IMPLEMENTATION OF MARRIAGE BY INDIGENOUS LAW TO YEI TRIBE COMMUNITIES

http://www.iaeme.com/IJMET/index.asp 37 editor@iaeme.com
International Journal of Mechanical Engineering and Technology (IJMET)
Volume 10, Issue 12, December 2019, pp. 37-47, Article ID: IJMET_10_12_005
Available online at http://www.iaeme.com/ijmet/issues.asp?JType=IJMET&VType=10&IType=12
ISSN Print: 0976-6340 and ISSN Online: 0976-6359
© IAEME Publication
ARDUINO BASED SMART IRRIGATION
SYSTEM AND PLANT LEAF DISEASE
DETECTION USING MATLAB
D. Rama Prabha, Ram Swaminathan, Kalepu Chaitanya, W. Razia Sultana*
Department of Electrical and Electronics Engineering,
Vellore Institute of Technology, Vellore, Tamil Nadu, India
*Corresponding Author Email: wraziasultana@vit.ac.in
ABSTRACT
The agriculture sector in India is going through a major crisis. One of the major
reason for this crisis, is lack of technological advancement in the farming sector, only
around 34% of the cultivated land is reasonably irrigated. Also, traditional method of
irrigation is very inefficient, leading to water wastage as well as other problems such
as water clogging, erosion of top soil etc. Another major reason is late identification
of plant diseases, which leads to partial or complete destruction of the crops. This
paper aims to address the issues by creating a model which can efficiently irrigate the
crops, thereby reducing energy, labour and water consumption. The system can also
identify potential diseases in the plants by observing the symptoms shown on the
leaves. The aim is to create a feasible, low cost implementable model, therefore by
using HC-05 modules; a wireless network has been created. The above proposed
system will also relay the moisture level content in soil and the status of motor from
time to time. This system uses micro-controller board based on ATmega328. It can
also help to detect the type of diseases occurring, by analysis of the leaves using K-
means segmentation algorithm in the MATLAB software. By the use of this system,
assistance in the optimum growth of plants is possible, can increase the yield and also
lessen the frequency of visit to the fields, enabling the farmer to focus on other
agricultural activities.
Keywords: Arduino, Irrigation, Disease detection, GSM module, hc-05 module
Cite this Article: D. Rama Prabha, Ram Swaminathan, Kalepu Chaitanya,
W. Razia Sultana, Arduino Based Smart Irrigation System and Plant Leaf Disease
Detection Using MATLAB. International Journal of Mechanical Engineering and
Technology 10(12), 2019, pp. 37-47.
http://www.iaeme.com/IJMET/issues.asp?JType=IJMET&VType=10&IType=12
1. INTRODUCTION
Agriculture is one of the most important sectors in India; around 60% of the total population
still depends on agriculture for their livelihood. The sector is facing major perils due to
various factors; one of them is lack of proper irrigation systems in the fields. In a tropical
monsoon country such as India, where the rainfall is scant and unreliable, and irrigation

