ABSTRACT

In the growth of agricultural countries like India, agriculture plays a crucial role. The growth of
the country has been hampered by problems surrounding agriculture. By modernizing the new
conventional methods of agriculture, the only solution to this issue is smart agriculture. IoT
sensors provide knowledge about the nature of agricultural fields and then, based on the input
farmer can take action. In our project we propose an Iot based advanced solution for monitoring
the soil condition/state and also atmosphere for efficient crop growth. The developed system is
capable of monitoring temperature, humidity, soil moisture level sensors connected NodeMCU
and using image processing we provide the detection of disease of the plant by capturing the
image of the plant leaf. Also, notification is sent to farmer’s phone using Wi-Fi about
environmental and Field Condition
INTRODUCTION

Agriculture is India's main profession and is the backbone of the Indian financial sector.
In addition to supplying food, agriculture gives rural people job opportunities on a wide scale in
underdeveloped and developing countries. It is the method by which the production and raising
of domestic animals provides fruit, fiber and many other desired items. Agriculture is the
primary source of livelihood for about 58 percent of the population of India. Smart cultivating is
the execution of different advances and gadgets like web, cloud and the IoT gadgets. As in this
day and age the populace is expanding and it should be around 9.7 billion by 2050 and to take
care of those billion people groups we have to improve the creation of harvests. The populace is
expanding and then again the agrarian land is diminishing because of different reasons like
industrialization, business markets and private structures are being made on those horticultural
terrains and to take care of these billions we have to increment out creation and this should be
possible by actualizing IoT in cultivating.

Climate changes will have significant impact on agriculture by increasing water demand
and limiting crop productivity in areas where irrigation is most needed. Irrigation system, rain
fed agriculture, groundwater irrigation is some of the methods introduced to produce healthier
crops which may not use water efficiently. In order to use water efficiently a smart system is
designed. In the system farmer need not make the water flow into fields manually, but the system
automatically does that efficiently. The technological advancements began to increase the
efficiency of production remarkably thus, making it a reliable system. The knowledge of
properties of soil determines the water supply to be driven in a smart way.

The practice of agriculture in a smart way helps to acquire knowledge of soil and
temperature conditions. Developing the smart agriculture using IoT based systems not only
increases the production but also avoids wastage of water. The soil moisture sensor, humidity
and temperature sensor continuously monitors the soil and environmental conditions, sends the
live data to smartphone via cloud service.

While raining, the moisture content may increase several times. A rain-drop detecting
sensor intimates the controller if there is rainfall, making the water supply to reduce or stop
depending upon the moisture content at the moment. The crop requirements such as amount of
humidity, temperature and moisture content are to be studied and can be installed again in the
controller to meet its circumstances.

Crop production problems are common in India which severely effect rural farmers, agriculture
sector and the country’s economy as a whole. In Crops leaf plays an important role as it gives
information about the quantity and quality of agriculture yield in advance depending upon the
condition of leaf. we proposed the system which works on preprocessing, feature extraction of
leaf images from plant village dataset followed by convolution neural network for classification
of disease and recommending Pesticides using Tensor flow technology. Agriculture is one field
which has a high impact on life and economic status of human beings. Improper management
leads to loss in agricultural products.
LITERATURE SURVEY

[1] Internet of Things (IoT) for Smart Precision Agriculture and Farming in Rural Area

The authors main aim of the project is to develop smart precision agriculture and farming with
the help of fog computing and the Wi-Fi based long distance connecting networks in IoT. The author
proposed scalable network architecture for monitoring and controlling agriculture in rural areas. The
developed project is costlier for the developing countries for the use of the system for smart farming [1].

Fig: IoT network architecture connecting rural areas

To focus on the specific requirements, author propose a scalable network architecture for
monitoring and controlling agriculture and farms in rural areas. Compared to the existing IoT-based
agriculture and farming solutions, the proposed solution reduces network latency up to a certain extent.
In this, a cross-layer-based channel access and routing solution for sensing and actuating is proposed.[1]
[2] IoT Based Smart Irrigation Monitoring and Controlling System.

The author developed a remote monitoring and controlling of drips through wireless sensor
network for smart irrigation system for the farm field. Smart farm irrigation system uses android phone
for controlling of the system for management of the farm irrigation water. Distribution of water supply
as per requirement for better use of the water supply for the farm crops. System is based on IoT that
uses real time input data to control the water supply in the farm Field [2].

