Using Cloud IOT For Disease Prevention in Precision Agriculture Using Cloud IOT For Disease Prevention in Precision Agriculture

Available online at www.sciencedirect.
com
Available online at www.sciencedirect.com
ScienceDirect
ScienceDirect
Available online at www.sciencedirect.com
Procedia Computer Science 00 (2018) 000–000
ScienceDirect www.elsevier.com/locate/procedia
www.elsevier.com/locate/procedia
The 9th International Conference on Ambient Systems, Networks and Technologies

The 9th International Conference on(ANT
Ambient
2018)Systems, Networks and Technologies
(ANT 2018)
Using Cloud IOT for disease prevention in precision agriculture
Using Cloud IOT for disease prevention in precision agriculture
Karim Foughaliaa* , Karim Fathallahbb, Ali Frihidabb
a
Karim Foughali * , Karim Fathallah , Ali Frihida
Communication System Laboratory Sys’Com, National Engineering School Of Tunis, University Tunis El Manar, Tunisia
a b
Communication System National Engineering
Laboratory Sys’Com,School Of Engineering
National Tunis, University
SchoolTunis El Manar,
Of Tunis, TunisiaTunis El Manar, Tunisia
University
b
National Engineering School Of Tunis, University Tunis El Manar, Tunisia
Abstract
Abstract
The application of decision support system (DSS) for potato late blight disease prevention has proven its benefit. In fact the DSS
The application
permits of decision
efficiency, support
minimizes cost andsystem (DSS) forimpact
environment potatoby
late blight disease
estimating prevention
the exact has proven
requirement its benefit.
fungicide quantityIntofact the DSS
apply. This
permits efficiency,
prediction minimizes
using weather cost and
condition environment
based late blightimpact bymodel.
forecast estimating
The the exact requirement
required fungicideisquantity
weather information collectedto from
apply.costly
This
prediction usingorweather
weather station condition
imprecise based
historical late
data. blight forecast
However, with themodel. The required
emergence weather
of the IOT, hugeinformation
number of lowis collected
cost andfrom costly
low power
weather stationcan
sensors nodes or easily
imprecise historicalindata.
be deployed However,
farmlands with to
in order thegather
emergence of the
a precise IOT, data.
climate huge Moreover,
number of the
lowcollected
cost and data
low can
powerbe
sensors nodes
forwarded can easily
by Internet be deployed
connection to thein so
farmlands in order
called cloud IOTtoframework.
gather a precise
In thisclimate data.present
paper we Moreover,
a newthe collectedofdata
prototype late can be
blight
forwarded
prevention by Internet
decision connection
support systemtobased
the sooncalled
sensorcloud IOTand
network framework. In this paper we present a new prototype of late blight
cloud IOT.
prevention decision support system based on sensor network and cloud IOT.
© 2018 The Authors. Published by Elsevier B.V.
© 2018 The Authors. Published by Elsevier B.V.
© 2018 The under
Peer-review Authors. Published by
responsibility of Elsevier B.V. Program Chairs.
the Conference
Conference
Peer-review under responsibility of the Program Chairs.
Peer-review under responsibility of the Conference Program Chairs.
Keywords: WSN; Cloud-IOT; DSS; Late blight.
Keywords: WSN; Cloud-IOT; DSS; Late blight.
1. Introduction
1. Introduction
Potato late blight has always been the most dangerous disease for potatoes in our regions 1. Since its appearance,
1
latePotato
blightlate
has blight has always
continued to worrybeen the mostSince
producers. dangerous disease
the mid 1970s, foritspotatoes
impact in
onour regions
crop . Since and
productivity its appearance,
quality has
late
beenblight hasincreasing.
steadily continued Theto worry producers.
disease Since
is declared the than
earlier mid in
1970s, its impact
the past, on crop productivity
the aggressiveness is higher and
and quality has
the control
been steadily
strategies put increasing.
in place areThe diseaseexceeded
regularly is declared
byearlier thanofin the
the scale the problem.
past, the aggressiveness is higher can
Phytophthora infestans and destroy
the control
the
strategies
foliage in put
a fewin days.
place itarecan
regularly exceeded
contaminate byand
tubers the make
scale them
of theunfit
problem. Phytophthora
for marketing: infestans
the loss can
can be destroy and
complete the
foliage in a few days. it can contaminate tubers and make them unfit for marketing: the loss can be complete and
* Corresponding author. Tel.: +21696488622; fax: +0-000-000-0000 .

