Note: Descriptions are shown in the official language in which they were submitted.
CA 03023336 2018-11-05
WO 2017/191065 PCT/EP2017/060275
Method and apparatus for high throughput testing of a treatment on one or more
characteristic of plants
Description
Technical field
This invention relates to evaluation of the effects of treatment on one or
more characteristics
of plants.
More in particular, the invention relates to a method and apparatus for
evaluation of the
effects of a treatment on one or more characteristics of plants.
Background art
When growing plants, e.g. cultivated for their seeds (also called seed crops)
for example rice,
wheat, barley, corn, soybean, canola, sunflower, millet and safflower, a major
goal is to
support the growth of the plant such that they produce a high yield in e.g.
seeds or biomass
or roots. Farmers support this growth amongst others by application of
fertilisers, herbicides,
pesticides, insecticides, bactericides, nematicides and/or fungicides.
In the development of these chemicals and/or biologicals for application, such
as fertilisers,
herbicides, pesticides, insecticides, bactericides nematicides and/or
fungicides, the testing of
those products and the testing of the formulations comprising those products
is an important
step. The screening of the effects of the application of these products and
their effective
amounts is traditionally done in a field setting using large amounts of
plants. Accelerated
systems for screening herbicide treatments were already developed as described
e.g. by
Stanley etal. in Forests 2014, 5, 1584-1595.
Tools for fast, accurate and efficient screening for effects of chemical
and/or biological
treatment on plant are a necessity for the plant growing industry. But also
other types of
treatment might provide the plant growing industry further insights.
Traditional methods for evaluating the effect of treatment on plants comprise
a phenotyping
of the plants, which involve labour-intensive procedures such as manual and
visual
measurements of dimensions, such as above and belowground biomass, pigments,
shapes,
growth, counting of plant parts, and weighing of plant parts such as
individual leaves,
inflorescences and seeds. Some of these operations require detaching the plant
parts of
interest from the subtending plant organs. Advancements in the phenotyping of
plants are
.. already available e.g. as described in WO 2010/031780 or WO 2013/001436.
Disclosure of Invention
It is therefore an object of the present invention to provide a device and
methods which at
least partially avoid the disadvantages and shortcomings of the systems and
methods known
from the prior art.
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
2
The present invention overcomes these shortcomings by providing an apparatus
and method
for evaluation of the effects of a treatment on the physical characteristics
of plants. In a
preferred embodiment, the apparatus and method provide for a high throughput
and fully
automated evaluation of the effects of a treatment on one or more
characteristics of plants.
The invention further relates to a method and apparatus for selecting the most
desired
genotypes based on scoring of one or more characteristics of treated plants,
and to a method
for rapid analysis of stress resistance of treated plant specimens. Biotic
stress can be caused,
for example, by bacterial, fungal, or viral disease, insects and nematodes.
Abiotic stress can
be caused, for example, by heat, drought, cold, wind, high salinity, and low
or too high nutrient
levels. The invention further relates to a method and apparatus to screen for
a desired
chemical or biological compound to treat a plant with, or a formulation
thereof, or for
determining the most optimal application regime of such chemical or biological
compound,
e.g. depending on the developmental stage of the plant. Examples of chemical
compounds
include fertilizers, herbicides, insecticides, pesticides, bactericides,
nematicides, compounds
inducing or inhibiting certain developmental steps in plant growth, and
nutrients; examples of
biological compounds include formulations comprising microorganisms or spores
with a
particular, preferably beneficial effect on plant growth (growth promotion,
treatment of
disease).
Devices and methods of this kind may be applied in all fields of agricultural
research and
commercial activities and in all fields of chemical and/or biological
technology related to plants
and plant specimens. Preferably, the device and methods according to the
present invention
may be applied to the technical field of testing of plants, testing of
chemical or biological
compounds and/or testing of methods for treatment of plants, such as one or
more of: testing
and/or evaluation of foliar application of biologicals and/or chemicals;
testing of resistance of
treated plants against specific types of stress; testing of specific
fertilizers, herbicides,
insecticides, pesticides, bactericides, nematicides and/or nutrients; the
testing of the effect
and/or effectiveness of specific treatment regimes, such as 1, 2, 3, 4, 5, 6,
7, 8, 9, 10, 11, 12,
13, 14, 15, 16, 17, 18, 19, 20 treatments of the plants or plant specimens
with fertilizers and/or
biocides. However, other applications of the present invention are possible.
Here, we describe a method and apparatus for analysis of the effects of a
treatment on one
or more characteristics of plants, which enhances dramatically the statistical
power of the
evaluation compared to traditional automated methods based on the "first in -
last out" or "first
in - first out" principles, because it provides for a real randomisation of
the plants thereby
reducing position and/or edge effects (i.e. effects on plant growth caused by
a particular
position at the growing location). The method and apparatus described herein
enable a high
throughput analysis of plants as these enable continuous processing and
treatment of the
provided plants. Preferably, the methods and apparatus of the present
invention comprise a
fully automated imaging and image processing step.
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
3
Further advantages of the invention will become apparent from the following
description.
Plants that can be evaluated by the present method and apparatus can be any
plant or
population having similar or differing genetic information. These plants may
be different
varieties, hybrids, inbreds and transgenic plants.
An aspect of the present invention provides a method for high throughput
evaluation of the
effects of a treatment on a plurality of plants.
The method comprises following steps: a plurality of plant containers wherein
at least one
plant is growing is provided as well as a container moving system to move said
plant
containers. A pre-treatment randomisation system is provided to randomise the
order of said
plant containers. And also a treatment system and a post-treatment
randomisation system
are provided. Each of said plant containers are moved by said container moving
system to
the pre-treatment randomisation system wherein the order of the plant
containers are
randomised. Thereafter the container moving system moves the containers to the
treatment
system, the treatment system then providing at least one treatment.
Thereafter, the
containers are moved by the container moving system to the post-treatment
randomisation
system wherein the order of the plant containers are randomised for a second
time.
Thereafter, the containers are moved by container moving system to a growing
location. The
post-treatment randomisation step assures that position or edge effects from
the growing
location on the groups of treated plants are minimised. The effect of the at
least one treatment
is then evaluated in a next step, at one or more timepoints after treatment.
The treatment in the above method can be any treatment a plant can undergo.
Such treatment
can be one or more of a foliar treatment, e.g. foliar treatment with a
chemical or biological
solution or powder, a foliar treatment with a coating solution or powder, a
foliar treatment with
a marker solution or powder; watering with a specified solution, such as
nutrient solution
and/or biocide solution; heating and/or cooling the plant in the plant
container; providing a
light application; shaking the plant container; a blowing application on the
plant; raining and/or
snowing and/or hailing application; high pressure or low pressure atmospheric
environment
application; fumigation or a gaseous application; or a sound or sonic
treatment. Other
treatments can also be a pollination action, or a hormonal application, or an
inoculation,
and/or insect and/or microbial and/or fungal infestation application.
Plants in a greenhouse are influenced by variations in the growth environment
caused by
variations in, for example, temperature, humidity, light, nutrient, and water
supply, which are
dependent on the location of the plant in the greenhouse. A plant at the outer
side of the
greenhouse is exposed to a different micro-environment than a plant at the
centre of the
greenhouse. Typically, the plants are set-up in plant containers in rows or on
tables in such
greenhouse and problems of environment-associated phenotype and/or metabolite
components are dealt with by moving the plant containers to another spot in
the greenhouse.