Fig: System Overview Based Smart Irrigation Monitoring and Controlling System

Smart farm irrigation system uses android phone for remote monitoring and controlling of drips through
wireless sensor network. ZigBee is used for communication between sensor nodes and base station.
Real time sensed data handling and demonstration on the server is accomplished using web based java
graphical user interface. Wireless monitoring of field irrigation system reduces human intervention and
allows remote monitoring and controlling on android phone. Cloud Computing is an attractive solution
to the large amount of data generated by the wireless sensor network. [2]

[3] Smart Farming Using IoT.

The developed a remote controlled vehicle for operations like spraying, cutting, etc. for
surveying and controlling of the devices smart farming devices. The Monitoring of the temperature,
humidity, soil condition did with the remote controlled vehicle. Distribution of water supply as per
requirement management is also getting done with remote controlled vehicle [3].

To increase the productivity, a novel design approach is presented in this paper. Smart farming
with the help of Internet of Things (IOT) has been designed. A remote controlled vehicle operates on
both automatic and manual modes, for various agriculture operations like spraying, cutting, weeding
etc. The controller keeps monitoring the temperature, humidity, soil condition and accordingly supplies
water to the field [3]

[4] Internet of Things Application for Implementation of Smart Agriculture System

The author tried to simultaneously generating more output from the same amount of input, but
there is need of improvement. The author controlled the operations like spraying, cutting, etc. with IOT
based smart device and do monitoring of farm field. The author done monitoring of farm field like
temperature, humidity, soil conditions [4].

This proposed wireless robot is equipped with various sensors for measuring different
environmental parameters. It also includes Raspberry Pi 2 model B hardware for executing the whole
process. The main features of this novel intelligent wireless robot is that it can execute tasks such as
moisture sensing, scaring birds and animals, spraying pesticides, moving forward or backward and
switching ON/OFF electric motor. The robot is fitted with a wireless camera to monitor the activities in
real time. The proposed wireless mobile robot has been tested in the fields, readings have been
monitored and satisfactory results have been observed, which indicate that this system is very much
useful for smart agricultural systems. [4]

[5] Design and Development of Precision Agriculture System Using Wireless Sensor Network

The author introduces a scalable solution for the smart farming. He introduced a smart irrigation
system for the farming to handle the irrigation in the farm [5]. Author introduces sensor network
architecture for the farming. He introduces the system as the main problem in the farming is bad
irrigation system to overcome the problem he introduces smart irrigation system and architecture [5].
He also manages the system with controlling of the water level in side water tank to distribute water in
the farm land efficiently [5].

Author proposed a suitable location for monitoring of crops. The sensing system is based on a
feedback e of temperature and moisture sensor control mechanism with a centralized control unit
which regulates the flow of water on to the field in the real time based on the instantaneous
temperature and moisture values. The sensor data would be collected in a central processing unit which
would take further action. Thus by providing right amount of water we would increase the efficiency of
the farm. The farmer can also look at the sensory data and decide course of action himself. We have
made the interface of our project keeping in view the educational and financial background of average
Indian farmer. [5]

[6] A control system in an intelligent farming by using Arduino technology.

Author introduces a smart farming system in which the system uses a GPS system for working
for the weather data for the farm Location. He introduces the smart farming architecture in the system
also collects the weather forecast data and predict the next possible condition for the farmland. He first
acquires the data from sensor and the GPS and predicts the next possible outcomes then process the
data give instruction to system, but the system is costlier to use the system [6].

Fig: A system architecture of the system of intelligent farming

[7] Smart Farming IoT in Agriculture.

The author developed smart farming system with help of Internet of things in which it uses a
server which receives the data from the farm field. The server takes the action on the farm field such as
watering the plants if water moisture is less in soil. He created a smart farming architecture for the poly
house farming in which if the air moisture is less then water get distribute to crops. If heat is more than
air, then condition is get on according to the system works.

The IoT sensors used in this model are air temperature sensor, soil pH sensor, soil moisture
sensor, humidity sensor, water volume sensor etc. In this paper they surveyed typical agriculture
methods used by farmers these days and what are the problems they face, they visited poly houses for
further more information about new technologies in farming. The proposed model is a simple
architecture of IoT sensors that collect information and send it over the Wi-Fi network to the server,
there server can take actions depending on the information [7]

[8] AgriSys: A smart and ubiquitous controlled environment Agriculture system.