D. Rama Prabha, Ram Swaminathan, Kalepu Chaitanya, W. Razia Sultana
becomes an important agricultural input. Also, it is important to create a scientific and
technically advanced irrigation method so that large tracts of land are not affected by salinity,
alkalinity and water logging-the ill effects of outdated practices. Currently, there are many
irrigation models which are coming up in this field, but their adoption rate is very less. As
most of the systems involve high investment and are complex in nature. Hence, there is a need
to create a low-cost, feasible system which can address the above issues and will benefit the
small and medium scale farmer’s. Also, an application has been created which can help in
detecting the diseases in the early stages, by carefully analysis of the affected region of the
leaves. Thereby, helping to identify the crop damaging diseases and this will in turn aid in the
process of taking preventive measures so that there is optimal growth of the plant and
productivity is high.
2. LITERATURE REVIEW
There has been work done in this field to improve the agriculture sector. By only using the
Arduino Uno, there has been an attempt to automate the irrigation system [1]. An attempt to
control the irrigation system by using Arduino-based phones has been proposed, the
application uses the GPRS (General Packet Radio Service) [3]. A system which focuses on
drip irrigation, and controls the pipes outlet with the use of Raspberry Pi and Arduino Uno
has been proposed [5]. In Malaysia, smart home garden irrigation system has been
implemented by using Raspberry Pi, but it cannot be applied for agriculture fields and large
tea gardens [6]. Another system has also been proposed for home garden irrigation uses the
Zig-Bee model for wireless communication, it sends across messages to the concerned parties
regarding the soil moisture level from time to time [7]. There have been efforts also made to
help in the identification of plant diseases. An FCM based clustering model to identify
potential diseases in rice crop has been made which can monitor plant growth [2]. Exclusively
for pomegranate’s leaf disease detection, a system using support vector machine has been
developed [4]. Some systems have attempted to simplify the process by using a voice
controlled irrigation system [8]. The only drawback with this is that it is not completely
automated; command has to be given for the system to work. Some papers have focused on
specific irrigation methods such as drip irrigation [9].
This paper also emphasises the use of a wireless network. Agriculture can be classified
into few major phases such as germinating seeds, irrigating etc. Some systems have
automated more than one process providing a multi-purpose system [10]. This paper has
completely focused on irrigation of sugarcane crop and fuzzy logic has been used for decision
making [11]. The same method though cannot be applied for other plants. Similarly, another
paper focuses only on the pest prediction and control of disease in apple trees [12]. Smartness
in the irrigation system together with smartness in energy system is designed [13]. A wireless
sensor network is created to efficiently monitor and identify the status of water for irrigation
and the water level of the crops [14]. A wireless sensor networks with simple processors has
been designed for smart irrigation purpose. A smart environment has been created to
implement the smartness in the agriculture systems [15].
A closed loop system for irrigation is developed to monitor the cotton plant. This system
identifies the water level in the cotton crop irrigation system [16]. In summer the chestnut
trees growth has been affected by the reduction in the soil water content. Using drip irrigation
and micro sprinkling method the chestnut trees are protected [17]. Each leaf have different
pattern and identifying individual leave is a challenging task. In this paper segmentation
algorithm has been applied to identify the leaves [18]. In pomegranate leaf phytoplasma
disease has been detected using phylogenetic analysis. The disease in the earlier stage of
pomegranate leaf is detected by molecular characterization [19]. For oleander leaves a

Arduino Based Smart Irrigation System and Plant Leaf Disease Detection Using MATLAB
Backpropagation Neural Network has been applied to help the cultivator to find the leaves. It
is an efficient tool to differentiate the genotypes [20].
This paper is an attempt which can successfully address the issues of cost and feasibility
of implementation of smart irrigation system for large fields, along with this system a model
which can aid the farmers in identifying the diseases by the analysis of the leaves using K-
means algorithm has been developed, the system can be trained for numerous types of crops
and disease.
3. SYSTEM DESCRIPTION
The flow chart of the smart irrigation setup of this system is show in Fig.1. This shows the
way in which order the model will work.
Figure 1 Shows the block diagram of the smart irrigation system
The soil moisture sensor is used to determine the amount of humidity content in the soil.
The analog value which has to be maintained can be set manually by trial and error method,
which takes into consideration the type of crop and other environmental conditions. If the soil
is wet, it will send a low input voltage and if the soil is dry, it will send high input voltage.
This analog value is converted to digital for further analysis, by the Arduino Uno (micro-
controller). By using a HC-05 blue tooth module this data is transferred to another micro-
controller which in turn is located at the rear end, near the water tank. It controls the operation
of the water pump. So, water from the tank is pumped into the farm and as soon as the
required water level is reached, the soil moisture senses the humidity level of soil and then
this information is relayed through the same process, this again eventually stops the water
pump. From time to time, the soil moisture level and the status of the motor is sent to the
farmer or the concerned parties, with the help of the SIM-800, the global system for mobile
communication (GSM) module to their mobile phones as a SMS.
For the next phase of the project, to identify the diseases associated with the plants. In this
paper, there is specific focus and analysis on the leaves, as most of the major diseases
occurring in the plant show symptoms on the leaves. Data of the leaves has been collected
which have been affected by various illnesses to train the system. After image acquisition, the
images have been enhanced by increasing contrast, using the partial stretching enhancement.
As shown in the flow chart, Fig.2:

