Elsevier

Biosystems Engineering

Volume 164, December 2017, Pages 49-67
Biosystems Engineering

Review
Close range hyperspectral imaging of plants: A review

https://doi.org/10.1016/j.biosystemseng.2017.09.009Get rights and content

Highlights

  • HSI is an automated, non-destructive and rapid alternative to explore plant traits.

  • It acquires both chemical and structural information related with plant traits.

  • Phenotyping, disease, species identification, foliar chemistry estimation.

  • Illumination effects are the major technical challenge to be dealt with.

The increasing need to develop a rapid understanding of plant functional dynamics has led to the employment of sensor technology for non-destructive assessment of plants. Hyperspectral Imaging (HSI) being an integration of two modalities, imaging and point spectroscopy, is nowadays emerging as a potential tool for rapid, non-destructive and automated close range assessment of plants functional dynamics both in terms of structure and physiology.

Firstly, this paper presents an overview of some basic concepts of close range HSI on plants, concerning the plant–light interaction, instrumental setup, and spectral data analysis. Furthermore, the work reviews recent advances of HSI for plant related studies under controlled experimental conditions as well as in natural agricultural settings. Applications are discussed on foliar content estimation, variety identification, growth monitoring, stress and disease-related studies, phenotyping and adoption of HSI in high-throughput phenotyping platforms (HTPPs).

Close range HSI is a challenging task and suffers from technical complexities related to external factors (e.g. illumination effects) as well as plant-related factors (e.g. complex plant geometry). The paper finally discusses some of the technical challenges related to the implementation of HSI in the close range assessment of plant traits.

Introduction

To increase the agriculture productivity, it is important to develop high-yielding crops which can adapt to future climatic conditions (Furbank & Tester, 2011). Moreover, the assessment of existing and to be developed high-yielding crop plants is crucial to understand how the detrimental effects of surrounding environment limit their growth and yield, further supporting the development of optimal plant varieties. Traditional methods used for plant assessment are still time-consuming, labour intensive and destructive in nature (Busemeyer et al., 2013).

The need for fast, non-destructive and high throughput alternative technologies for plant assessment has lead to the development of new and the re-use of different existing sensor technologies from various scientific domains (Li, Zhang, & Huang, 2014). One such emerging sensor technology for non-destructive, rapid and automated assessment of plants is the hyperspectral imaging (HSI) (Matsuda, Tanaka, Fujita, & Iba, 2012). The application of HSI can be found in diverse domains of research such as remote sensing (Blackburn, 2007), foods (Mishra et al., 2015, Mishra et al., 2016, Wu and Sun, 2013), microbiology (Gowen, Feng, Gaston, & Valdramidis, 2015) and pharmaceutical sciences (Gendrin, Roggo, & Collet, 2008). In particular, in remote sensing, vegetation monitoring has been studied using HSI for many years (Blackburn, 2007), and has motivated the use of HSI for exploring plants at close range.

A HSI system integrates a spectrograph that records reflectance in a wide range of the spectrum, including the ultraviolet (UV), visible (VIS) and near-infrared (NIR) into a digital sensor (Bock, Poole, Parker, & Gottwald, 2010). Data is generated in the form of a 3D spatial map of spectral variation: the first two dimensions provide the spatial information and a third dimension accounts for the spectral information. Being an integration of imaging and conventional spectroscopy, HSI can obtain complementary information from both domains. While point spectroscopy gathers information to understand the physiology of the plants (Montes, Melchinger, & Reif, 2007), the information from imaging technology is used to understand the structural dynamics (Apelt et al., 2015, Bucksch et al., 2014, Dhondt et al., 2014). In combination, HSI has the potential to extract integrated spatial and spectral information related to the plant's functional dynamics regarding both structure and physiology (Bergsträsser et al., 2015, Kuska et al., 2015, Mahlein et al., 2012a, Rascher et al., 2011, Ustin and Gamon, 2010).

Various emerging applications of HSI related to plants biochemistry estimation (Vigneau, Ecarnot, Rabatel, & Roumet, 2011), stress detection (Rumpf et al., 2010, Mahlein et al., 2010, Mahlein et al., 2013), species identification (Kumar, Skidmore, & Mutanga, 2010) and Phenotyping (Leucker et al., 2017, Wahabzada et al., 2016) have gained the interest of plant biologists and agronomist all over the globe, covering the need for a fast, non-destructive and visually interpretable technology (Fiorani et al., 2012, Rascher et al., 2011). Furthermore, HSI has been increasingly integrated with high-throughput phenotyping platforms (HTPPs) for controlled and automated assessment of plants (Busemeyer et al., 2013). HTPPs perform qualitative and quantitative determination of functional plant traits, such as plant growth and biomass yield, resulting from the interaction of the genetic characteristics of the plant with surrounding controlled environmental conditions (Furbank & Tester, 2011). These functional traits are reflected into different regions of the reflectance spectrum and can be estimated through HSI.