Most of the commercially available systems work, in case of a row set-up, in a
first-in, first-
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
4
out or a first-in, last out way. This is not providing a real randomisation of
the plants or plant
containers. This is overcome by the system and method of the invention. The
advantage of
the method and apparatus of the invention is that the pre-treatment
randomisation uses the
originating cultivation location as the randomising factor and that the post-
treatment
randomisation uses the treatment as the randomising factor. This first
randomisation
averages the effect of the growing location before treatment and the second
randomization
after treatment will randomize the treated plants again such that the later
measurement of the
one or more characteristics will be done on truly randomized plants which are,
after treatment,
also growing at a cultivation location having its specific micro-environment.
In addition, where
a heterogenous population of plants are grown (e.g. a population of plants
having different
genotypes, a population of plants after a mutagenesis treatment or a set of
transgenic events)
a further randomisation is possible at the sowing or planting stage, to
minimize position or
edge effects from the growing location.
The treatment is preferably done on a group of several plants being subjected
to the same
treatment (a block). As such the method of the invention provides for a fully
automated system
to perform randomized block trials to test the effect of a specific treatment
or to identify the
best performing treatment for a desired or undesired effect out of a group of
treatments; e.g.
a foliar application with the same substance in differing dilutions.
In a preferred embodiment, the method of the invention is a high throughput
evaluation of the
effects of a foliar treatment, preferably a spraying, comprising the steps as
provided above.
In an even more preferred embodiment, the method then also provides a drying
system
wherein the plant containers pass the drying system after the spraying
treatment. In another
preferred embodiment, the treatment is a watering of the plant container with
a defined
solution, e.g. a defined nutrient solution or a defined biocide solution.
Preferably, each plant is linked to a unique identifier. Preferably, the
identifier may be or may
comprise, but is not limited to, one or more of the following identifiers: a
barcode; a
contactless electronic identifier, i.e. an identifier comprising at least one
piece of information,
which may be read from the identifier, preferably without any physical contact
between a
reading mechanism, preferably a reader, and the identifier, most preferably
the identifier may
be at least one radio frequency identification tag (RFID tag). However,
alternatively or
additionally, other types of identifiers are possible. The information may be
a simple
identification, e.g. a plant specimen and/or a genotype and/or growth
conditions and/or
treatment. In general, the at least one identifier not necessarily has to be
in physical contact
with the plant, but should be assigned to a respective plant in any
unambiguous way.
The evaluation can be done by visual scoring or by sampling or by imaging.
Preferably, the method further provides at least one imaging system. The plant
containers
than also pass through this imaging system while performing the method.
Imaging can be
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
done before and/or after the treatment. Preferably, the evaluation of the
effect of the treatment
is made by use of this imaging system. Such evaluation of the effect of the
treatment is
preferably done when the plants had some time to obtain the full effect of the
treatment. In
the case of a foliar application, the evaluation is preferably done after the
plants had the
5 chance to grow in the randomised block pattern on the growing location.
The plants can then
be imaged by use of the imaging system of the method.
Preferably, the imaging system comprises at least one detector.
The term detector, as used in the present invention, may imply any type of
detector, preferably
a detector for electromagnetic waves. The term electromagnetic radiation, as
used in the
present invention, may comprise light in the visible range, X-ray, UV,
infrared and near-
infrared, thermal and terahertz radiation. It may comprise monochromatic
electromagnetic
(EM) radiation as well as a broad spectrum EM-radiation and it may comprise
incoherent EM-
radiation as well as coherent EM-radiation, polarised and unpolarised EM-
radiation. Other
types of electromagnetic waves are also possible. More preferably the detector
may comprise
a detector for light in at least one spectral wave length region selected from
a visible, an
infrared and ultraviolet wavelength region and most preferably a camera. The
camera may
be a digital camera, preferably with spatial and/or time resolution.
Preferably, the detector acquires at least one spatially resolved image. One
or more
characteristics are then measured from said image by appropriate software
which then
provides resulting information.
Preferably, the imaging system is imaging one or more characteristic of each
of the plants.
The imaging system then also provides for analysing the images for the one or
more
characteristic of each of the plants by computer processing and for
associating the resulting
information with the unique identifier information for each of the plants
respectively.
In an even more preferred embodiment, the method also comprises a step of
directing
electromagnetic waves to the plant, such that the plant emits or reflects
electromagnetic
waves. The plant is then imaged by the detector at different wavelengths
wherein images
comprising pixels are obtained. These images recorded at different wavelengths
are aligned
on the basis of the pixels, such that a 3-dimensional image is generated. The
3-dimensional
image, the image cube, comprises 2 spatial dimensions and 1 spectral
dimension. In the next
step of this method, a customary predictive mathematical model combining the
weighted
contributions of the different wavelengths is used to obtaining a
multispectral or hyperspectral
imaging cube of the plant and the one or more characteristics is then measured
from said
imaging cube by appropriate software.
The electromagnetic waves emitted or reflected from the plant are preferably
transmitted light.
In another preferred embodiment, the electromagnetic waves emitted or
reflected from the
plant are reflected light.
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
6
In a preferred embodiment the images are collected at many different narrow
wavebands in
the near infrared range of the light spectrum, preferably between 900 and 1800
nm.
In an even more preferred embodiment, the method comprises the hyperspectral
or
multispectral imaging described above in combination with a 3D imaging, which
provides a
4-dimensional image.
The term image, as used in the present invention, may imply any type of
images, preferably
at least two-dimensional images. The images may be optical images. The images
may
comprise transmission images and/or shadow images and/or reflection images.
The images
may be generated by detecting an emission signal, e.g. a fluorescence and/or
phosphorescence signal. Thus, the images may be generated by chlorophyll
fluorescence
measurements and/or selectable marker fluorescence measurements. The signal
which may
be used to generate an image may be discrete in time or may be a continuous
signal. Other
types of images are also possible as e.g. described hereunder. From the above
it follows that
the term imaging, as used in the present invention, may imply any way of
acquiring images.
The one or more characteristics are measured from the image by appropriate
software. If
desired, algorithms may be used to evaluate the measured one or more
characteristics.
In a preferred embodiment, the imaging system images plants while the plants
are being
moved and turned at the same time in a controlled manner to be able to take
images of all
sides of the plant and store them in digital format, for example as described
in WO
2010/031780. By turning the plants while imaging them, an additional step of
orienting the
plant with its maximal radial axis towards the imaging device can be avoided,
and a more
complete picture of the plant is obtained.
The one or more characteristics comprise one or more of, but are not limited
to, an observable
physical manifestation of the plant, a phenotypic trait, metabolic trait,
colour, greenness, yield,
growth, biomass, maturity, a transgenic trait, flowering, nutrient use, water
use, or effects of
disease, pests, and/or stress. Preferably, the one or more characteristics
comprise one or
more of area, height, width, leaf angle, number of leaves, presence and/or
number of
inflorescences, number of shoots, and branching pattern. In another preferred
embodiment,
the one or more characteristics comprise one or more of different metabolites,
and might
entail the assessment of the presence or absence of a specific metabolite,
number of
metabolites, the amount of a specific metabolite,...