Figure 2 Shows the block diagram of the Image Processing technique
In the project, 13 parameters are used for the feature extraction of the image, such as: -
Contrast, Correlation, Energy, Homogeneity, Mean, Standard deviation, Entropy, RMS (Root
mean square), Variance, Smoothness, Kurtosis, Skewness, IDM (Inverse difference
movement).
The above system can be trained for various crops depending on the requirements. This
model can help to predict the diseases quite early based on the symptoms appearing on leave
to reduce the damage caused.
4. SYSTEM REQUIREMENTS
To make this system many software and hardware components are essential. Some of them
are-
4.1. Soil Moisture Sensor
The soil moisture sensor used here is YL-69 (Fig.3), the advantage it offers is high sensitivity
and low power consumption. It works on the principle of conductivity or resistivity of the
soil, the current passing through the two probes helps to determine the moisture content of
soil. The indicator factor is that if there is high resistance, then the water content in soil is less,
and vice-versa.
Figure 3 Soil moisture sensor (YL-69)
The size of this sensor is 60*20*5mm, depth of detection is 37mm; the temperature under
which it can operate is around 10 to 30 degrees Celsius and the operating voltage ranges from
3.3-5 Volts.
4.2. GSM Module
GSM Module is basically a GSM (Global System for Mobile) Modem which is connected to a
PCB board. The board also has pins to attach mic and speaker, to take out +5V or other values
of power and ground connections. It is used in the project for wireless data transmission for
alerting and messaging purposes. This module is compatible with Arduino UNO and can be
interfaced quite easily. The main purpose of this module is to relay the information to the
mobile from control unit and vice-versa. The SIM-800 is a complete Quad-Band GSM/GPRS
solution in a SMT type; it can be embedded for customer applications. It can transmit data,

voice or SMS as and when required with very low power consumption. The module can
support Quad-band 850/900/1800/1900MHz. It is controlled by the AT (AT stands for
attention) commands. The supply voltage range is around 3.4 to 4.4 voltage and the device
can operate from -40 to 85 degrees Celsius.
4.3. Arduino Uno
This is an easy to use open-source electronics platform based on hardware and software. The
Arduino boards are able to read inputs such as, a finger on a button, light on a sensor and
other similar activities and can turn it into an output that is it can activate a motor, turn on a
LED, send across information etc. This micro-controller board is based on the ATmega328P
(datasheet). It has 14 digital input/output pins (of which 6 can be used as PWM outputs), a 16
MHz quartz crystal,6 analog inputs, a USB connection, a power jack, a reset button and an
ICSP header. It can be connected to a computer with a USB cable or it can be powered with
an AC-to-DC adapter or a battery to start it.
An open-source known as Arduino Software (IDE) is used to create code and transfer it to
the board. It can run on Windows, Mac OS X, and Linux. The Arduino language is a set of
C/C++ function that can be called from the code.
4.4. HC-05 Module
This is an easy to use Bluetooth SPP (Serial Port Protocol) module. It is designed for
transparent wireless serial connection setup. This Bluetooth module can be used in a master or
slave configuration, which makes it a feasible solution for wire-less communication. The
default factory setting is Slave for this specific module. The role of the module can be
configured only by the AT commands only. Only the master module can initiate a connection
to the other devices, whereas the slave cannot.
Figure 4 Master- Slave HC-05 Module
This Module comes with an integrated antenna, an edge connector and programmable
(input and output) control. The IEEE specific code for the Bluetooth is IEEE802.15, it has 6
pins, and the default pin-code for auto-pairing is “1234”. By default, it gets connected
automatically to the last device in power. The slave default baud rate is 9600, data bits are 8
and there is no parity.
4.5. MATLAB Software
MATLAB (matrix laboratory) is a high performing language for technical computing. A
proprietary programming language developed by MathWorks. The MATLAB allows matrix
manipulations, plotting of functions and graphs, also for data extrapolation, image processing,
implementation of algorithms, creation of user interfaces, and interfacing with programs

