sensing for food and agriculture in vtt ltd
TRANSCRIPT
VTT TECHNICAL RESEARCH CENTRE OF FINLAND LTD
Sensing Technology for
Food and Agriculture
Philippe Monnoyer, Ph.D.
Sensors in Food and Agriculture, 29-30.11.2016, Møller Centre, Cambridge, UK
VTT Ltd Technical Research Center of Finland
Finland
33
Finland in the world
• Highest Social
Progress Index
(2016)
• Safest Country (2016)
• Greenest (EPI 2016)
• Most powerful
passport (2014)
• Best european intnl
student satisfaction
(2014)
• Least gender gap
(2014)
• OECD education
ranking (2010)• Least corruption
(2014)
• Helsinki quality of
life (2013)
• 4th: Global
Innovation Index
• …
st
2
nd
3d
44
Main exports: Electrotechnical goods, metal products,
machinery, transport equipment, wood and paper products,
chemicals
99 years old RepublicPresident Mr Sauli Niinistö
5.5Millions (18/km2)
Member 1955
Member 1995-40.7 C/-40.7 F +32.8 C/91 F
5
DATA
2013
VTT LtdTechnical Research Centre of Finland
730.12.16 7
BASIC RESEARCH
APPLIED RESEARCH
DEVELOPMENT
UNIVERSITIES
INDUSTRY
VTT
VTT’s status as performer of R&D work
830.12.16 8
VTT Technical Research Centre of
Finland Ltd
Net turnover and other operating
income 272 M€ for VTT Group in 2015(VTT Group’s turnover 185 M€ in 2015)
Personnel 2,470 (VTT Group 31.12.2015 )
Unique research and testing
infrastructure
Wide national and international
cooperation network
We use
4 million
hours of brainpower a
year to develop
new technological
solutions
36% of Finnish
innovations
include VTT
expertise
TOP 2VTT is second
most active
patenting
organisation in
Finland (2014) A leading R&D organisation
in Nordic countries
We provide expert services
for our domestic and
international customers
and partners, both in
private and public sectors
Click to edit Master title style
Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
930.12.16 9
Some VTT spin-offs and start-ups
10
Agriculture
1130.12.16 11
Top Agricultural Products
12
Milk consumption
1330.12.16 13
1000 €More income
for farmersEach cow, Each year
Milk Quality
1430.12.16 14
HOW?Selection
Food
Health
Reproduction and fertility
…
Hay
Artturi Virtanen: Nobel prize in Chemistry 1945
"Artturi Virtanen - Facts". Nobelprize.org. Nobel Media AB 2014. Web. 24 Nov 2016.
<http://www.nobelprize.org/nobel_prizes/chemistry/laureates/1945/virtanen-facts.html>
1730.12.16 17
Fodder
10,000
Barn fires
Total damages
Billions
Quanturi is the forerunner of the Farming IoT
19
Alerts
Cloud service
Visualization
The Quanturi solution
20
Sensor probes
Reader
Quanturi is the prize winner of
the Innov’Space from the
international agricultural fair in
France in September 2016
21
Haytech by Quanturi video https://goo.gl/bpMkUg
Measuring
Milk
Quality
2323
Spectrometer Microspectrometer
Fragile
Heavy
Big
Expensive
Robust
Light
Small
Cheap
2430.12.16 24
Existing MEMS or Piezo FPI technology platforms
0,2 0,3 µm0,4 0,6 0,8 1 2 3 4 5 6 7 8 10 11 12 13 14
VisibleUV NIR SWIR LWIRMWIR
CMOS & CCD Image Sensors
InGaAs
InSb
MCT
0,82-1,05
PbSe
0,40-0,58
0,50
1,5-2,0
1,7-2,2
1,95
1,3-1,7
1,77
1,5
3,0-3,7
2,8-3,4
3,3
3,3
3,2-4,0
4,0-5,03,3
4,2
8,5-10,7
7,3-9,6
9,3
9,3
0.4-1.0 1.0-2.5
0.50-0.85
Material science
Biology
Medical, Anticounterfeit,
Space, Remote sensing, PoC
Pharma,
Food,
Agriculture
Gases,
Plastics MineralsMinerals
Forestry
4.0-5.0
Poly-Si-SiO2
0,58-0,80 0,950,36-0,50
0,47-0,65
0,52-0,71
5.5-7.5
PbTe-ZnSe
7.0-11.0
PbTe-ZnSe0.36-0.4
0.29-0.34
Metallic
mirrors, Ti,
Ag
Dielectric
mirrors
1.0-2.5
Poly-Si-SiO2
2530.12.16 25
After ~200,000 hours of R&D …
350 nm
UV
11 µm
TIR
Microspectrometry
& Hyperspectral
Imaging Solutions
from
+ Color Xray based on Medipix
CERN technology!