In a preferred embodiment, the methods of the present invention can be used to
detect any
characteristic of the plants that can be measured by imaging. The images may
be taken from
aboveground plant parts and/or or plants roots. The aboveground plant parts
may be one or
.. more of shoots, leaves, tillers, inflorescence, flowers, seeds. In one
preferred embodiment,
the one or more characteristic is one or more of a quantitative trait, a
biochemical trait and a
morphological trait. In an even more preferred embodiment, the biochemical
trait is selected
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
7
from the group comprising of oil composition, protein composition,
carbohydrate composition,
fibre composition, oil content, protein content, carbohydrate content, starch
content, fibre
content, dry weight and water content. In another even more preferred
embodiment, the
morphological trait is selected from plant architecture, plant size, plant
shape, aboveground
biomass, plant colour, plant growth rate, leaf surface texture, plant weight,
plant integrity, leaf
integrity, leaf colour, leaf shape, leaf size, leaf growth rate, belowground
biomass, root growth
rate, root thickness, root length, root anchorage, inflorescence architecture,
flower size,
flower shape, flower colour, flower surface texture, flower weight, flower
integrity, endosperm
size, germ size, seed shape, seed size, seed colour, seed surface texture,
seed weight, seed
density, and seed integrity. As used herein, integrity is correlated to
susceptibility or
resistance to any one of diseases, insect infestation, fungal infestation or
environmental
stress. In an alternative preferred embodiment, the quantitative trait is
selected from amount
of (green) leaves, amount of roots, such as amount of hairy roots and/or
branched roots,
amount of florets, amount of seeds, amount of empty seeds, amount of
branching, weight of
seeds, total weight of seeds and/or fill rate. However, other types of
parameters and/or
combinations of the named parameters and/or other parameters may be possible,
e.g.
aboveground biomass per plant and per area; belowground biomass per area;
content of oil,
starch and/or protein in aboveground biomass (e.g., seeds or vegetative parts
of the plant);
number of flowers (florets) per plant; or modified architecture, such as
increase stalk
diameter, thickness or improvement of physical properties (e.g. elasticity,
soil penetration
capacity of roots). Where applicable, changes in characteristics of the plants
can be
measured over time or by comparison to suitable control plants such as wild-
type or reference
plants, or untreated plants. Untreated plants can for example be plants that
received no
treatment at all, or that were treated, in case of spraying, with water only
or with the
formulation in absence of the active ingredient(s). Persons skilled in the art
are aware of
selecting proper control plants.
In another preferred embodiment, the methods of the present invention can be
used to detect
any characteristic of the plants that can be measured by taking a small sample
of the plants.
The sample can be taken from any part of the plant, such as aboveground parts,
such as a
leaf or a flower, or from belowground parts, such as roots or root microbiome.
The samples
are then analysed by metabolic profiling for one or more characteristics.
In a preferred embodiment, the method further comprises a step of analysing
the resulting
information for the one or more characteristic for the one or more plant(s) to
determine the
impact of the treatment.
The methods of the present invention can be used to analyse the impact of
genetic
modifications on plants in combination with the evaluation of the effects of a
treatment, and
in particular on one or more characteristics of the plant, and selecting a
plant with a genetic
modification of interest. Such method comprises following steps: first a
plurality of plants with
differing genotypes are grown. Preferably each plant is associated with an
identifier, more
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
8
preferably a machine readable identifier that distinguishes the plant from
other plants. Plants
are treated and images or samples are obtained using the methods described
herein and
these images or samples are then analysed for one or more characteristic,
and/or trait as
described above, to determine the impact of the treatment in relation to the
genetic
modification. A selection can then be made for a plant or seeds thereof with a
genetic
modification of interest in relation to the treatment. If desired, algorithms
may be used to
select and evaluate the measured one or more characteristics and the results
statistically
analysed to identify plants with genetic modifications of interest, for
selecting the best
performing candidates or for selecting candidates having any given
characteristics for any
given further treatment.
The creation of genotypic variation can be based on genetic modifications made
in the lab,
but can also rely on the production of genetic alterations that can be
obtained by techniques
including recombination through classical crossing, chemical mutagenesis,
radiation-induced
mutation, somatic hybridisation, inter-specific crossing and genetic
engineering. The obtained
plants can be compared to other non-transgenic plants, to transgenic plants
and/or to
corresponding control plants. Following the creation of genotypic variation,
selection of those
genotypes having the most desirable agronomic one or more characteristics is
performed.
The information resulting from the measurement of the one or more
characteristics from the
image or sample by appropriate software is preferably also associated to the
identifier.
The invention provides in another of its aspects a process for evaluating and
recording of the
effects of a treatment on one or more characteristics of a plant, comprising
the steps of
identifying the plant, treating and imaging the plants using the methods
described herein,
determining the one or more characteristics and recording results in a
prescribed format in a
computer database together with the plant identifier.
The computer database compiled by subjecting plants to a process as aforesaid
may be
interrogated and enables rapid comparison of characteristics from a multitude
of different
plants and thus permits rapid determination of seeds from which further plants
may be derived
which yield seeds having desired characteristics.
The invention provides in another of its aspects a process for comparing one
or more of the
characteristics of the plants in a batch of treated plants with corresponding
characteristics of
batches of plants subjected to another treatment, in which a computer database
compiled by
subjecting batches of plants to a process according to the last preceding
paragraph but one
is interrogated concerning said one or more characteristics.
Likewise, the computer database compiled by subjecting plants to a process as
aforesaid
may be interrogated and enables rapid comparison of characteristics from a
multitude of
different plants and thus permits rapid determination of compounds with which
the plants
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
9
were treated and which compounds have desired characteristics. Such compounds
can be
chemical compounds or biological compounds. In a similar way, the computer
database can
be interrogated to determine optimal dosing or treatment regimes, dependent on
how the
treatment of the plants was set up. The invention thus provides in another of
its aspects a
process for comparing one or more chemical or biological compounds having
effect on the
characteristics of plants in a batch of treated plants when compared to
corresponding
characteristics of batches of untreated plants or of plants subjected to
another treatment.
Preferably, the method further provides for a step wherein one or more plants
are selected
for further use in a plant breeding or advancement experiment or for
introducing further
modifications in transgenic plants. Likewise, the method further provides for
a step wherein
one or more compounds are selected for further use in an advancement
experiment, or
wherein a dosing or treatment regime is further optimised. Examples of
advancement
experiments include further optimisation experiments, or further selection
experiments,
further screening experiments and the like.
Another aspect of the invention provides an apparatus for high-throughput
application of a
treatment on a plurality of plant containers wherein at least one plant is
growing, wherein the
apparatus comprises a container moving system to move the plant containers; a
pre-
treatment randomisation system to randomise the order of the plant containers;
a treatment
system; and a post-treatment randomisation system. Each of the plant
containers move into
the apparatus by the container moving system and to the pre-treatment
randomisation
system. The plant containers are then randomised by that pre-treatment
randomisation
system, where after the containers move further on the container moving system
to the
treatment system which then treats the plants in the plant containers.
Thereafter, the plant
containers move via the container moving system to the post-treatment
randomisation system
which performs a second randomisation of the plant containers. Thereafter the
plant
containers are moved out of the apparatus by the container moving system.
Preferably, the
plant containers are then moved to a plant growing location, such as a
greenhouse or
screen house.
The treatment in the above apparatus can be any treatment a plant can undergo.
Such
treatment can be one or more of a foliar treatment; watering with a specified
solution, such
as nutrient solution and/or biocide solution; heating and/or cooling the plant
in the plant
container; providing a light application; shaking the plant container; a
blowing application on
the plant; raining and/or snowing and/or hailing application; high pressure or
low pressure
atmospheric environment application; fumigation or a gaseous application, or a
sound or
sonic treatment. Other treatments can also be a pollination action, or a
hormonal application,
or an inoculation, and/or insect and/or microbial and/or fungal infestation
application.
Typically, the treatment is used in a screening program, where for example one
or more plants
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
are tested for a response to the treatment, or where one or more chemical
compounds or one
or more biological agents are tested for their effect on plants or on plant
growth.
