IRJET- Crop Yield Prediction and Disease Detection using IoT Approach

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 06 | June 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 334
Crop Yield Prediction and Disease Detection Using IOT Approach
Akshada Sakhare1, Tanuja Patil2 , Priti Giri3, Riya Gulame4
1,2,3,4Computer Engineering, PDEA’s College of Engineering
5Prof. R.B. Rathod Dept. of Computer Engineering, PDEA’s College of Engineering ,Maharashtra, India.
-----------------------------------------------------------------------***--------------------------------------------------------------------
Abstract - Today energy resources are becoming
inadequate and therefore more worthwhile. In association
with the population advance over last century, the need
for finding new, more decisive, and continual methods of
agricultural farming and food production has become
more biting. To simplify this process, we are designing,
building, and classifing a system for rigor agriculture
which afford farmers with useful data about the disease
prevention, the water stockpile, and the general
information of the diseases in a good enough, easily
available manner. Our system aims to make farming and
sprinkling more worthwhile as the farmer is able to make
better informed decisions and thus save time and
resources. The diversity of location and climatic effects
upon agricultural farming, along with other environmental
parameters over time makes the farmer’s decision-making
process more difficult and requires additional factual
grasp. Implementing wireless sensor networks for
observing weather parameters and bringing together this
information with a user-personalized service may enable
farmers to deed their knowledge in an dynamic way in
order to get the best results from their agricultural
farming. The system can scale based on each farmer’s
claims and the resulting imposition of collected
information may express a beneficial resource for
forthcoming use, in addition to its use for real-time
decision making. The design of the precision agriculture
system consist of a prototype solution about the sensor
platform and a personalized service that can be used in
diverse ways and by several articles.
Keywords:
Graphical passwords, Social engineering, Distortion
I. Introduction
As the world is contributing towards advanced
technologies and fulfillment it is a paramount goal to trend
up in agriculture also. Many researches are completed in
the field of agriculture. Most projects proclaim the use of
wireless sensor network collect data from different
sensors use at various nodes and send it over the wireless
obligation. The collected data grant the information about
the assorted environmental factors. Observing the
environmental factors is not the outright result to raise the
yield of crops. There are number of other factors that
reduce the yield to a greater quantity.
In India around 80% of people depend upon farming.
Smart Agriculture is one of the clarification to this
problem. To representing appearance of this project
concludes water
Management, weather forecasting, canal controlling in
both automatic and manual modes and all these data are
saved and visible in a mobile application. The alert SMS
and notification is sending to the user based on the fixed
benchmark. By regulating all these operations by a mobile
which is connected to internet and it will supply better
performed by interfacing sensors, Wireless Fidelity etc.
II. Literature Survey
There are many existing strategies and approaches are
used for prediction and detection of disease, for that
different types of algorithms are implemented.
[6]It is systems which can predict the more accuracy using
meteorological data. Nowadays, there are a lot of yield
prediction models, that more of them have been generally
classified in two group: a) Statistical Models, b) Crop
Simulation Models of Artificial Intelligence (AI),
[8]This Recently, application research aimed to assess
these new data mining techniques and apply them to the
various variables consisting in the database to establish if
meaningful relationships can be found.
[14]The results of this study indicate that the ARMA model
is preferable over other time series models considered in
this paper. The implication of the finding in this study is
significant for insurance underwriters responsible for
constructing area-based yield insurance that can benefit
the Micro insurance market of smallholder farmers and for
institutions that rely on those forecasts in providing
capital.

III. Proposed Methodology
This project is implemented using Rasberry Pi as a
controller. Here we are using Hardware like moisture
sensor and Motor On and off switch.
In this work the experiments are executed significant and
well admitted grade algorithm KNN are enforced to the
dataset. There veracity is obtained by assess the datasets.
In recommended system the farmer will enter his crop
name in the system and when system detect the climate or
weather change ,then System will automatically predict
and advise the farmer that which disease will taint to your
crop as well as the system will also gives a distinct
methods to prevention.
This project support us to conduct the moisturize level
and where we can use in the Society easily. The percentage
of moisturize is preserve by sensor which is present inside
the soil and the data will store in the database using
mobile application. Confer to that motor will be work
automatic and manual. If the moisturizer level is low
automatically motor gets switched on if it’s up to fill ,Then
it will shut down the motor. Apart of this the farmer will
also get disease information by only putting the disease
name in the system. This information include the
prevention methods ,how to cure the disease, which plants
can be affected by this disease and on which weather it
can be happen or have the chances to come etc. along with
their images.
Fig. (1) Architectural Diagram of system
1. Soil Moisture Sensor:
The volumetric water present in soil is measured by
moisture sensors.In consideration of the
explicit geometric assessment of free moisture which
build upon removing, drying, and weighting of a sample,
moisture sensors sense the volumetric water
circumlocutorily with the help of some other property of
the soil, like as electrical resistance, dielectric constant, or
synergy with neutrons, as a proxy for the moisture
content. The affiliation among the sensed property and
soil moisture must be graded and may differ depending on
substancial factors such as soil type, temperature,
or electric conductivity. Reverse
transfered microwave radiation is afflicted by the soil
moisture and is used for remote sensing in hydrology and
agriculture. lightweight probe equipments can be pre-
owned by farmers or gardeners.
Soil moisture sensors have general attribute to sense
volumetric water . Another class of sensors sense other
property of moisture in soils known as water potential.
These sensors are commonly referred as soil water
potential sensors and incorporate densitometers and
gypsum blocks.
Fig. (2) Soil moisture sensor
2.Rasberry PI:
Raspberry Pi is a basic computer that was originally
intended to help spur interest in computing. Raspberry Pi
is single circuit board which provide ability to mix and
match software according to the work they wish to do.
There are some sensors for the Raspberry Pi that can
measure humidity, temperature and other values. The
value received from sensor is processed by rasberry pi
and the conditioned signal pass to the data base for
further acquisition.