BUILDS YOUR CUSTOM HSI CAMERA
2630.12.16 26
Spectroscopic milk sensing
Measurement performed with
VTT MEMS based
Microspectrometry
2727
Comparison with ICAR accuracy requirements
ICAR = International Committee for Animal Recording
http://www.icar.org/
In our model, RMSECV for fat and protein weight-% is 0.26 and 0.20,
respectively. These two values can be roughly compared to the
standard deviation limits stated in the table above.
2830.12.16 28
IoT
of
dugs
www.spectralengines.com
Data management, storing,
analysis and sensor fusion
Elisa IoT™
Internet of FarmingDairy business in Europe 150 B€
24 million cows in 1 million farms
75 million agricultural IoT devices by 2020
Improve production efficiency of
farms by monitoring milk quality
Real-time fat and protein
information from individual cows
*) Business Insider, “Why IoT, Big Data & Smart Farming is the Future of Agriculture”, October 2016.
3030
What about undesired substances in milk ?
Antibiotics ?
Antibiotic resistance
Yogurt, cheese preparation affected (200 M€/yr in EU)
Copyright © 2016 by BioMensio
Handheld multi-analyte biosensor platformRobust, Multiplexed and Label-free
SensorSample
Microfluidic cartridge
Label-free
Cost efficient
Digital
Handheld
Robust
Integrated Multi-analyte
Easy to use
Hand held platform
Disposable
cartridge
Reader
UI
Focus:
• Antibiotics in milk and meat juice
• Mycotoxins from grain products
3230.12.16 32
Fungi Fusarium
producing Mycotoxins
in cereal crops
25% of the world's
crops are
affected by
mycotoxins each
year, with annual
losses of around
1 billion metric
tons of foods and
food products.
Testing needed
through the food
supply chain
3330.12.16 33
HappyCow EU project – A FITBIT FOR COWS
Common abnormalities and
behavioural phenomena to be
detected:
Estrus
Diseases
Lameness
Rumination
Acceleration and magnetic
field in 3D >base station
>cloud
433 MHz
x100 modules/base station
300 m range
The system has been developed in cooperation with Connecterra (NL), University of Wageningen (NL) and University of Aarhus (DK)
3430.12.16 34
Prize winning solution – Web summit 2015
Connecterra was chosen as the top startup for 2015 in the
Alpha category.
https://blog.websummit.net/introducing-2015s-best-startups-its-
our-pitch-powered-by-audi-winners/
188,000
lakes in
Finland
3630.12.16 36
Blue-Green Algae• Link to fertilizers
• Several toxins
• Economic impact US only: $75M (Larkin&Adams, 2007)
• Swimming, Inhalation, Swallowing
• Affects fishes (depleted oxygen)
• Serious cases of human and livestock poisoning and deaths in a number of
countries worldwide (e.g., in Brazil, Australia and North America)
Some of the more common toxins produced by blue-green algae can cause allergic-
type reactions such as rashes, eye, nose, throat irritation, asthma, headaches, fever,
nausea, stomach cramps, vomiting and diarrhea
Other toxins can impact internal organs, and can cause gastroenteritis, tissue
damage, muscle weakness and paralysis in severe exposure cases
Central nervous system and can cause seizures, paralysis, respiratory failure or
cardiac arrest
37
Paper based test for toxic Cyanobacteria from
water
• Paper-based test for
consumers to detect toxic
cyanobacteria
• Paper based diagnostic
platform
• Printed fluidic channels
on paper (VTTs IPR)
• Inkjet deposited
biomolecules
• Adjustable platform which
can be adapted for various
detection chemistries
38
Mapping of
the
cyanobacteri
a in the lake
Lohjanjärvi.