The treatment is preferably done in a block of several plants being subjected
to the same
5 .. treatment. As such the apparatus of the invention provides for a fully
automated system to
perform randomized block trials to test the effect of a specific treatment or
to identify the best
performing treatment for a desired or undesired effect out of a group of
treatments; e.g. a
foliar application with the same substance in differing dilutions or differing
formulations. The
post-treatment randomisation step will randomly distribute the plant
containers with the
10 treated plants over the growing location, thereby minimising position or
edge effects of the
growing location.
In a preferred embodiment, the apparatus of the invention enables a high
throughput
evaluation of the effects of a foliar treatment, preferably a spraying,
comprising the steps as
provided above. In an even more preferred embodiment, the apparatus also
provides a drying
system wherein the plant containers pass the drying system after the spraying
treatment.
In another preferred embodiment, the treatment is a watering of the plant
container with a
defined solution, e.g. a defined nutrient solution or a defined biocide
solution.
The apparatus may further also comprise one or more identifier reader(s) to
identify an
identifier linked to the plant in the plant container, the reader preferably
providing output in
digital form. Examples of such a reader are, but are not limited to, a barcode
reader, a
transponder reader and an RFID reader. In a preferred embodiment, the
apparatus further
comprises at least one electronic code reading device to identify an
identifier linked to said
plant. The identifier reader is preferably integrated by use of software in a
computer device
and fed therefrom to the database. The database may be manipulated to inspect
and
compare data to determine various characteristics of the plant. The apparatus
preferably
comprises as many identifier readers as needed to be able to determine the
identity of a plant
at any position in the apparatus: before and/or after the pre-randomisation
stage, at the
treatment stage, before and/or after the post treatment randomisation stage.
In another preferred embodiment, the apparatus comprises also at least one
imaging system.
Preferably, the imaging system comprises one or more detectors. Even more
preferably, the
imaging system comprises at least one detector, most preferably at least one
digital camera.
The term detector, as used in the present invention, may imply any type of
detector, preferably
a detector for electromagnetic waves. The term electromagnetic waves, as used
in the
present invention, may comprise light in the visible range, infrared and near-
infrared light. It
may comprise monochromatic light as well as a broad spectrum of light and it
may comprise
incoherent light as well as coherent light. Other types of electromagnetic
waves are also
possible. More preferably the detector may comprise a detector for light in at
least one
spectral wave length region selected from a visible, an infrared and
ultraviolet wavelength
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
11
region and most preferably a camera. The camera may be a digital camera,
preferably with
spatial and/or time resolution. More preferably, the camera is a line scan
camera.
In a preferred embodiment, the treatment system and the imaging system can
operate
independent from each other or they can co-operate, thus giving the apparatus
a great
versatility: for example, plants can be moved from the growing location to the
imaging system
for analysis, and subsequently moved back to the growing location, or plants
can be moved
from the growing location to the treatment system for treatment and
subsequently moved
back to the growing location, or plants can be moved from the growing location
to the imaging
system for analysis, and subsequently moved to the treatment system for
treatment (or vice
versa) and subsequently moved back to the growing location. The apparatus thus
allows one
set of plant containers being imaged in the imaging system and another set of
plant containers
simultaneously being treated in the treatment system without interfering with
each other.
The apparatus may further comprise at least one image analysis device. The
image analysis
device may be adapted to perform at least one image analysis of at least one
of the images,
preferably the image analysis device may be adapted to derive at least one
characteristic of
the plant. The one or more characteristics are measured based on analysis of
the image by
appropriate software. If desired, algorithms may be used to evaluate the
measured one or
more characteristics.
The apparatus further may have at least one database for recording data
regarding the plants
and the treatment or treatments performed on each plant. The data preferably
may be at least
one of the following: at least one image of the plant; at least one or more
characteristics or
trait derived from at least one image of the plant; information from the
identifier; treatment
performed on the plant. As outlined above, the at least one characteristic or
trait may
comprise one or more parameters characterizing the characteristic or trait of
the plant.
The one or more characteristics comprise one or more of an observable physical
manifestation of the plant, a phenotypic trait, a metabolic trait, colour,
greenness, yield,
growth, biomass, maturity, a transgenic trait, flowering, nutrient use, water
use, or effects of
disease, pests, and/or stress. Preferably, the one or more characteristics
comprise one or
more of area, height, width, leaf angle, number of leaves, presence and/or
number of
inflorescences, number of shoots, and branching pattern.
In a preferred embodiment, the apparatus of the present invention can be used
to detect any
characteristic of the plants that can be measured by imaging. The images may
be taken from
aboveground plant parts and/or or plants roots. The aboveground plant parts
may be one or
more of shoots, leaves, tillers, inflorescence, flowers, seeds. A system for
root imaging is
described in WO 2010/031780.
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
12
In another preferred embodiment, the apparatus of the present invention can be
used to
detect any characteristic of the plants that can be measured by taking a small
sample of the
plants. The sample can be taken from any part of the plant, such as
aboveground parts, such
as a leaf or a flower, or from belowground parts, such as roots or root
microbiome. The
samples are then analysed for example by metabolic profiling for one or more
metabolic
characteristics, or by microbial analysis to determine the presence or absence
of certain
microorganisms, or for morphologic analysis on tissue or cellular level.
The at least one characteristic or trait may preferably be chosen from: one or
more of a
quantitative trait, a biochemical trait and a morphological trait. In a
preferred embodiment, the
biochemical trait is selected from the group consisting of oil composition,
protein composition,
carbohydrate composition, fibre composition, oil content, protein content,
carbohydrate
content, starch content, fibre content, dry weight and water content. In
another preferred
embodiment, the morphological trait is selected from plant architecture, plant
size, plant
shape, aboveground biomass, plant colour, plant growth rate, leaf surface
texture, plant
weight, plant integrity, leaf integrity, leaf colour, leaf shape, leaf size,
leaf growth rate,
belowground biomass, root growth rate, root thickness, root length, root
anchorage,
inflorescence architecture, flower size, flower shape, flower colour, flower
surface texture,
flower weight, flower integrity, endosperm size, germ size, seed shape, seed
size, seed
colour, seed surface texture, seed weight, seed density, and seed integrity.
As used herein,
integrity is correlated to susceptibility or resistance to any one of
diseases, insect infestation,
and fungal infestation.
In an alternative preferred embodiment, the quantitative trait is selected
from amount of
(green) leaves, amount of roots, such as amount of hairy roots and/or branched
roots, amount
of florets, amount of seeds, amount of empty seeds, amount of branching,
weight of seeds,
total weight of seeds and/or fill rate. However, other types of parameters
and/or combinations
of the named parameters and/or other parameters may be possible, e.g.
aboveground
biomass per plant and per area; belowground biomass per area; content of oil,
starch and/or
protein in aboveground biomass (e.g., seeds or vegetative parts of the plant);
number of
flowers (florets) per plant; or modified architecture, such as increase stalk
diameter, thickness
or improvement of physical properties (e.g. elasticity).
In another preferred embodiment, the one or more characteristics comprise one
or more of
different metabolites, and might entail the assessment of the presence or
absence of a
specific metabolite, number of metabolites, the amount of a specific
metabolite,...
The apparatus according to the invention permits derivation of data about
plant
characteristic(s) or traits after treatment without human intervention other
than perhaps
providing the treatment to the plants. It may be used for a variety of
purposes and is especially
useful for evaluation of biocides, such as foliar application of herbicides in
differing dilutions.
In such use, the apparatus provides an integrated automatic process for
evaluating one or
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
13
more characteristic and/or phenotype of a treated plant or plants. By use of
the apparatus
one may derive in a single operation desired data about key parameters of
interest to the
plant breeder such as optimal dilution of a specific compound or effective
amounts of a
compound, or optimal formulation of a specific compound. Other purposes for
which the
apparatus can be used include evaluation of growth promoting substances, in
the form of a
chemical compound or as a microbial suspension.