written in other languages. The program is usually written in C/C++ or Java, the operating
system can be Windows, Linux and macOS. MATLAB has a group of application-specific
arrangements called toolboxes. Important to most clients of MATLAB, toolboxes enable the
client to learn and apply specific technology. Toolboxes are complete accumulations of
MATLAB functions (M-records) that stretch out the MATLAB condition to tackle specific
classes of issues. Regions in which the toolboxes are accessible include control systems,
fuzzy logic, signal processing, wavelets, simulation, neural networks and many others.
5. RESULT AND ANALYSIS
The first step is to connect the soil moisture sensor with the Arduino board, in which the code
is dumped. By, trial and error method the analog value which has to be maintained can be set;
this value will be based on crop requirements and other environmental factors which are taken
into consideration. The code in the Arduino Software (IDE) is written in C language. The
output displayed on the screen, is as shown in Table-1: -
Table 1 Basic output of the system
S.NO Soil Condition Sensor Output GSM Output Motor State
1 Dry High Low soil Moisture
detected Motor
turned ON
On
2 Wet Low Soil Moisture is
Normal Motor
turned OFF
Off
This information has to be transmitted again to the farmer or the concerned parties, so that
they are always aware above the condition of the soil and the status of the pump. So, that their
activities related to the farm can be planned accordingly, to achieve this objective. The GSM
module has been interfaced with the Arduino board, so that data is sent to the mobile in the
SMS format. This was done with the help of the AT commands. The message being sent to
the mobile of farmer can be in any language and can be changed according to our needs. The
SMS sent to the mobile number signified in the code is as show the Fig.5: -
Figure 5 Screenshot of the SMS
The devised master-slave system using the HC-05 module enables wireless data transfer
between the two ends. The Arduino board near the soil moisture sensor end detects the
humidity content of the soil, and then triggers the master module, which in turn relays the
message to the other pair of HC-05 module located at the other end. This information is then
processed by the Arduino board present at the other end, near the tank. Again, based on the

condition of soil (the input data received) the pump is turned on or switched off by the micro-
controller. The system worked well without any problems.
For the disease detection, the MATLAB software with SVM (Support Vector Machine)
has been used. The system has been trained for plant diseases related to grape and strawberry.
During training, for feature extraction we considered 13 parameters as shown in the Table-2: -
Table-2 Table comprising of the parameters considered for disease detection system
contrast Correlation energy homogeneity Mean SD entropy rms variance smooth kurtosis skewness IDM
0.078876 0.978321 0.762589 0.974878 14.84385 47.81168 1.709878 5.574773 2150.696 1 15.59777 3.632011 255
0.466835 0.865708 0.796721 0.959196 14.15012 48.13958 1.365848 4.313618 1632.216 1 15.7654 3.674427 255
0.367584 0.910197 0.757318 0.962547 16.44411 51.41943 1.667891 5.340374 2305.041 1 13.79264 3.402522 255
0.541238 0.751034 0.538239 0.922202 17.97166 37.66352 2.582884 7.403698 1306.813 1 10.4951 2.588339 255
0.512776 0.710321 0.894702 0.971681 17.1185 35.52045 2.843172 10.45046 1162.225 1 27.60328 4.682011 255
0.697626 0.873892 0.487259 0.910412 31.56037 56.45961 2.982981 8.114045 2844.325 1 4.400836 1.612926 255
0.488618 0.958014 0.268706 0.940314 71.85277 83.07288 5.120415 11.46161 5682.676 1 1.826999 0.649677 255
0.430913 0.896565 0.765986 0.965598 17.43762 52.46393 1.878881 5.728899 2052.447 1 12.8361 3.273623 255
0.576072 0.909153 0.710405 0.958389 23.81361 60.20883 1.673428 5.436237 3230.313 1 6.958173 2.334859 255
0.746201 0.90982 0.52789 0.900746 40.04733 73.85749 2.911928 7.533026 4465.177 1 3.993152 1.587334 255
0.88943 0.826304 0.818493 0.96507 16.41812 55.65341 1.300243 4.322796 2841.536 1 12.32038 3.301014 255
0.413971 0.970172 0.410585 0.972992 76.61838 97.98215 3.843924 9.364152 6332.289 1 1.517145 0.599022 255
0.08629 0.94908 0.884848 0.993442 8.575495 35.45269 0.717587 2.515133 1110.396 1 18.75243 4.105919 255
1.045052 0.816676 0.619172 0.920878 26.94916 60.87399 2.481799 6.955969 3471.252 1 6.628791 2.20753 255
The image data set which has been used to train the system has been obtained from plant
village social media site. Then random pictures of leaves affected by diseases and that of
healthy ones forms the training set for this model. The images are segmented using the K-
means clustering algorithm. This algorithm works to minimize the objective function, here the
objective function is:-
J=∑ ∑ i
(j)
-cj
2
Where the function is a chosen distance measure between a given data point and the
cluster centre, this an indicator of the distance of the n data points from their respective cluster
centres. The algorithm helps to separate the image into groups or clusters, for further analysis.
In this case, a set of three different clusters are obtained, from which it is required to choose
the ROI (region of interest), to identify the type of disease as shown in Fig.6: -
Figure 6 Screenshot of the output screen