The HSI can be used to image the complete plants, as is required in a monitoring experiment, as well as to image plucked leaves from the plants to perform quantitative or qualitative analysis of interest. Individual plant pixels captured by HSI provide spectral information related to the chemical composition of the plant, i.e. its physiological status (Behmann, Mahlein, Rumpf, Römer, & Plümer, 2015). To understand the effects of artificial or natural environmental factors, the plant spectra can be explored and used for model development. For example, in the case of stress related studies, stress resulting from a drought induces degradation of the leaf chlorophyll content leading to an increase in reflectance over the whole VIS spectrum (400–700 nm). Similar information generated by HSI can be used for identification, quantification and spatial representation of nutrient deficiencies, diseases, and drought under controlled experiments in a greenhouse or under uncontrolled conditions in an agriculture field (Sankaran, Mishra, Ehsani, & Davis, 2010).

The aim of the paper is to provide the reader with an overview of the principles, instrumentation and data handling of close range HSI of plants, followed by a review of recent applications of plant assessment using HSI. Finally, the major challenges associated with the implementation of close range HSI for plants are discussed, and the paper is concluded with some future directions. All the applications reviewed in the paper were studies performed after the year 2000 and were selected from six different domains related to exploration of plants, i.e. foliar biochemistry estimation, leaf monitoring, species identification, stress detection, phenotyping and HTPPs.

Section snippets

Interaction of plants with light

The interaction of light (electromagnetic radiation (EMR)) with plants differs according to the light frequencies. Since the leaves are mainly responsible for the photosynthetic activity, the interaction of light with the leaves is of particular interest (Jacquemoud & Ustin, 2001). For green leaves, the relevant regions of EMR are the VIS region (400–700 nm), responsible for absorption of light by photosynthetic pigments; the NIR (700–1100 nm), dominated by absorption by dry matter; and the

Imaging sensor

A HSI setup consists four main parts: the light source, the objective lenses, the imaging spectrograph and the area detector. The selection of the light source is crucial to ensure good performance and reliability of any optical inspection system. Typically, halogen lamps are most widely used in HSI systems for indoor applications related to plants (Behmann et al., 2014b, Matsuda et al., 2012, Yu et al., 2014). Halogen lamps are broadband illumination sources covering the visible and NIR

Recent applications

Recent applications of HSI, ranging from simple biochemical content estimation to monitoring of plant growth under controlled environment (Phenotyping) are presented in Table 1. The major applications were related to foliar content estimation, disease detection, variety identification, leaves monitoring, stress related studies, phenotyping and adoption of HSI in HTPPs. These applications are further discussed in the following subsections.

Technical challenges

In the last decade, HSI is increasingly being applied for assessment of functional plant traits in close range settings. Nevertheless, as an emerging application, it faces some major challenges limiting its full potential. Many of the limitations are either related to the technical complexity of the imaging setup or the nature of the samples to be imaged.

To date, the most common imaging setup used for close range HSI acquisition is based on a push-broom movement of the sensor. In this

Conclusion

The ability of HSI to provide integrated spatial and spectral information offers a perfect combination of rapid non-destructive assessment of plant traits. In the VIS region of the electromagnetic spectrum, the spectral information is majorly dominated by pigments while in the NIR, the information is related to different biochemical constituents such as water, starch, protein etc. A wide range of emerging applications of HSI, on whole plants or on plucked leaves, such as foliar content

Acknowledgement

Puneet Mishra acknowledges the financial support from INDIA4EU2 project of Erasmus Mundus during his stay at Technical University of Madrid, Spain (http://www.india4eu.eu/). Mohd Shahrimie M. A. acknowledges the support of the Academic Staff Training Scheme (ASTS) of the Universiti Sains Malaysia and the Ministry of Higher Education of Malaysia.

References (121)

  • J.-B. Feret et al.

    Prospect-4 and 5: Advances in the leaf optical properties model separating photosynthetic pigments

    Remote Sensing of Environment

    (2008)
  • F. Fiorani et al.

    Imaging plants dynamics in heterogenic environments

    Current Opinion in Biotechnology

    (2012)
  • R.T. Furbank et al.

    Phenomics technologies to relieve the phenotyping bottleneck

    Trends in Plant Science

    (2011)
  • Y. Ge et al.

    Temporal dynamics of maize plant growth, water use, and leaf water content using automated high throughput rgb and hyperspectral imaging

    Computers and Electronics in Agriculture

    (2016)
  • P. Geladi et al.

    Hyperspectral imaging: Calibration problems and solutions

    Chemometrics and Intelligent Laboratory Systems

    (2004)
  • C. Gendrin et al.

    Pharmaceutical applications of vibrational chemical imaging and chemometrics: A review

    Journal of Pharmaceutical and Biomedical Analysis

    (2008)
  • M. Gerhards et al.

    Water stress detection in potato plants using leaf temperature, emissivity, and reflectance

    International Journal of Applied Earth Observation and Geoinformation

    (2016)
  • A.A. Gowen et al.