The apparatus furthermore may comprise a control system which may be adapted
to control
and/or to drive the imaging system and/or transporter and/or the conveyor belt
systems and/or
the image analysis device and/or the reader and/or the database and/or a power
supply. The
control system may comprise a computer and electrical and/or signal
connectors, preferably
electrical lines and interfaces.
Preferably, the imaging system is shielded from natural daylight. Light inside
the imaging
system may be provided by a standardized set of lamps of which the intensity
can be
controlled.
Images taken in the imaging system can be processed on-line using imaging
analysis
software to extract information on the plants and preferably, the processed
data as well as
the images get linked to the unique identifier of the corresponding plant and
even more
preferably, downloaded to a computer.
In a preferred embodiment, the imaging system comprises the following:
¨ at least one digital camera with sensitivity in the visual, infrared
and/or near-
infrared range;
¨ at
least one spectrograph composed of an optical dispersing element such as a
grating or prism to split the light into many narrow, adjacent wavelength
bands,
said spectrograph being placed before the camera and being tuneable so that
specific wavebands can be selected and transmitted to the camera in a
predetermined sequence;
¨ at least one suitable optical lens;
¨ at least one light source with suitable spectral composition in the near
infrared
range to illuminate said plant,
¨ computer hardware elements and connections to the different previous
elements
and
¨
dedicated software elements for driving signal outputs and inputs from and to
the
hardware elements, and automatically perform the different steps of the method
described herein.
Such imaging is often referred to in literature as imaging spectroscopy, which
is the
simultaneous acquisition of spatially co-registered images in many spectrally
contiguous
bands. The image produced by such imaging spectroscopy is similar to an image
produced
by a digital camera, except each pixel has many bands of light intensity data
instead of just
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
14
three bands: red, green, and blue. In the art, the wording "hyper spectral"
data sets are
described as being composed of relatively large number (e.g., 100-1000)
spectral bands of
relatively narrow bandwidths (e.g., 1-10 nm), whereas, "multi-spectral" data
sets are usually
fewer (e.g., 5-10) bands of relatively large bandwidths (e.g., 70-400 nm).
In a preferred embodiment, the imaging system comprises a hyperspectral
camera. In
another preferred embodiment, the imaging system comprises a multispectral
camera.
Another aspect of the present invention provides for the use of an apparatus
as described
herein for evaluating the effects of a treatment on a plurality of plants.
Preferably, such an
apparatus is used in the methods as described herein.
In another aspect, the apparatus as described herein can be used in a method
for comparing
the effects of different treatments on similar plants.
In another aspect, the apparatus as described herein can be used in a method
for comparing
the effects of different growth conditions of plants in relation to the
effects of a treatment on
a plurality of plants.
In another aspect, the apparatus as described herein can be used for screening
a population
of plants by measuring the effects of a treatment, for example in a breeding
experiment.
In another aspect, the apparatus as described herein can be used in a method
for testing the
effects of treatment of plants, such as one or more of: testing of specific
fertilizers and/or
nutrients; testing of specific pesticides, the selection and/or breeding of
plants having one or
more desired properties in response to the effects of a specific treatment;
the testing of the
effect and/or effectiveness of specific treatments, such as 1, 2, 3, 4, 5, 6,
7, 8, 9, 10, 11, 12,
13, 14, 15, 16, 17, 18, 19, 20 treatments of the plants or plant specimens
with fertilizers,
nutrients and/or biocides.
In an alternative aspect, the apparatus as described herein can be used in a
method for
phenotyping and/or metabolic profiling in relation to a treatment, for
selecting the most
desired genotypes based on phenotype and/or metabolic profile scoring of the
treated plant.
In another alternative aspect, the apparatus as described herein can be used
in a method for
analysis of stress resistance of treated plant specimens.
Summarizing the ideas of the present invention, the following embodiments are
preferred:
Embodiment 1: Method for high throughput evaluation of the effects of a
treatment on a
plurality of plants said method comprising following steps:
- providing a plurality of plant containers 120 wherein at least one plant
121 is growing
- providing a container moving system 130 to move said plant containers
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
- providing a pre-treatment randomisation system 140 to randomise the order
of said
plant containers
- providing a treatment system 150, and
- providing a post-treatment randomisation system 160,
5 wherein each of said plant containers 120 moves by said container moving
system 130
to said pre-treatment randomisation system 140,
said pre-treatment randomisation system 140 randomising the order of plant
containers
121,
said container moving system 130 then moving said containers 121 to said
treatment
10 system 150,
said treatment system 150 providing at least one treatment,
thereafter said container moving system 130 moving said containers to said
post-
treatment randomisation system 160,
said post-treatment randomisation system 160 performing a second randomisation
of
15 the order of said plant containers,
said container moving system 130 moving said containers 120 to a growing
location
and subsequently evaluating the effect of said treatment.
Embodiment 2: Method according to embodiment 1, wherein said plant in said
plant container
is linked to a unique identifier 180.
Embodiment 3: The method according to any one or more of the previous
embodiments, said
method further providing at least one imaging system 190, wherein said plant
containers
pass through said imaging system.
Embodiment 4: The method according to the previous embodiment, wherein said
plant
containers pass through an imaging system 190 before and/or after said
treatment.
Embodiment 5: Method according to any one or more of the embodiments 3 or 4,
wherein
said evaluation of the effect of said treatment is made by use of said imaging
system
190.
Embodiment 6: The method according to any one or more of the previous
embodiments,
wherein said imaging system comprises one or more detectors 191.
Embodiment 7: The method according to the previous embodiment, wherein said
detector
comprises a camera, preferably a digital camera.
Embodiment 8: Method according to any one or more of the previous embodiments
3 to 7,
said imaging system 190 further providing:
¨ directing electromagnetic waves on said plant thereby forming
emitted or reflected
electromagnetic waves from said plant;
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
16
¨ imaging said plant at different wavelengths by said detector 191 thereby
obtaining
images comprising pixels;
¨ aligning said images recorded at different wavelengths on the basis of
said pixels,
thereby generating a 3-dimensional image, said 3-dimensional image comprising
2 spatial dimensions and 1 spectral dimension;
¨ using a customary predictive mathematical model combining the weighted
contributions of the different wavelengths, thereby obtaining a multispectral
or
hyperspectral imaging cube of said plant;
¨ measuring one or more characteristic from said imaging cube by
appropriate
software.
Embodiment 9: The method according to the previous embodiment, wherein said
images are
collected at many different narrow wavebands in the visual, infrared and/or
near infrared
range of the light spectrum, preferably between 900 and 1800 nanometres.
Embodiment 10: Method according to any one or more of embodiments 6 to 9,
wherein said
detector acquires at least one spatially resolved image, the method further
providing for
measurement of one or more characteristic from said image by appropriate
software
providing resulting information.
Embodiment 11: The method according to any one or more of the previous
embodiments 3
to 10, wherein said imaging system 190 is imaging one or more characteristic
of said
plant and analysing the images for the one or more characteristic of the plant
by
computer processing and associating the resulting information with the unique
identifier
180 information for said plant 121.
Embodiment 12: Method according to any one or more of embodiments 9 to 11,
wherein the
one or more characteristic comprises one or more of an observable physical
manifestation of the plant, a phenotypic trait, metabolic trait, colour,
greenness, yield,
growth, biomass, maturity, a transgenic trait (i.e. a trait altered by the
presence of a
transgene), flowering, nutrient use, water use, or effects of disease, pests,
and/or
stress.