Project HSI-
stereo in
collaboration
with
University of
Jyväskylä,
FGI, Luode
consulting,
Lentokuva
Vallas
Image © Ilkka
Pölönen,
Anna-Leena
Erkkilä
Or with Hyperspectral Imaging
© Natural Resources Institute Finland
Precision fertilization task based on hyperspectral imaging
39 30.12.2016
1000*
1000
*35.5 442.0 BiomassNitrogen
BiomassNNEED
www.dronefinland.fi
40
Forests
Sources: Forest Europe: State of Europe’s Forests 2015
100 Millions ha
20% of Finnish exports. 11 b€
4130.12.16 41
Pinus
peuce
Pinus
sylvestris
Abies
koreana
Picea
koreana
Picea
jezoenis
Picea
abies
Pseudotsuga
menziesii
Picea
omorika
Quercus rubra
Examples
of tree
species
Composite of bands with
1578, 925 and 636nm
• 400-1000 nm VNIR camera
• 1100-1600 nm SWIR camera
R.Näsi et al., UAS BASED TREE SPECIES IDENTIFICATION
Processing flow for Hyperspectral data
Field reflectance
reference
Data capture
Image pre-
processing
Image
orientation
DSM
generation
Radiometric
block
adjustment
Individual tree
detection
Spectral
characteristics
of trees
ClassificationMosaic
generation
Forest
reference
plots
ALS DTM
R.Näsi et al., UAS
BASED TREE
SPECIES
IDENTIFICATION
Forest classification
Näsi, R., Honkavaara, E., Lyytikäinen-Saarenmaa, P., Blomqvist, M., Litkey, P.,
Hakala, T., ... & Holopainen, M. (2015). Using UAV-Based Photogrammetry and
Hyperspectral Imaging for Mapping Bark Beetle Damage at Tree-Level. Remote
Sensing, 7(11), 15467-15493Image courtesy Roope Näsi from National Land Survey of Finland
Bark Beetle Outbreaks
Grains
Control crop pricing
Control protein for aninal feed
…
• On site handheld measurement of grains, which can determine fast
and accurately the concentrations of key determinants of harvest
value and processing cost:
• Protein
• Carbohydrates (starch + sugars + fibers)
• Oil
• Moisture
• Positioning (GPS) and wireless internet connection in order to
enable value added services
• Data collection and analysis (“big data”)
• SW products which improve productivity
World First Truly Handheld
46
GrainSense Diffuse reflection Diffuse transmission
Detector + Silicon - InGaAs (1.0 < λ < 2.4µm) + Silicon
% of mass sampled + 100% - Tiny (<<1%) +/- ~50%
Sample presentation error + None +/- Moderate (with mechanics) - Large (”pin holes”)
#Kernels required + 10 to 100 +/- Many thousands - Thousands
Wet sample possible? + Yes +/- Within limits (mech. problems) - No
Optical Penetration: Method Comparison
• GrainSense addresses 1 billion plus
market potential worldwide with a Cloud
based service offering and a hand-held
device
• Pricing very competitive and affordable by
individual farmers
• GrainSense pay back time within one
harvesting season (less than 6 months)
Out in Market 2017
4930.12.16 49
Potato grower’s app by Solentum B.V.
Solentum B.V., a provider of science based tools and services, is an
independent daughter company of HZPC Holland B.V, the world
leader in potato breeding
Solentum B.V. has developed a mobile application to support the
potato growers
By means of a photo taken with the mobile phone, the length, width,
and the measured weight per potato is calculated. Based on the
entered data of the field, the expected yield is given.
VTT contributed in developing the detection algorithm for the
Solentum’s application
5030.12.16 50
Printed visual indicators at VTT Ltd
Detecting:
• O2
• H2S
• Ethanol
• Aldehydes and ketones
Poultry freshness (H2S)
With UPM (Finland)
The development work carried out in EU project SusFoFlex “Smart and sustainable food packaging
utilizing flexible printed intelligence and materials technologies” funded by EU FP7/2007-2013 under
grant agreement no 289829 (2012-2014)
Ethanol sensor
coupled with RFID
For fresh cut fruits
Anti-tampering label INTACT
Clean Card PRO®
Clean surfaces test
with Orion Diagnostica
51
IMS beads with signal enhancers
Raman spectrum of Listeria innocua
Uusitalo, S., et al. "Detection of Listeria innocua on roll-to-roll produced SERS
substrates with gold nanoparticles." RSC Advances 6.67 (2016): 62981-62989.
Listeria detection by Surface-enhanced Raman
spectroscopy (SERS)
Low cost R2R polymer based SERS substrate
Bacteria accumulated by immunomagnetic
separation beads for enhanced signal intensity
The limit of detection in the range of ∼104 CFU ml−1
S.Z. Oo, R.Y. Chen, S. Siitonen, V. Kontturi, D.A. Eustace, J. Tuominen, S. Aikio, and M.D.B.
Optics Express Vol. 21, Issue 15, pp. 18484-18491 (2013).
5252
Roll-to-roll patterned
SERS substrates by UV-nanoimprint lithography
Production by roll-to-roll printing and vacuum deposition of metal
Roll of sensors are cut into smaller pieces that are disposable,
sensor price ~ 1€
Sensor area is about 1 cm X 1cm
Analysis off-line by Raman microscopy and on-line by fiber
connected analyzers
Sensor nanostructures Devices for analysis
S.Z. Oo, R.Y. Chen, S. Siitonen, V. Kontturi, D.A. Eustace, J. Tuominen, S. Aikio, and M.D.B. Charlton: Disposable plasmonic plastic SERS sensor.
Optics Express Vol. 21, Issue 15, pp. 18484-18491 (2013).
53
Acknowledgements
• Roope Näsi , Eija Honkavaara,
Anttoni Jaakkola et al. (fgi.fi, nls.fi)
• Jere Kaivosoja et al. (luke.fi)
• Joni Leinvuo, biomensio.com
• Jarkko Antila, spectralengines.com
• Nadine Pesonen, quanturi.com
• Edvard Krogius, grainsense.com
• VTT: Heikki Saari, Altti Akujärvi,
Leena Hakalahti, Maria Smolander,
Thea Sipiläinen-Malm, Himadri
Majumdar, Liisa Hakola, Kaarle
Jaakkola, Kaj Nummila,…
vttresearch.com
@dronefinland
@PhMonnoyer