The author developed a smart farming system they called it as AgriSys. AgriSys is a smart and ubiquitous
controlled-environment agriculture system. Author aims at providing the benefits through the AgriSys
smart farming system are increased the productivity of crops, enhanced safety of farm field, easier
agriculture procedures for farming, instant interventions around the clock, advanced lifestyle. The
author also said AgriSys is found to be superior in its edibility, ease of maintenance, web interface, and
portability

In this paper, they have proposed a smart Agriculture System (AgriSys) that can analyze an agriculture
environment and intervene to maintain its adequacy. The system deals with general agriculture
challenges, such as, temperature, humidity, pH, and nutrient support. In addition, the system deals with
desert-specific challenges, such as, dust, infertile sandy soil, constant wind, very low humidity, and the
extreme variations in diurnal and seasonal temperatures. The system interventions are mainly intended
to maintain the adequacy of the agriculture environment. For a reduced controller complexity, the
adoption of fuzzy control is considered. The system implementation relies on state-of-art computer
interfacing tools from National Instruments as programmed under LabVIEW [8]

[9] Plant Disease Detection Using Image Processing

Identification of the plant diseases is the key to preventing the losses in the yield and quantity of
the agricultural product. The studies of the plant diseases mean the studies of visually observable
patterns seen on the plant. Health monitoring and disease detection on plant is very critical for
sustainable agriculture. It is very difficult to monitor the plant diseases manually. It requires
tremendous amount of work, expertize in the plant diseases, and also require the excessive
processing time. Hence, image processing is used for the detection of plant diseases. Disease
detection involves the steps like image acquisition, image pre-processing, image segmentation,
feature extraction and classification. This paper discussed the methods used for the detection of
plant diseases using their leaves images. This paper also discussed some segmentation and
feature extraction algorithm used in the plant disease detection. [9]
[10] A Review on Leaf Disease Detection using Image Processing

Identification of the leaf diseases is the key to preventing the losses in the yield and quantity of
the agricultural product. The studies of the leaf diseases mean the studies of visually observable
patterns seen on the leaf. Health monitoring and disease detection on leaf is very critical for
sustainable agriculture. It is very difficult to monitor the leaf diseases manually. It requires
tremendous amount of work, expertize in the leaf diseases, and also require the excessive
processing time. Hence, image processing is used for the detection of leaf diseases. Disease
detection involves the steps like image acquisition, image pre-processing, image segmentation,
feature extraction and classification. This paper discussed the methods used for the detection of
leaf diseases using their leaves images. [10]

BACKGROUND

According to years’ census around 70% of people are from villages in India and they depend
upon agriculture. Agriculture in Indian villages has been the foremost profession for the people
since the ancient period, as they earn their livelihoods from agriculture. Many farmers reduced
cultivating due to migration of people from rural to urban, lack of interest, scarcity of agriculture
fields and water. One of the major challenge in agriculture is the proper monitoring of Soil
health, the air temperature and humidity, proper irrigation at right time.

EXSISTING SYSTEM

In the present era one of the greatest problem faced by the world is water scarcity and
agriculture being a demanding occupation consumes plenty of water. Therefore, a system is required
that uses water judiciously. Smart irrigation systems estimate and measure diminution of existing plant
moisture in order to operate an irrigation system, restoring water as needed while minimizing excess
water use.

PROPOSED SYSTEM
 The proposed device uses a microcontroller (NodeMCU) that has a Wi-Fi module over it
(ESP8266). Blynk mobile App is used as the user interface. Soil moisture sensor, humidity and
temperature sensor (DHT11), rain gauge sensor, LED’s along with relay are used. This relay is
connected to a water pump which pumps water to the crops when the relay is ON. The soil
moisture sensor senses the moisture level of the soil, depending on the level of moisture,
NodeMCU decides whether to water the crop or not. By using appropriate functions and
conditional statements in the code written for the NodeMCU functioning, the watering of the
crop starts by NodeMCU making relay ON when the moisture content is below a threshold value
and is made OFF when there is enough moisture content in the soil.

The humidity and temperature sensor gives the humidity and temperature values of the
atmosphere which determine whether the crop is suitable for growth. Some crops grow only in
particular weather conditions and some give better yield only for a particular temperature range.
The raindrop sensor measures the intensity of rain. If there is enough rainfall to provide soil with
required water, the crops are not watered. Even after raining, if the crops are not having
sufficient water then water is pumped again by making relay ON. Here we also check the ph for
the water as the water contains some acidic content in it or no, so that plant cannot be affected.
Data reaches the blynk cloud from NodeMCU through Wi-Fi. From this app, the farmer can
control the relay through various buttons and switches. When the NodeMCU gets the command
from the app then the appropriate analysis is done and the pump motor is controlled. The data
again travels through Wi-Fi again in the same path.
Block diagram

Power Supply

Soil Moisture
Sensor Relay

Rain gauge

sensor Pump
NODEMCU

DHT11
Blynk

LED’s

Mobile App
The main aim is to design a system which is efficient and which provide disease name and pesticides
name as fast as possible.