* E-mail address:author.
Corresponding Tel.: +21696488622; fax: +0-000-000-0000 .
Foughali.karim@gmail.com
E-mail address: Foughali.karim@gmail.com
1877-0509 © 2018 The Authors. Published by Elsevier B.V.
Peer-review
1877-0509 ©under
2018responsibility
The Authors. of the Conference
Published Program
by Elsevier B.V. Chairs.
1877-0509 © 2018 The Authors. Published by Elsevier B.V.

10.1016/j.procs.2018.04.106
576 Karim Foughali et al. / Procedia Computer Science 130 (2018) 575–582
2 Karim Foughali/ Procedia Computer Science 00 (2018) 000–000
very quick. To fight this terrible disease, farmers are forced to devote a significant part of their means of production,
including the repeated use of phytosanitary products. The economic consequences (cost of products and their
application on crops), environmental (risks of drift of fungicidal slurries, quantities and risks of the products
themselves), social (time of monitoring and implementation of means of protection by farmers) and societal (image
of agriculture damaged by the repeated presence of spraying equipment in the fields) are important It is also
necessary to take into account the losses caused to the crops by the mildew in spite of this fungicidal protection.
Historically, it has been considered that fungicides are sufficient to control this pathogen, this statement having
shown in recent years certain limits, especially during wetter summers or difficult conditions at the beginning of the
season (winter without frost, presence of many regrowths in other cultures, problem of piles of sorting gap, ...). The
frequency of years with very strong pressures of mildew is higher and higher.
The ever-increasing aggressiveness of late blight populations, coupled with the almost widespread use of
susceptible varieties, is leading to serious contamination of crops despite the repeated and massive use of plant
protection products. In this context, thanks to the information they integrate , Decision Support Systems (DSS) help
farmers implement effective treatment programs.
These allow farmers to increase the effectiveness of control methods without increasing the risks. In other words,
the DSS do not systematically seek to reduce the number of treatments but to control the mildew effectively (with a
sufficiently wide margin of safety). DSS increases the effectiveness of control methods and can also be used to
justify the application of fungicides.
The latter DSS require the collection of weather data from a costly weather station infrastructure or based of
historical data. With the climatic change, historical data has become more and more useful in order to a precise
forecasting of the climatic condition. Moreover in developed countries we can't rely on weather forecasting
infrastructure in order to receive more precise climatic information.
Based on the latter observation the state of art DSS can't be useful in North African countries. The advent of the
Internet of Things enabled the interconnection of tremendous numbers of tiny and low cost climatic sensors. This
offers the possibility to farmers’ community to deploy a low-cost weather station network on their farmlands.
Moreover a new emerged service like Cloud-IOT enables a freely accessible web-service which permits a plug and
play of the remotely deployed sensors. The objectives of this study were to develop and implement a Cloud-IOT
DSS for late blight enabling the use of location-specific weather data to drive disease forecasters and a mechanistic
model of the late blight disease, in order to provide real-time (in-season) support for late blight management.
2. Related works
IOT in agriculture
In order to increase agricultural production, several IOT platforms have been developed for crop monitoring,
disease prediction, and control of water consumption in irrigation. In 2 Balamurugan et al proposed an IOT
application to control the evolution of temperature and soil moisture from sensor network deployed in the
agricultural field using an IOT platform based on Raspberry Pi thus scavenging the analysis and monitoring the data
received. In 3 Min-ShengLiao et al have developed an IOT application for monitoring environmental factors in
orchids greenhouse agricultural field. This application integrates a system based on the image processing of orchid
leaves allowing the follow-up and analysis of leaf growth in real time. In 4 Payero et al proposed a low-cost IOT
application for monitoring soil moisture in a wheat field. This system is based on a network of sensors and an IOT
platform, the latter displays in real-time the values of soil moisture.
Light blight DSS
In the field of agriculture, several decisions support systems have been implemented to help farmers control their
crops and to implement effective treatment programs against diseases specifically against late blight disease. Among
them can be mentioned PLANT-Plus5, MILEOS6, NegFry7, ProPhy8 these models are based on climatic variables to
predict late blight and in case of infection, evaluate and measure the evolution of the disease. In our study we will
Karim Foughali et al. / Procedia Computer Science 130 (2018) 575–582 577
Karim Foughali/ Procedia Computer Science 00 (2018) 000–000 3
use the SIMCAST9,16 model given its simplicity and efficiency, this model has as input the number of consecutive
hours where the humidity is greater than 90% for a day and the average temperature. Based on these meteorological
data it assesses the risk of occurrence of the disease by calculating a parameter called "Blight units" the latter when
it reaches a defined threshold is notified agriculture to begin treatment against the disease of late blight. The
SIMCAST model incorporates two logistic functions (2) (3). we used a common approach to simulate the
temperature dependence of the growth of plants, the beta function (1) (2) which uses the three temperatures
characteristic for the simulated species, the minimum temperature T m, the maximum temperature T M and the
optimum temperature To for growth (in our case light blight disease development). The increase of blight units with
length of the period is sigmoidal, hence we approached this with a sigmoidal function (3).
log( 2)
 (1)
T -T
log( M m )
T0 - Tm
Where Tm =-1.9453 represent the minimal temperature for growth