On selecting the cluster region which shows the diseased parts, the type of disease, the
percentage of affected region of the leaf and the accuracy of system in identifying the disease
is obtained. This system is able to produce up to 95 percent accuracy of linear kernel, as
shown in the Fig.7:
Figure 7 Screenshot of the final output provided by diseases detection system
The following table shows the output for few of the leaves (healthy and unhealthy) which
have been tested by using the disease detection algorithm. The classification results, along
with the other parameters have been compiled and are as follows for the leaves shown in
Table-3:
Table 3 Classification results for various leaves
Figure Classification Result Accuracy of
Kernel
Affected
Region
Grape healthy 95.1613 NA
Grape Leaf blight
(Isariopsis Leaf Spot)
96.7742 42.3764

Strawberry healthy 96.7742 NA
Strawberry Leaf scorch 96.7742 53.8165
The graph below shows the relation between mean square error (mse) and epoch. It is the
performance characteristic graph of the system (Fig.8): -
Figure 8 Screenshot of the performance graph of leaf disease detection system
6. CONCLUSION
In this paper, the process of creating a feasible, low cost smart irrigation system which can
lower power, labour and water consumption has been discussed. By the use of this system, the
frequent visits to the farm will come down greatly, enabling the farmer to focus on other
activities. This system will also, update the farmer from time-to-time about the status of the
pump and the water content level of soil, by sending a message to the mobile with the help of
GSM Module. This model will also help to identify the type of diseases occurring in the plant,
by using the images of the leaves. The system will be able to determine the specific type of
disease based on the symptoms shown in the leaves. This system has been made in mind
keeping the scattered and less land holdings possessed by Indian farmers, so the emphasis is
more on cost effectiveness and simplicity. If implemented this system can also increase the
fertility level of the soil, and will aid in maintaining the optimal growth of plant. Even
problems occurring due to bad irrigation systems, such as run off of the fertilizers and top soil