For that purpose, we use two phase:

1st is training phase and 2nd is testing phase.

In 1st phase: Image acquisition, Image Pre-processing and CNN based training.

In 2nd phase Image acquisition, Image Pre-processing, Classification and disease identification and
pesticides identification. For experimentation purpose we have used Plant Village datasets. The data
records contain 54,309 images. The images span 14 crop species: Apple, Blueberry, Cherry, Corn, Grape,
Orange, Peach, Bell Pepper, Potato, Raspberry, Soybean, Squash, Strawberry, Tomato. It contains
images of 17 fungal diseases, 4 bacterial diseases, 2 mold(oomycete) diseases, 2 viral disease, and 1
disease caused by a mite. 12 crop species also have images of healthy leaves that are not visibly affected
by a disease. The following figures show the overall model of the system and sample images from
dataset.
IMAGE ACQUISITION

For training, Image is taken from database. And for testing, you can take image from camera at real
time but in this project, we made a particular folder on desktop from that image will be fetched by GUI
screen and send through java web services i.e. tomcat server to server side system on which pre-
processing is done and later on algorithm test that particular image.

Image Pre-Processing

Image should be processed before sending to the algorithm for testing and training purpose. For that
purpose, in this project image is scaled or resize into 150 x 150 dimension. As we used color image so
that we don’t need any color conversion techniques and that pre-processed image is directly passed to
algorithm for training and testing purpose.

Convolutional Neural Network

Once pre-processing is done, then CNN is used for training purpose and after that we get trained model.
That CNN method is written with help of tensor flow. By using this model, we classify the image that
system is getting after pre-processing of testing image. Then we get particular disease name or healthy
leaf name if there is no disease on that leaf and that disease name is send to GUI screen and with the
help of that disease name we get particular pesticide name which help farmer to take respective action
in order to decrease percentage of disease.

SYSTEM REQUIREMENT

Software Requirement

 Arduino IDE
 C++
 Blynk
 Weather API

Hardware Requirement

 Nodemcu
 DHT11
 Servo motor
 buzzer
 Soil Moisture
 Rain gauge
 Relay
 Pump
 12v Adaptor
 Power Supply

CONCLUSION
Using NodeMCU, IoT technology is used to sense and analyse temperature, humidity level, soil
moisture level, rain condition and pump motor. All these values are transmitted using Wi-Fi to
the smart phone. Appropriate water is pumped and rain is also used effectively because of the
use of this system. This system is very useful for farmers because they need to pump water
regularly and check the status of each crop. Farmers can know the values of humidity,
temperature and soil moisture from anywhere in the world and if the pump motor is ON on their
smartphones via the blynk app.

REFERENCES

[1] Nurzaman Ahmed, Debashis De , Senior Member, IEEE, and Md. Iftekhar Hussain, Member, IEEE;
“Internet of Things (IoT) for Smart Precision Agriculture and Farming in Rural Areas” IEEE INTERNET OF
THINGS JOURNAL, 2018.

[2] Shewta B. Saraf, Dhanashri H. Gawali, “IoT Based Smart Irrigation Monitoring And Controlling
System”, May 2017.

[3] Arshia Bhattacharjee;“Smart farming using IOT”2017 8th IEEE Annual Information Technology,
Electronics and Mobile Communication Conference (IEMCON).

[4] K. Lokesh Krishna; Omayo Silver ”Internet of Things application for implementation of smart
agriculture system”2017 International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud)
(I-SMAC).

[5] S. R. Nandurkar, V. R. Thool, R. C. Thool, “Design and Development of Precision Agriculture System
Using Wireless Sensor Network”, IEEE International Conference on Automation, Control, Energy and
Systems (ACES), 2014.

[6] N. Putjaika, S. Phusae, A. Chen-Im, P. Phunchongharn and K. Akkarajitsakul, "A control system in an
intelligent farming by using arduino technology," 2016 Fifth ICT International Student Project
Conference (ICT-ISPC), Nakhon Pathom, 2016, pp. 53-56.

[7] Rahul Dagar, Subhranil Som, Sunil Kumar Khatri, “Smart Farming – IoT in Agriculture”, IEEE 2018.
[8] A. Abdullah, S. A. Enazi and I. Damaj, "AgriSys: A smart and ubiquitous controlled-environment
agriculture system," 2016 3rd MEC International Conference on Big Data and Smart City (ICBDSC),
Muscat, 2016, pp. 1-6.