TM =29.6298 represent the maximal temperature
T0 =17.1268 represent the optimal temperature
2(T - Tm ) (T0 - Tm )  (T - Tm ) 2
 (t )  (2)
(T - Tm ) 2
a
s(t )  (3)
1  e (bct )
Where a=7.235, b=0.5270 , c=4.6006
3. Background
3.1. Precision agriculture
Today, agriculture is facing several problems such as water scarcity, plant diseases and so on. Wireless sensor
networks can be a solution to deal with these problems. Wireless sensor networks can monitor different
environmental factors, such as temperature and humidity, to maximize production and minimize costs. It makes it
possible to decide the right quantity of the fertilizers at the good moment which makes it possible to reduce the
costs, the quantity of water necessary for the plants. In addition, this system14,15 allows farmers to remotely control
their plants in a computerized manner.
3.2. WSN
A wireless sensor network is composed of a set of on-board processing units, called "motes", communicating via
wireless links. WSN collects a set of environmental parameters surrounding the motes, such as temperature,
humidity, and pressure. This collected data will be transmitted to the gateway which provides a connection to the
wired world where we can collect, process, analyze and present the measurement data via a web interface or via an
IOT platform. Wireless Sensor Networks are a first-class tool for better understanding of cropping processes by
providing real-time measurements that are essential for optimal management of agricultural parcels. WSN enables
remote monitoring of large geographic areas without infrastructure without the need for expensive cabling, and
allows many localized parameters (temperature, humidity, freeze detection, failures, etc.) to be identified in a very
short period of time. This data is routed to a decision center that can act in an automated way by sending commands
remotely.
3.3. Cloud IOT
With new Standard 6LowPan7, all previously unconnected wireless sensor networks could integrate directly
using public IPv6 addresses on each node. Millions of devices are now connected to the internet. A new generation
of the Internet has appeared and a great title is already announced "the Internet of Things". With this latest concept
many Cloud-IOT like (Xively10, ThingWors11, ubidots12 ) have been used to manage connected devices and track
their operating status. These platforms IOT offer several services to users such as the visualization of data from
different sensors, the storage and the security of the data collected.
4. Architecture system Design
The architecture of our supervision system see Fig. 1 is composed of three thirds. A third WSN containing the
various nodes deployed in our field of study, a third gateway which conveys the data collected from the network via
internet, and a third based on a cloud-IOT and a DSS to analyze and display the data.
Fig. 1. Architecture system Design.