will come down along with that other problems such as water clogging and alkaline water
beds will be reduced.
REFERENCES
[1] Devika C. M., Bose K., Vijayalekshmy S. Automatic plant irrigation system using
Arduino. IEEE International Conference on In Circuits and Systems (ICCS), 2017, 1,
384-387.
[2] Devi R., Hemalatha R., Radha S., (2017, February). Efficient decision support system
for Agricultural application, Third International Conference on Advances in Electrical,
Electronics, Information, Communication and Bio-Informatics (AEEICB), 2017,1, 379-
381.
[3] Pavithra D.M.S, .Srinath, GSM based Automatic Irrigation Control System “for Efficient
use of Resources and crop Planning by using an Android Mobile, IOSR Journal of
Mechanical and Civil Engineering, 2014, 11(4), 49-55.
[4] Panchal S. S., Sonar R., Pomegranate Leaf Disease Detection Using Support Vector
Machine, International Journal of Engineering and Computer Science, 2016, 5(6), 10-
15.
[5] Agrawal N., Singhal S., Smart drip irrigation system using raspberry pi and Arduino,
International Conference on Computing, Communication & Automation, Noida, 2015, 1,
928-932.
[6] Ishak, S. N., Malik, N. A., Latiff, N. A., Ghazali, N. E., Baharudin, M. A. Smart home
garden irrigation system using Raspberry Pi. IEEE 13th International Conference on
Communications (MICC), Malaysia , 2017, 13, 101-106.
[7] Al-Ali A.R., Qasaimeh M., Al-Mardini M., Radder S., Zualkernan I.A., ZigBee-based
irrigation system for home gardens, International Conference on Communications, Signal
Processing, and their Applications (ICCSPA), Sharjah, 2015, 1, 1-5.
[8] Katrojwar S. B., Taksande V. K., Voice controller application for irrigation. 2nd
International Conference on Communication and Electronics Systems (ICCES), 2017, 40-
41.
[9] Mohanraj I., Gokul V., Ezhilarasie R., Umamakeswari A., Intelligent drip irrigation
and fertigation using wireless sensor networks, Technological Innovations in ICT for
Agriculture and Rural Development (TIAR), 2017, 2, 36-41.
[10] Arvind G., Athira V. G., Haripriya H., Rani R. A., Aravind S., Automated irrigation
with advanced seed germination and pest control, Technological Innovations in ICT for
Agriculture and Rural Development (TIAR), 2017, 1, 64-67.
[11] Alfin A. A., & Sarno R., Soil irrigation fuzzy estimation approach based on decision
making in sugarcane industry, 11th International Conference on Information &
Communication Technology and System (ICTS), 2017, 137-142.
[12] Shivling, D. V., Sharma, S. K., Ghanshyam, C., Dogra, S., Mokheria, P., Kaur, R., &
Arora, D. (2015, October). Low cost sensor based embedded system for plant protection
and pest control. International Conference on Soft Computing Techniques
Implementations (ICSCTI), 2015, 179-184.
[13] Antonia Nasiakou, Manolis Vavalis, Dimitris Zimeris, Smart energy for smart irrigation,
Computers and Electronics in Agriculture, 2016, 129(1), 74-83.
[14] Francesco Fabiano Montesano, Marc W. Van Iersel, Francesca Boari, Vito Cantore,
Angelo Parente, Sensor-based irrigation management of soilless basil using a new smart
irrigation system: Effects of set-point on plant physiological responses and crop
performance, Agricultural Water Management, 2018, 203, 20-29.

[15] Murat Dener, Cevat Bostancioğlu, Smart Technologies with Wireless Sensor Networks,
Procedia - Social and Behavioral Sciences, 2015, 195, 1915-1921.
[16] G. Vellidis, M. Tucker, C. Perry, C. Kvien, C. Bednarz, A real-time wireless smart sensor
array for scheduling irrigation, Computers and Electronics in Agriculture, 2008, 61(1),
44-50.
[17] M. Mota, T. Pinto, F. Raimundo, A. Borges, J. Caco, J. Gomes-Laranjo, Relating plant
and soil water content to encourage smart watering in chestnut trees, Agricultural Water
Management, 2018, 203, 30-36.
[18] Liankuan Zhang, Paul Weckler, Ning Wang, Deqin Xiao, Xirong Chai, Individual leaf
identification from horticultural crop images based on the leaf skeleton, Computers and
Electronics in Agriculture, 2016, 127, 184-196.
[19] Mohammad Salehi, Seyyed Alireza Esmailzadeh Hosseini, Rasoul Rasoulpour, Elham
Salehi, Assunta Bertaccini, Identification of a phytoplasma associated with pomegranate
little leaf disease in Iran, Crop Protection, 2016, 87, 50-54.
[20] Ada Baldi, Camilla Pandolfi, Stefano Mancuso, Anna Lenzi, A leaf-based back
propagation neural network for oleander (Nerium oleander L.) cultivar identification,
Computers and Electronics in Agriculture, 2017, 142, 515-520.

IMPLEMENTATION OF MARRIAGE BY INDIGENOUS LAW TO YEI TRIBE COMMUNITIES

More Related Content

What's hot (18)

Similar to IMPLEMENTATION OF MARRIAGE BY INDIGENOUS LAW TO YEI TRIBE COMMUNITIES (20)

More from IAEME Publication (20)

Recently uploaded (20)

IMPLEMENTATION OF MARRIAGE BY INDIGENOUS LAW TO YEI TRIBE COMMUNITIES