Fig. (3) Rasberry Pi
3. Data Base:
MySQL: MySQL, is the most famous Open Source SQL
database management system, is advanced, dispersed, and
backed by Oracle Corporation. The MySQL Web site
(http:www.mysql.com) administer the immediate prior
information regarding MySQL software.
MySQL is a database management system.
A database is a efficient batch of data. It may be any thing
at all from a simple shopping bill to a picture gallery or
the broad amounts of information in a collective network.
To reckon, approch, and process data stored in a computer
database, you need a database management system such
as MySQL Server. Since computers are excellent at
handling huge amounts of data, database management
systems play a fundamental role in computing, as stand
alone service, or as parts of other applications.
4. Android application:
An Android app is a software application fuctioning on the
Android podium. Because the Android podium is built for
mobile like devices, a typical Android app is invented for a
smartphone or a tablet PC functioning on the Android OS.
5. KNN Algorithm:
A simple way of accomplish this is to use K-nearest
Neighbor.
K-nearest neighbor algorithm (KNN) is element of
administered research that has been used in many
applications like field of data mining, statistical pattern
understanding and many others.
KNN is a approach for allocating objects based on closest
training examples in the feature space.
An object is allocated by a mass vote of its neighbors. K is
always a positive integer. The neighbors are taken from a
set of objects for which the proper allocation is
recognized.
It is normal to use the Euclidean distance, though other
distance measures such as the Manhattean distance could
in principle be used instead.
Points given to compute algorithm on K-nearest neighbors
are as follows:
1. Calculate the parameter K = number of nearest
neighbors ahead. This value is eliptic or acceptable.
1. 2. Calculate the distance between the query-
instance and all the training samples. You can use any
distance algorithm.
2. 3. Sort the distances for all the training samples
and determine the nearest neighbor based on the K-th
minimum distance.
3. 4. Since this is supervised learning, get all the
Categories of your training data for the sorted value which
fall under K.
4. 5. Use the majority of nearest neighbors as the
prediction value.
Advantages:
Robust to noisy training data (especially if we use inverse
square of weighted distance as the “distance”)Effective if
the training data is large.
K-nearest neighbor (Knn) algorithm pseudocode:
Let (Xi, Ci) where i = 1, 2……., n be data points. Xi denotes
feature values & Ci denotes labels for Xifor each i.
Assuming the number of classes as ‘c’
Ci ∈ {1, 2, 3, ……, c} for all values of i
Consider x is a point for which label is uknown, and we
would like to determine the label class using k-nearest
neighbor algorithms.
Knn Algorithm Pseudocode:
1. Calculate “d(x, xi)” i =1, 2, ….., n; where d denotes
the Euclidean distance between the points.
2. Arrange the calculated n Euclidean distances in
non-decreasing order.

3. Let k be a +ve integer, take the first k distances
from this sorted list.
4. Find those k-points corresponding to these k-
distances.
5. Let ki denotes the number of points belonging to
the ith class among k points i.e. k ≥ 0
6. If ki >kj ∀ i ≠ j then put x in class i.
6. MATHEMATICAL MODEL:
Let W be the whole system which consists
Input = {U, M, dp, D}.
1. Let u is the set of number of users or Farmer.
U= {U1, U2….Un}.
2. M is the moisturize level of soil
3. Dp is a crop name .
4. D is a disease name
Procedure:
Step 1: The farmer will enter his crop name in the system
and when system detect the climate or weather change
,then System will automatically predict and notify the
farmer that which disease will infect to your crop as well
as the system will also gives a different methods to
prevention.
Step 2:When sensor sense the moisture in the soil,
according to that motor will be work automatic and
manual. If the moisturizer level is low automatically motor
gets switched on if it’s up to fill ,then it will shut down the
motor.
Step 3: The farmer will enter the disease name and system
will return all the information related to that disease
(including their prevention methods, cure methods and
images)
Output: Predict the weather and provide disease
prevention methods, turn motor on/off by sensing the
moisturize level, disease information.
A. Registration
 The Farmer will register to the system with
normal information.
 At the time of registration Farmer will enter valid
Email-ID and Password
B. Login
 For login to the system, Farmer will enter the
Email and password, if entered details are correct
then the system will redirect him to home page
otherwise it will shows an error message.
After Login:
1. Disease prevention:Farmer will enter the his crop
name.
On climate change the system will predict which
disease will affect the crop and notify the farmer
what prevention methods he has to take.
2. Moisturizer
The Sensor sense the moisturize level of the soil
and turn the motor on and off accordingly.
3. Disease information
The farmer will enter the disease name as input and
system will return all the information related to that
disease as an output. This information includes
disease prevention methods, cure methods their brief
information along with the images.
C. Logout
Farmer Logout from system.
Hardware Requirements:
System : Pentium IV 2.4 GHz.
Hard Disk : 40 GB.
Ram : 512 Mb.
Software Requirements:
 Operating system : Windows XP/7.
 Coding Language : JAVA/J2EE, Hibernate.
 IDE : Java eclipse, Android.
 Web server : Apache Tomcat 7.
 Front End : JSP, CSS etc.