4.1 Tier of sensor Network
In our application this tier aims to collect information from the sensors. For measurements, we deployed five
wasmposte nodes. . Each sensor node contains a medium (Waspmote 868 SMA 4.5 DBI), which is interconnected
with the XBee 802.15.4 Pro SMA 5dBi module that acts as an IEEE 802.15.4 communication interface and a
Waspmote agriculture sensor board PRO to define the links between different sensors. each node contains a
temperature sensor and a humidity sensor. These sensors communicate via a Zigbee interface (using XBee802.15.4
SMA 5dBi) and interconnected with a Gateway which allows the transfer of the measurements collected by the
sensors to the cloud-IOT platform.
4.2 Tier Gateway
The gateway is a device that has the distinction of having two network interfaces. It connects the wireless sensor
network to a more traditional network, in our internet network internet. Indeed, usually the sensor network is only
used to trace the measurements, the applications processing this information being executed on the machine of the
end user. In this application we used a gateway named Meshlium the latter is a multi-protocol router since it has five
interfaces (2.4GHz Wifi, 5GHz Wifi, Blutooth, ZigBee and 3G / GPRS) is used to collect all data from sensor nodes
and store them in the cloud.
4.3 Tier IOT cloud
In the paradigm of the Internet of Things (IoT), many of the objects around us will be on the network in one form
or another. These results in the generation of huge amounts of data that must be stored processed and presented. As
a result, several platforms have been put online offering users several services such as storage and filtering of data
from different devices. In our application we used the platform Ubidots which allows on the one hand to supervise
the data coming from the wasmpote network in real time and on the other hand offers a notification service to warn
the farmers. These alerts are based on the SIMCAST prediction model
5. Deployment
5.1. Flied description
The deployment field is located in the Ras Jebel region in northern Tunisia. Like other regions of Tunisia, this
region is facing a problem of resource degradation and, among other things, contamination of the drinking water on
which a large part of its rural population relies for its supply and the emergence of late blight disease that infects
potato crops. As a result, it has served as a test case for conducting investigations and analyzes with a view to
developing an evaluation and intervention strategy that can subsequently be applied to other agricultural
productions.
5.2. Implementation step
The application we have developed consists of three parts the first part is to collect the data from the sensors
deployed in the potato field and transfers them to the Gateway. The second part is to transfer the data to the cloud
and the third part is to analyze the data and integrate a prediction model "SIMCAST" to predict the risk that could
affect the plants. To carry out this process we began by programming the sensor nodes by flashing a code written in
C language, and then we implemented the SIMCAST model at the IOT platform. The latter is described by the
algorithm below which has as input the average temperature per day and the number of hours where the humidity
exceeds 90% and as output blight units.
Algorithm:
Int Function Calculate_Blight_Unit (NumberHour, Temp)
{
IF temp>=23 and temp<=27 THEN

{
if (NumberHour>=7 and NumberHour<=9) THEN { Return (1) }
}
{
}
{
if (NumberHour =7 ) THEN { Return (1) }

if (NumberHour=10) THEN { Return (3) }
}

{
}
Return(0)
5.3. Monitoring Interface
The user interface represents a dashboard for agriculture this interface displays to the user the different nodes
deployed in the agricultural field fig 3. In addition this interface represents the evolution of the different values in
real time of the environmental parameters the temperature and the humidity as indicated in fig 2. Depending on the
number of consecutive hours in which the humidity is greater than 90% during a day and the average temperature,
the SIMCAST model evaluates the risk of appearance of the disease by calculating a parameter called "Blight unit".
When blight unit reaches threshold 30 we have programmed a trigger to notify agriculture fig.4 to start treatment
against late blight disease.
Fig. 2. Temperature and humidity values
Fig. 3. Monitoring interface
Fig. 4. Alert interface.