    Recent applications of hyperspectral imaging in microbiology

    Talanta

    (2015)
  • A. Gowen et al.

    Hyperspectral imaging an emerging process analytical tool for food quality and safety control

    Trends in Food Science and Technology

    (2007)
  • S. Jacquemoud et al.

    Prospect: A model of leaf optical properties spectra

    Remote Sensing of Environment

    (1990)
  • S. Jay et al.

    A physically-based model for retrieving foliar biochemistry and leaf orientation using close-range imaging spectroscopy

    Remote Sensing of Environment

    (2016)
  • R.F. Kokaly et al.

    Spectroscopic determination of leaf biochemistry using band-depth analysis of absorption features and stepwise multiple linear regression

    Remote Sensing of Environment

    (1999)
  • W. Kong et al.

    Fast detection of peroxidase (pod) activity in tomato leaves which infected with botrytis cinerea using hyperspectral imaging

    Spectrochimica Acta A: Molecular and Biomolecular Spectroscopy

    (2014)
  • M. Lara et al.

    Monitoring spinach shelf-life with hyperspectral image through packaging films

    Journal of Food Engineering

    (2013)
  • Z.-Y. Liu et al.

    Characterizing and estimating fungal disease severity of rice brown spot with hyperspectral reflectance data

    Rice Science

    (2008)
  • J. Luo et al.

    Savitzky–Golay smoothing and differentiation filter for even number data

    Signal Processing

    (2005)
  • A.-K. Mahlein et al.

    Development of spectral indices for detecting and identifying plant diseases

    Remote Sensing of Environment

    (2013)
  • M. Min et al.

    Design of a hyperspectral nitrogen sensing system for orange leaves

    Computers and Electronics in Agriculture

    (2008)
  • P. Mishra et al.

    Application of independent components analysis with the {JADE} algorithm and {NIR} hyperspectral imaging for revealing food adulteration

    Journal of Food Engineering

    (2016)
  • J.M. Montes et al.

    Novel throughput phenotyping platforms in plant genetic studies

    Trends in Plant Science

    (2007)
  • H. Onoyama et al.

    Potential of hyperspectral imaging for constructing a year-invariant model to estimate the nitrogen content of rice plants at the panicle initiation stage

  • G. Polder et al.

    Measuring ripening of tomatoes using imaging spectrometry

  • F. Apelt et al.

    Phytotyping4d: A light-field imaging system for non-invasive and accurate monitoring of spatio-temporal plant growth

    The Plant Journal

    (2015)
  • N. Arens et al.

    Non-invasive presymptomatic detection of Cercospora beticola infection and identification of early metabolic responses in sugar beet

    Frontiers in Plant Science

    (2016)
  • D. Bannon et al.

    Harsh environments dictate design of imaging spectrometer

    Laser Focus World

    (2005)
  • P. Baranowski et al.

    Hyperspectral and thermal imaging of oilseed rape (brassica napus) response to fungal species of the genus alternaria

    PLoS One

    (2015)
  • R. Barnes et al.

    Standard normal variate transformation and de-trending of near-infrared diffuse reflectance spectra

    Applied Spectroscopy

    (1989)
  • J. Behmann et al.

    Generation and application of hyperspectral 3d plant models

  • J. Behmann et al.

    A review of advanced machine learning methods for the detection of biotic stress in precision crop protection

    Precision Agriculture

    (2015)
  • Bellwether. https://www.danforthcenter.org/scientists-research/core-technologies/phenotyping. (Last accessed 7 October...
  • S. Bergsträsser et al.

    Hyperart: Non-invasive quantification of leaf traits using hyperspectral absorption-reflectance-transmittance imaging

    Plant Methods

    (2015)
  • G.A. Blackburn

    Hyperspectral remote sensing of plant pigments

    Journal of Experimental Botany

    (2007)
  • C. Bock et al.

    Visual rating and the use of image analysis for assessing different symptoms of citrus canker on grapefruit leaves

    Plant Disease

    (2008)
  • C. Bock et al.

    Plant disease severity estimated visually, by digital photography and image analysis, and by hyperspectral imaging

    Critical Reviews in Plant Sciences

    (2010)
  • A. Bucksch et al.

    Image-based high-throughput field phenotyping of crop roots

    Plant Physiology

    (2014)
  • J. Burger

    Hyperspectral NIR image analysis volume 2006

    (2006)
  • J. Burger

    Replacement of hyperspectral image bad pixels

    NIR News

    (2009)
  • L. Busemeyer et al.

    Breedvisiona multi-sensor platform for non-destructive field-based phenotyping in plant breeding

    Sensors

    (2013)
  • Z.-P. Chen et al.

    Extracting chemical information from spectral data with multiplicative light scattering effects by optical path-length estimation and correction

    Analytical Chemistry

    (2006)
  • D. Cozzolino et al.

    Applications and developments on the use of vibrational spectroscopy imaging for the analysis, monitoring and characterisation of crops and plants

    Molecules

    (2016)
  • Cited by (212)

    View all citing articles on Scopus
    View full text