Embodiment 13: Method according to the previous embodiment, wherein said one
or more
characteristic is one or more of a quantitative trait, a biochemical trait and
a
morphological trait.
Embodiment 14: Method according to any one or more of the embodiments 8 to 12,
wherein
the one or more characteristic comprises one or more of area, height, width,
leaf angle,
number of leaves, presence and/or number of inflorescences, number of shoots,
and
branching pattern.
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
17
Embodiment 15: Method according to any one or more of embodiments 8 to 12,
said method
further comprising a step of analysing the resulting information for the one
or more
characteristic of the one or more plant(s) to determine the impact of the
treatment.
Embodiment 16: The method of embodiment 1, wherein the plurality of plants
comprise one
or more transgenic plants.
Embodiment 17: The method of embodiment 1, wherein one or more plants are
selected for
further use in a plant breeding or advancement experiment or for introducing
further
modifications.
Embodiment 18: The method of embodiment 7, wherein the images and/or
information are
taken of above ground plant parts and/or of plant roots.
Embodiment 19: The method of embodiment 18, wherein the above ground plant
parts
comprise shoots, leaves, tillers, inflorescence, flowers, seed, or any
combination
thereof.
Embodiment 20: Method according to any or more of the previous embodiments,
wherein the
treatment is a foliar spraying treatment.
Embodiment 21: Method according to the previous embodiment, said method
further
providing a drying system 170, said method further comprising a step wherein
the plant
containers pass a drying system after the foliar spraying treatment.
Embodiment 22: Method according to any one or more of the embodiments 1 to 19,
wherein
the treatment is a watering of the plant container with a defined solution.
Embodiment 23: Method according to the previous embodiment, wherein the
defined solution
is a defined nutrient solution.
Embodiment 24: Method according to embodiment 22, wherein the defined solution
is a
defined biocide solution.
Embodiment 25: Apparatus for high throughput application of a treatment on a
plurality of
plant containers 120 wherein at least one plant 121 is growing,
said apparatus comprising:
- a container moving system 130 to move said plant containers;
- a pre-treatment randomisation system 140 to randomise the order of said
plant
containers
- a treatment system 150, and
- a post-treatment randomisation system 160,
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
18
wherein each of said plant containers 120 moves into said apparatus by said
container
moving system 130 to said pre-treatment randomisation system 140, the order of
plant
containers being randomised by said pre-treatment randomisation system 140
before
moving further on said container moving system 130 to said treatment system
150
wherein said plant containers 120 are treated, the plant containers 120 then
move via
the container moving system 130 to the post-treatment randomisation system
160, the
post-treatment randomization system 160 providing a second randomisation of
the
order of said plant containers and
thereafter the plant containers 120 being moved by said container moving
system 130
out of said apparatus.
Embodiment 26: Apparatus according to the previous embodiment, wherein said
apparatus
further comprises a unique identifier reader 181.
Embodiment 27: Apparatus according to any one or more of the embodiments 25 or
26, said
apparatus further comprising at least one imaging system 190.
Embodiment 28: Apparatus according to the previous embodiment, wherein said
imaging
system 190 comprises one or more detectors 191.
Embodiment 29: Apparatus according to the previous embodiment, wherein said
imaging
system comprises at least one detector, preferably at least one digital
camera.
Embodiment 30: Apparatus according to any one or more of the embodiments 25 to
29,
wherein said treatment system 150 comprises a foliar spraying system 151.
Embodiment 31: Apparatus according to the previous embodiment, wherein said
apparatus
further comprises a drying system 170.
Embodiment 32: Apparatus according to any one or more of the embodiment 25 to
29,
wherein said treatment system 150 comprises a watering system 152.
Embodiment 33: Apparatus according to any one or more of embodiments 27 to 32,
wherein
said imaging system 190 comprises:
¨ at
least one digital camera with sensitivity in the visual, infrared and/or near-
infrared range;
¨ at least one spectrograph composed of an optical dispersing element such
as a
grating or prism to split the light into many narrow, adjacent wavelength
bands,
said spectrograph being placed before the camera and being tuneable so that
specific wavebands can be selected and transmitted to the camera in a
predetermined sequence;
¨ at least one suitable optical lens;
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
19
¨ at least one light source with suitable spectral composition in the near
infrared
range to illuminate said plant;
¨ computer hardware elements and connections to the different previous
elements;
¨ dedicated software elements for driving signal outputs and inputs from
and to the
hardware elements, and automatically perform the different steps of the method
described in any one of embodiments 1 to 24 and 39 to 41.
Embodiment 34: Use of an apparatus according to any one or more of embodiments
27 to 33
for evaluating the effects of a treatment on a plurality of plants.
Embodiment 35: Use of an apparatus according to any one or more of embodiments
27 to 33
in the method of any one of the embodiments 1 to 24 and 39 to 41.
Embodiment 36: : Use of an apparatus according to any one or more of
embodiments 27 to
33 in a method for comparing the effects of different growth conditions of
plants in
relation to the effects of a treatment on a plurality of plants.
Embodiment 37: Use of an apparatus according to any one or more of embodiments
27 to 33
in a method for phenotyping and/or metabolic profiling, for selecting the most
desired
genotypes based on phenotype or metabolite scoring in the evaluation of the
effects of
a treatment on a plurality of plants.
Embodiment 38: Use of an apparatus according to any one or more of embodiments
27 to 33
in a method for testing the effects of treatment of plants, such as one or
more of: testing
of specific fertilizers and/or nutrients; testing of specific biocides, the
selection and/or
breeding of plants having one or more desired properties in response to the
effects of
a specific treatment; the testing of the effect and/or effectiveness of
specific treatments,
such as 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,20
treatments of
the plants or plant specimens with fertilizers, nutrients and/or biocides.
Embodiment 39: Method of embodiment 2, wherein said unique identifier 180
communicates
with unique identifier reader 181.
Embodiment 40: Method according to any or more of embodiments 1 to 19, wherein
the
treatment is on above-ground plant parts.
Embodiment 41: Method according to embodiment 22, wherein the defined solution
comprises growth promoting substances, either as chemical compound or as a
microbial suspension.
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
Embodiment 42: Apparatus according to any one or more of the embodiments 25 to
29,
wherein said treatment system 150 comprises a spraying system 151 suitable for
spraying above-ground plant parts.
5
In order that the invention may become more clear there now follows a
description to be read
with the accompanying schematic drawings of apparatuses and methods according
to the
invention and their use in a process according to the invention selected for
description to
10 illustrate the invention by way of example. The examples are by way of
illustration alone and
are not intended to completely define or to otherwise limit the scope of the
invention.
Description of the figures
Figure 1 is a schematic view of one embodiment of a method and apparatus for
high
15 throughput evaluation of the effects of a treatment on a plurality of
plants.
Figures 2A and 2B is a schematic view of a method and apparatus for high
throughput
evaluation of the effects of a spraying treatment on a plurality of plants.
Figures 3A and 3B shows a schematic view of a method and apparatus for high
throughput
evaluation of the effects of a combined spraying and drying treatment on a
plurality of plants.
20 Figure 4 shows a schematic view of a method and apparatus for high
throughput evaluation
of the effects of a watering treatment on a plurality of plants.
