1. Toward Global Agricultural Cloud
Masayuki HIRAFUJI* **
Yasuyuki HAMADA*
Tomokazu YOSHIDA*
Atsushi ITOH *
Takuji KIURA *
* NARO National Agriculture and Food Research Organization
** University of Tsukuba

2. “Big Data” Has Been Dream in Agriculture
Plant growth is complex system.
Environment is complex system.
Maximization of income
Minimization of pollution
Maximization of plant growth
Modeling by learning
Analysis between genome and phonotype

3. Nonlinear Regression Models Using Artificial Neural Networks (studied since 20 years ago)
Predicted Yield
Recommendation of fertilizer …
Last year’s application of fertilizer
Accumulated air temperature.
Accumulated soil temperature
Accumulated soil moisture
Last year’s
yield

4. Field Servers for Continuous Data Collection

5. Low-cost USB DNA Sequencer

6. Nanopore Technology

7. More phenotypic data is needed for breeding.

8. Phenomics vs. Genomics
Gene + ome = Genome Genome + ics = Genomics Phenotype + ome = Phenome Phenome + ics = Phenomics

9. Genome Data >> Phenome Data by High-throughput Phenotyping
Nanopore Sequencer
Sensors in Fields
Genotypic Data << Phenome Data Environment Data
Genome Data >> Phenome Data
Environment Data

10. by Open Field Server (Open-FS)
Massive Deployment
by Open Field Server (Open-FS)
Wi-Fi
LED garden light with IR sensor
Solar panel
Photo sensors
Inside temperature sensor
Soil temperature sensor
Soil moisture sensor

11. Field Twitter (Open-FS) Has Been Improved.
樹体水分センサ

12. Towards A Field Phenomics Center
Wi-Fi Router
1km
Phenotype data
Calibration data for remote sensing
Memuro Campus of HARC, NARO

13. Tweeting data

14. Tweeting data

15. Tweeting data

16. Collecting Microscopic Data by A Smartphone with A Macro Lens
A macro lens for iPhone
Stomata on beet leaves can be measured.

17. Products with Twitter

18. Plant Sensor

19. Data of Agricultural Machinery

20. Data stream on agricultural machinery
Petition (GPS)
Speed
Power
Fuel consumption
Steering
Vibration
Yield
Fertilizer
Chemical

21. Reprinted from the Proceeding of AgEng 2011 , pp.294, 2011
XML by iGreen Project for Agricultural Machine EU (Germany) leading. USA has a same project (AgGateway) ・
Reprinted from the Proceeding of AgEng 2011 , pp.294, 2011

22. Farm management data

23. Contents of FIX-pms Common Data Format for Farm Management Data

24. How Can We Combine Data? API of Cloud Services Can Be A Method.

25. Let’s Make Big Data for Agriculture!
Applicayions
Developing
New Businesses
Decision Support
System
Precision Farming
GAP
Models
API （Application Interface）
New Apps
and
Businesses
CLOP: CLoud Open Platform in agriculture
API
API
API
API
Consortium
Faming Data
Field Data
Sensor data
of Agr-Machines
Others
移動監視
SNS
Smartphones
UAV
Satellites
etc.
Sensor Networks
Such As Field Server
ISO11783

26. All Data Provided As API
API of Cloud Services
Satellites
UAV
Smartphone
Variable rate fertilization
Harvester equipped with yield sensor
Sensor data

27. Mashape: Cloud API Hub

28. Mash-Up Using API for Agricultural Data (FIX-pms)
on CLOP
FIX
FARMS
APRAS

29. Big Data Will Be Created by Using API of Apps
FIX
FARMS
APRAS

30. The Best Condition Can Be Found on Nonlinear Models
Predicted yield
Big data
Yield
Fertilizer
Soil temperature
Soil moisture : …
Last year’s
yield
Last year’s application of fertilizer
This year’s
application of
fertilizer
Accumulated soil moisture

31. Conclusion CLOP is conceptual framework for API mash-up.
CLOP must be flexible, and will include all.
ANN can utilize big data.
ICT companies should provide open API.
Let’s make big data together.
Let’s make API of agricultural apps.
Let’s open “How to use API”.