectral imaging

KW - Image segmentation

KW - Phenotyping

KW - Raspberry

UR - http://www.scopus.com/inward/record.url?scp=85033383150&partnerID=8YFLogxK

U2 - 10.1186/s13007-017-0226-y

DO - 10.1186/s13007-017-0226-y

M3 - Article

VL - 13

JO - Plant Methods

JF - Plant Methods

SN - 1746-4811

IS - 1

M1 - 74

ER -