110: apparatus for high throughput treatment of a plurality of plants
120: plant container
121: plant
130: container moving system (not shown)
140: pre-treatment randomisation system
141: pre-spraying randomisation system
150: treatment system
151: spraying system
152: watering system
160: post-treatment randomisation system
161: post-spraying randomisation system
170: drying system
180: unique identifier (not shown)
181: unique identifier reader
190: imaging system
191: detector (not shown)
192: image
193: image analysis device
200: control system
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
21
Examples
Figure 1A is an example of a method and apparatus for high throughput
evaluation of the
effects of a treatment on a plurality of plants. The plants 121 are growing in
plant containers
120 at a growing location, in this example a greenhouse. Such greenhouse
provides to the
plant containers an environment of controlled climatic conditions with
controlled supply of
nutrients and feed water. But it goes without saying that the plants in the
plant containers in
such greenhouse are influenced by variations in the growth environment (micro-
climate
variations) caused by variations in, for example, temperature, humidity,
light, nutrient, and
water supply, which are depending on the location of the plant container in
the greenhouse
(position effects). A plant in a plant container at the outer side of the
greenhouse or of a
growth table is exposed to a different micro-environment than a plant in a
plant container at
the centre of the greenhouse or at the centre of a group of plants (edge
effects). Typically,
the plant containers are set-up in rows or on tables in such greenhouse and
problems of
environment-associated phenotype and/or metabolite components are dealt with
by moving
the plant containers to another spot in the greenhouse. Most of the
commercially available
systems work, in case of a row set-up, in a first-in, first-out or a first-in,
last out way. This is
not providing a real randomisation of the plants or plant containers. This is
overcome by the
system and method of the invention.
The apparatus 110 (not depicted) comprises a container moving system 130 (not
depicted)
to move the plant containers 120. The plant containers 120 are moved from the
growing
location into apparatus 110. The apparatus further also comprises a pre-
treatment
randomisation system 140 to randomise the plant containers; a treatment system
150; and a
post-treatment randomisation system 160. In the method of the invention each
of the plant
containers 120 move into the apparatus by the container moving system 130 and
to the pre-
treatment randomisation system 140. The plant containers are then randomised
by that pre-
treatment randomisation system 140, which used the originating cultivation
location as the
randomising factor. Thereafter the containers 120 move further on the
container moving
system 130 to the treatment system 150 which then treats the plants 121 in the
plant
containers. Thereafter, the plant containers 120 containing the treated plants
move via the
container moving system 120 to the post-treatment randomisation system 160
which
performs a second randomisation of the plant containers, but now the treatment
is the
randomising factor. Thereafter the plant containers are moved out of the
apparatus by the
container moving system. Preferably, the plant containers are then moved to a
plant growing
location, such as a greenhouse or screenhouse. Depending on the treatment, the
effect of
the treatment on the plants in the plant containers is evaluated immediately
or after a certain
time after the plant containers left the apparatus of the invention. Such
evaluation can be
done visually by scoring the plants at the plant growing location or can be
done in an
automated way. Such automation might entail imaging the plants at the growing
location from
above the plants or can be performed by bringing the plant containers to an
imaging system,
as e.g. described in W02010/031780, where the plants are imaged.
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
22
In the exemplary embodiment of figure 1B, the plant containers are imaged
before the pre-
randomisation step and after treatment and post-treatment randomization. This
second
imaging can be done immediately after the randomization or only after some
time wherein
the plants had the chance to develop the (full) effect of the treatment at the
growing location.
One or more characteristics of the plant is measured from the images by
appropriate
software.
If desired, algorithms may be used to evaluate the measured one or more
characteristics.
The imaging system 190 comprises a detector 191. In this example of figure 1B,
the detector
191 is a digital camera.
The imaging system of figure 1B may further comprise at least one image
analysis device
193 (not shown). The image analysis device 193 may be adapted to perform at
least one
image analysis of at least one of the images 192, preferably the image
analysis device 193
may be adapted to generate at least one characteristic or trait of an imaged
plant. The term
generate according to the present invention may refer to deriving e.g. from
the image
analysis.
The apparatus of the invention may further also comprise an identifier reader
181 (not shown)
to identify an identifier linked to a plant in a plant container or even a
group of plants in a plant
container. Such a reader can be a barcode reader, a transponder reader and/or
an RFID
reader. In a preferred embodiment, the apparatus further comprises at least
one electronic
code reading device to identify an identifier linked to said plant.
The apparatus further may have at least one database (not shown) for recording
data
regarding the plant, the treatment and the effect of the treatment, i.e. the
one or more
characteristics of the plant after treatment and or the difference in the one
or more
characteristics before and after treatment of a particular plant. The data
preferably may be at
least one of the following: at least one image of the plant aboveground and/or
belowground;
at least one characteristic or trait derived from at least one image of the
plant; at least one or
more characteristic derived from metabolite analysis of a sample taken from a
plant,
information from the identifier; information on the treatment; information on
the time after
treatment when determination of the effect takes place. As outlined above, the
at least one
characteristic or trait may comprise one or more parameters characterizing the
phenotype of
the plants. In a preferred embodiment, the methods of the present invention
can be used to
detect any characteristics of the plants that can be measured by imaging. The
images may
be taken from aboveground plant parts and/or or plants roots. The aboveground
plant parts
may be one or more of shoots, leaves, tillers, inflorescence, flowers, seeds.
In one preferred
embodiment, the characteristic is one or more of a quantitative trait, a
biochemical trait and a
morphological trait. In an even more preferred embodiment, the biochemical
trait is selected
from the group consisting of oil composition, protein composition,
carbohydrate composition,
amino acid composition, fibre composition, oil content, protein content,
carbohydrate content,
starch content, amino acid content, secondary metabolite content, fibre
content, dry weight
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
23
and water content. In another even more preferred embodiment, the
morphological trait is
selected from plant architecture, plant size, plant shape, branching,
aboveground biomass,
plant colour, plant growth rate, leaf surface texture, plant weight, plant
integrity, leaf integrity,
leaf colour, leaf shape, leaf size, leaf growth rate, belowground biomass,
root growth rate,
root thickness, root length, root branching, root anchorage, inflorescence
architecture, flower
size, flower shape, flower colour, flower surface texture, flower weight,
flower integrity,
endosperm size, germ size, seed shape, seed size, seed colour, seed surface
texture, seed
weight, seed density, and seed integrity. As used herein, integrity is
correlated to
susceptibility or resistance to any one of diseases, insect infestation, and
fungal infestation.
In an alternative preferred embodiment, the quantitative trait is selected
from amount of
(green) leaves, amount of roots, such as amount of hairy roots and/or branched
roots, amount
of florets, amount of seeds, amount of empty seeds, amount of branching,
weight of seeds,
total weight of seeds and/or fill rate. However, other types of parameters
and/or combinations
of the named parameters and/or other parameters may be possible, e.g.
aboveground
biomass per plant and per area; belowground biomass per area; content of oil,
starch and/or
protein in aboveground biomass (e.g., seeds or vegetative parts of the plant);
number of
flowers (florets) per plant; or modified architecture, such as increase stalk
diameter, thickness
or improvement of physical properties (e.g. elasticity).
The apparatus furthermore may comprise a control system 200 which may be
adapted to
control and/or to drive the imaging system 190 and/or container moving system
130 and/or
the image analysis device 193 and/or the reader 181 and/or the database and/or
a power
supply. The control system 200 may comprise a computer and electrical and/or
signal
connectors, preferably electrical lines and interfaces.
Images 192 taken with the imaging system 190 can be processed on-line using
imaging
analysis software to extract information on the one or more characteristics of
the plant and
preferably, the processed data as well as the images get linked to a unique
identifier and
even more preferably, downloaded to a computer.