 Back End : MySQL as database
server.
IV. Prototype model:
Fig. (4) Prototype model
Initially, the soil moisture sensor detect the certain
parameters which include moisture, phosphorus,and
nitrogen present in the soil.we get required ratios as in
form of input values which will help us for prediction.
The prototype model has 3 buttons on it , when we put
sensor in soil that time it shows the NPQ ratio on mobile
application. That NPQ ratio will setup by using these 3
buttons on model as per requirement.
Secondly, motor get started and there water level is
calculated.And simultaneously the wheather will detected
through the mobile application,by the detection of
wheather the approximate prediction is done about which
type of disease will affect the crops .This will provide
proper guidelines and accordingly farmers will give
proper prevention and cure to save the crops.
Lastly , the predicted disease will found out and required
solution will given form data base.
Conclusion
In the advance, a Novel System Facilitate: IoT Based
Agriculture Stick for Live adherence Soil Moisture has
been recommended using Arduino or Rasberry pi 3. The
sensors has high efficiency and accuracy in fascinating the
live data of soil moisture. The system enables effective
soil, water, moisture, parameters has been observing and
updating using IOT. This implement adequate soil
maintenance and Disease prevention mechanism. This
conquered the manual operations required to observe and
maintain the agricultural farms in both automatic and
manual. The system enables the farmer to search about
the various diseases.
References
[1]Adams, R., Fleming, R., Chang, C., McCarl, B., and
Rosenzweig, 1993 ―A Reassessment of the Economic
Effects of Global Climate Change on U.S. Agriculture,
Unpublished: September.
[2] kevin bhalodia, mansing rathod,2018
android application for crop yield prediction and crop
disease detection
[3] rushika ghadge, juilee kulkarni , 2018 prediction of
crop yield using machine learning
[4]Adaptation to Climate Change Issues of Longrun
Sustainability." An Economic Research
[5]Barron, E. J. 1995."Advances in Predicting Global
Warming‖.The Bridge (National Academy of Engineering).
25 (2, Summer): 10-15.
[6] Raorane A.A.1, Kulkarni R.V.2
Agricultural Crop Yield Prediction Using Artificial Neural
Network Approach
[7]Basu, Majumder, A., Bera, B. and Rajan, A. 2010.
Teastatistics: Global scenario. Int. J. Tea Sci.8: 121-124.
[8]Raorane A.A.1, Kulkarni R.V. Data Mining: An effective
tool for yield estimation in the agricultural sector
[9]Brack, D. and M. Grubb. 1996. Climate Change,
"ASummary of the Second Assessment Report of the
IPCC."FEEM (Fondazione ENI Enrico Mattei, Milano
Italy)newsletter, 3, 1996
[10] M.Soundarya, R.Balakrishnan,” Survey on
Classification Techniques in Data mining”, International
Journal of Advanced Research in Computerand
Communication Engineering Vol. 3, Issue 7, July 2014.
[11] D Ramesh , B Vishnu Vardhan, “Data mining
technique and applications to agriculture yield data”,
International Journal of Advanced Researchin Computer

and Communication Engineering Vol. 2, Issue 9,
September 2013 .
[12] Gideon O Adeoye, Akinola A Agboola, “Critical levels
for soil pH, available P, K, Zn and Mn and maize ear-leaf
content of P, Cu and Mn insedimentary soils of South-
Western Nigeria”, Nutrient Cycling in Agroeco systems,
Volume 6, Issue 1, pp 65-71, February 1985.
[13] D. Almaliotis, D. Velemis, S. Bladenopoulou, N.
Karapetsas, “Appricot yield in relation to leaf nutrient
levels in Northern Greece”, ISHS ActaHorticulturae 701:
XII International Symposium on Apricot Culture and
Decline .
[14] Askar Choudhury, Illinois State University James
Jones, Illinois State University Crop yield prediction using
time series models

IRJET- Crop Yield Prediction and Disease Detection using IoT Approach

More Related Content

What's hot (20)

Similar to IRJET- Crop Yield Prediction and Disease Detection using IoT Approach (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Crop Yield Prediction and Disease Detection using IoT Approach