6. Conclusion
In this work has been presented a system of decision support for Late Bight disease, using the information they
integrate to help the agriculturists implement effective treatment of the disease. In this developed application we
used a sensor network and a cloud server to store the information about temperature and humidity. Then using a
decision support system to notify the farmer by SMS when the first attack of disease "Late Blight" will take place.
We can improve our application by developing an IOT platform that collects the images of the plants and analyzes
the sudden transformations during the development of the disease either on the leaves in the form of pale brown and
pale green margins or on the leaves. The fungus leaves brown spots that can cause the destruction of young plants or
the breakage of adult stems.
References
1. Hansen J.G., Lassen P. Monitoring of potato late blight caused by Phytophthora infestans (Mont.) de Bary (Pi-Monitoring ver 1.1.1. User
Manual). Danish Institute of Agriculture Sciences, 2000, p 24.
2. C.R. Balamurugan ; R. Satheesh. Fronteiras. Development of Raspberry pi and IoT Based Monitoring and Controlling Devices for Agriculture
Journal of Social, Technological and Environmental Science v.6, n.2, may-august. 2017 .p. 207-215
3.Min-ShengLiao, Shih-Fang Chen, Cheng-Ying Chou, Hsun-Yi Chen, Shih-Hao Yeh, Yu-Chi Chang, Joe-Air Jiang:
On precisely relating the growth of Phalaenopsis leaves to greenhouse environmental factors by using an IoT-based monitoring
system. Computers and Electronics in Agriculture;2017. p. 125-139.
4. Payero, J.O., Mirzakhani-Nafchi, A., Khalilian, A., Qiao, X. and Davis, R. (2017) Development of a Low-Cost Internet-of-Things (IoT)
System for Monitoring Soil Water Potential Using Watermark 200SS Sensors. Advances in Internet of Things, Vol.7 No.3, July 2017 71-86.
5. Jan Hadders. Experience with a late blight DSS (Plant-Plus) in a starch potato area of theNetherlands in 1995 and 1996. First Workshop of an
European Network for Development of an Integrated Control Strategy of potato late blight Lelystad, The Netherlands, October 1996.
6. Jaunatre V, Gaucher D , Mileos a decision-making aid that evolves. 4ème Conférence Internationale sur les Méthodes Alternatives en
Protection des Cultures Evolution des cadres réglementaires européen et français Nouveaux moyens et stratégies Innovantes. Nouveau Siècle,
Lille, France..2011. pp. 104–108.
7. Leonard, R., L.J. Dowley, B. Rice & S. Ward, Comparison of the NegFry decision support system with routine fungicide application for the
control of potato late blight in Ireland. Potato Research. 2001. pp 327–336.
8. Spits HG, Wander JGN, Kessel GJT . DSS development focused on variety resistance in The Netherlands 2003. Proceedings of the eighth
workshop of an European network for development of an integrated control strategy of potato late blight. PPO special report no 10, pp 215–
222
9.. Grunwald , G. R. Montes, H. L. Salda~na, O. R. Covarrubias, W. E. Fry. Potato late blight management in the toluca valley: Field
validation of simcast modified for cultivars with high field resistance, Plant Disease, 2002. vol. 86, no. 10, pp. 1163-1168.
10. N. Sinha, K. E. Pujitha and J. S. R. Alex, "Xively based sensing and monitoring system for IoT," Computer Communication and Informatics
(ICCCI), 2015 International Conference on, Coimbatore, 2015, pp. 1-6.
11. T. Baranwal, N. and P. K. Pateriya. Development of IoT based Smart Security and Monitoring Devices for Agriculture. 6th International
Conference - Cloud System and Big Data Engineering, 2016
12. Francisco Javier Ferrández Pastor, Juan Manuel García Chamizo, Mario Nieto-Hidalgo, Jerónimo Mora Pascual, José Mora-Martínez:
Developing Ubiquitous Sensor Network Platform Using Internet of Things: Application in Precision Agriculture. Sensors 2016.
13. G. Mulligan, The 6LoWPAN architecture, Proceedings of the 4th workshop onEmbedded networked sensors - EmNets ’07, 2007.p. 78
14. Patil V.C., Al-Gaadi K.A., Biradar D.P. Rangaswamy M: Internet of things (Iot) and cloud computing for agriculture: An overview.
Proceedings of Agro-Informatics and Precision Agriculture . 2012, pp 292-296.
15. Shi Y., Wang Z., Wang X., Zhang S. Internet of Things Application to Monitoring Plant Disease and Insect Pests. International Conference
on Applied Science and Engineering Innovation .2015 ,p 31-34.
16. Wang E, Engel T. Simulation of Phenological Development of Wheat Crops. Agric Syst. 1998; pp 1-24.

Using Cloud IOT For Disease Prevention in Precision Agriculture Using Cloud IOT For Disease Prevention in Precision Agriculture

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Using Cloud IOT For Disease Prevention in Precision Agriculture Using Cloud IOT For Disease Prevention in Precision Agriculture

Încărcat de

Drepturi de autor:

Formate disponibile

Available online at www.sciencedirect.

The 9th International Conference on Ambient Systems, Networks and Technologies

* Corresponding author. Tel.: +21696488622; fax: +0-000-000-0000 .

1877-0509 © 2018 The Authors. Published by Elsevier B.V.

Light blight DSS

Where Tm =-1.9453 represent the minimal temperature for growth

3.1. Precision agriculture

3.3. Cloud IOT

4. Architecture system Design

Fig. 1. Architecture system Design.

4.1 Tier of sensor Network

4.2 Tier Gateway

4.3 Tier IOT cloud

5.1. Flied description

5.2. Implementation step

IF temp>=23 and temp<=27 THEN

if (NumberHour =7 ) THEN { Return (1) }

IF temp>=3 and temp<=7 THEN

5.3. Monitoring Interface

Fig. 2. Temperature and humidity values

Fig. 3. Monitoring interface

Fig. 4. Alert interface.

S-ar putea să vă placă și