In a third exemplary embodiment, as shown in figure 2A, the apparatus and
method is the
same as the one described in figure 1A, but the treatment is a spraying
treatment 151. The
spraying treatment can be done on the at least one plant 121 in the plant
containers 120, on
a one by one plant container basis, or the spray treatment 151 can be done in
block of multiple
plants in containers. In the exemplary embodiment of figure 2B, the block
consists of 12 plant
containers. But the skilled person will acknowledge that any amount of plant
containers can
be taken to form a block for the concurrent spraying of the plant containers.
Spraying can be
done top-down and/or on the side of the plant to be treated. As described
above, depending
on the treatment, the effect of the treatment on the plants in the plant
containers is evaluated
immediately or after a certain time after the plant containers left the
apparatus of the invention.
Such evaluation can be done visually by scoring the plants at the plant
growing location or
can be done in an automated way. Such automation might entail imaging the
plants at the
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
24
growing location from above the plants or can be performed by bringing the
plant containers
to an imaging system, as e.g. described in W02010/031780, where the plants are
imaged.
Figure 3A shows a further exemplary embodiment, based on the example as shown
in figure
2A, wherein the method further also comprises a step of drying 170 of the
plants in the plant
containers. The apparatus 110 therefore comprises a drying tunnel or similar
device. The
spraying and drying treatment can be done on the at least one plant 121 in the
plant
containers 120, on a one by one plant container basis, or the spray treatment
151 can be
done in block. In the exemplary embodiment of figure 3B, the block consists of
12 plant
containers. But the skilled person will acknowledge that any amount of plant
containers can
be taken to form a block for the concurrent spraying and drying of the plant
containers. As
described above, depending on the treatment, the effect of the treatment on
the plants in the
plant containers is evaluated immediately or after a certain time after the
plant containers left
the apparatus of the invention. Such evaluation can be done visually by
scoring the plants at
the plant growing location or can be done in an automated way. Such automation
might entail
imaging the plants at the growing location from above the plants or can be
performed by
bringing the plant containers to an imaging system, as e.g. described in
W02010/031780,
where the plants are imaged.
The skilled person will understand that also an imaging step as described in
figure 1B can be
added to the method of figures 2A, 2B, 3A, 3B. The plant containers are than
imaged before
the pre-randomisation step and after treatment and post-treatment
randomization. This
second imaging can be done immediately after the randomization or only after
some time
wherein the plants had the chance to develop the (full) effect of the
treatment at the growing
location. The one or more characteristics of the plant is measured from the
images by
appropriate software. If desired, algorithms may be used to evaluate the
measured one or
more characteristics.
Figure 4 shows a further exemplary embodiment wherein the apparatus and method
is the
same as the one described in figure 1A, but the treatment is a watering
treatment 152. The
watering treatment will be a specific solution provided to the plant
containers and this solution
will comprise specific nutrient solution or biocide solution, or any other
solution or suspension
that affects plant growth. Such defined solution can then be provided via
watering to at least
one plant container 120, on a one by one plant container basis, or the
watering treatment 151
can be done in block. The skilled person will acknowledge that any amount of
plant containers
can be taken to form a block for the concurrent watering of the plant
containers. As described
above, depending on the treatment, the effect of the treatment on the plants
in the plant
containers is evaluated immediately or after a certain time after the plant
containers left the
apparatus of the invention. Such evaluation can be done visually by scoring
the plants at the
plant growing location or can be done in an automated way. Such automation
might entail
imaging the plants at the growing location from above the plants or can be
performed by
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
bringing the plant containers to an imaging system, as e.g. described in
W02010/031780,
where the plants are imaged.
The skilled person will understand that also an imaging step as described in
figure 1B can be
5 added to the method of figure 4.
In a preferred embodiment, the imaging system 190 comprises the following:
- at least one digital camera with sensitivity in the visual, infrared and/or
near-infrared
range;
10 ¨
at least one spectrograph composed of an optical dispersing element such as a
grating or prism to split the light into many narrow, adjacent wavelength
bands,
said spectrograph being placed before the camera and being tuneable so that
specific wavebands can be selected and transmitted to the camera in a
predetermined sequence;
15 ¨ at least one suitable optical lens;
¨ at least one light source with suitable spectral composition in the near
infrared
range to illuminate said plant with light,
¨ computer hardware elements and connections to the different previous
elements
and
20 ¨
dedicated software elements for driving signal outputs and inputs from and to
the
hardware elements, and automatically perform the different steps of the method
described herein.
Such imaging is often referred to in literature as imaging spectroscopy, which
is the
25
simultaneous acquisition of spatially co-registered images in many spectrally
contiguous
bands. In the art, the wording "hyper spectral image cubes" are described as
multichannel
images being composed of many spectrally contiguous spectral bands of
relatively narrow
bandwidths (e.g., 1-10 nm), whereas, "multi-spectral" images are usually fewer
(e.g., 5-10)
bands of relatively large bandwidths (e.g., 70-400 nm).
The imaging system at least comprises a detector 191. Such detector may be a
hyperspectral
camera. In another preferred embodiment, the imaging system comprises a
multispectral
camera.
In another exemplary embodiment, the method according to the invention further
involves the
following steps:
- Collection of digital images of individual plants before and/or after
treatment. One
image of each individual plant is collected by use of a normal RGB colour
camera.
- Generation of one or more characteristics using appropriate software.
- Determination of the pixels belonging to the plant organs, as opposed to
the non-plant
background. This is achieved using standard image processing algorithms, such
as
intensity thresholding, in which the pixel values differing from predetermined
background values are considered as belonging to the plant object.
CA 03023336 2018-11-05
WO 2017/191065
PCT/EP2017/060275
26
- Determination of pixels belonging to the one or more characteristics, as
opposed to the
rest of the plant organs. This is achieved by standard image processing
algorithms,
such as morphological segmentation, in which objects are identified as e.g.
seed or
non-seed, flower, leaf, when their geometrical properties correspond to
predefined
specifications.
- Calculation of the metric properties per each individual object
identified in the image,
based on the combined properties of all individual pixels constituting each
object.
These properties include amongst other physical dimensions in the 2
dimensional
space and amount of plant characteristics.
In a further exemplary embodiment, the method of the invention involves the
following steps:
- Identification of each plant or group of plants being measured by means
of
unambiguous coding system. Ideally the coding system is of a type that can be
read
electronically, e.g. barcode, or transponder tag.
- Collection of digital images of individual plants. Many images of the
same individual
plants are collected at many different narrow wavebands in the near infrared
range of
the light spectrum, namely between 900 and 1700 nm.
- Generation of hyper-spectral image cube by alignment of the images
recorded at the
different wavelengths in order to generate a 3 dimensional image comprising 2
spatial
dimensions (x, y) and 1 spectral dimension (z). From such images, a spectrum
of light
absorption for each pixel in the two-dimensional space can be generated.
- Estimation of the amount of dry matter and basic chemical composition
corresponding
to each pixel, based on a customary predictive mathematical model combining
the
weighted contributions of the different wavelengths at each pixel.
- Determination of the pixels belonging to the plant organs, as opposed to
the non-plant
background. This is achieved using standard image processing algorithms, such
as
intensity thresholding, in which the pixel values differing from predetermined
background values are considered as belonging to the plant object.
- Determination of pixels belonging to the one or more characteristics, as
opposed to the
rest of the plant organs. This is achieved by standard image processing
algorithms,
such as morphological segmentation, in which objects are identified as e.g.
seed or
non-seed, leaf, flower, when their geometrical properties correspond to
predefined
specifications.
- Calculation of the metric properties per each individual object identified
in the spectral
image, based on the combined properties of all individual pixels constituting
each
object. These properties include: physical dimensions in the 2 dimensional
space,
estimated dry weight, and estimated chemical composition.
