Professional Documents
Culture Documents
Submitted by:
Chethan Kumar S
18CSE015
At first, I would like to thank the ACU, B G Nagara, for having this seminar as
part of its curriculum, which gave me a wonderful opportunity to work on my research
and presentation abilities and BGSIT for providing me with such excellent facilities,
without which, this seminar could not have acquired the shape it has now done.
I would like to thank, Ms. Ancy Y Das, Asst. Prof., Dept. of CSE for his
continuous support, advice and guidance.
Last but not the least my profound thanks to the Teaching staff and Non-Teaching
staff of the Department of Information Science and Engineering for their help and
patience.
CHETHAN KUMAR S
18CSE015
i
ABSTRACT
As initiatives to increase Japan’s declining food self-sufficiency rate and
revitalize the field of agriculture, the concept of smart agriculture and urban agriculture
are currently being implemented. Automation via artificial intelligence is expected to
overcome the labor shortage inthe agricultural industry. However, the number of skilled
farmers who can contribute to gathering crop growth data required for machine
learning is restricted, and these data are also limited to local and environmental-based
conditions. In this paper, we propose a system for obtaining composite growth data in
various environments and crops targeted for home gardens and paddy fields. An
interactive cultivation sensing system consisting of IOT-based technologiesis designed
and realized to ensure the continuous growth of crops in optimum conditions daily.
With this, progress will be made in determining the efficient cultivation conditions for
machine learning, and in finding solutions to future problems of agriculture.
ii
TABLE OF CONTENTS
Title Page No.
ACKNOWLEDGEMENT i
ABSTRACT ii
TABLE OF CONTENTS iii
LIST OF FIGURES iv
CHAPTER 1: INTRODUCTION 1
3.1.1 Disadvantages 5
3.4 Applications 9
iii
LIST OF FIGURES
iv
Interactive cultivation system for the future IOT-based agriculture 2021-2022
CHAPTER 1
INTRODUCTION
To promote agriculture as a further growth industry, the Japanese government has
devised the ―Policy Package for Enhancing Competitiveness of Japan’s Agriculture‖ in
November 2016. This policy incorporates the direction of new agricultural administration
reforms which includes decreasing agriculture input costs, strategically exploring international
markets and reinforcing human resources. Despite the initiatives taken by the government,
modern agriculture in Japan still face limitations with the aging agricultural labor force and labor
shortages.
According to the latest data by the Ministry of Agriculture, Forestry and Fisheries
(MAFF), the number of agricultural workers in 1995 is 4.14 million, and it has then decreased by
half to 2.1 million by 2015 . The average area of cultivated land per business unit has steadily
expanded from 1.6 hectares in 1995 to 2.5 hectares in 2015. It is possible to manage plant
cultivations in their optimum environmental conditions for both smart agriculture and urban
agriculture, provided that there would be some constraints. The actual number of skilled farmers
who can contribute to gathering plant growth data required for machine learning in AI is limited.
These plant growth data are also restricted to local and environmental-based conditions.
In this project, targeting home gardens and paddy fields, we propose and design a system for
acquiring composite growth data in various environments and crops. An interactive cultivation
sensing system is designed to ensure the continuous growth of crops daily. With this, progress
will be made in determining the efficient cultivation conditions for machine learning, and in
finding solutions to future problems of agriculture.
In this project, targeting home gardens and paddy fields, we propose and design a system
for acquiring composite growth data in various environments and crops. An interactive
cultivation sensing system is designed to ensure the continuous growth of crops daily. With this,
progress will be made in determining the efficient cultivation conditions for machine learning,
and in finding solutions to future problems of agriculture. In this project, targeting home gardens
and paddy fields, we propose and design a system for acquiring composite growth data in various
environments and crops.
The system also provides basic remote function where users could control the watering
system, and the Light Emitting Diodes (LEDs) via a web-based application. The analyzed data
are then gathered and displayed in the form of graphs, charts or figures for the user’s
understanding. A smart home garden irrigation system is developed, where all sensor devices are
connected to Raspberry Pi to obtain the data in paper . These data are then monitored and
analyzes. The watering activity is done using Android application as a Graphical User Interface
(GUI). At certain critical condition such as low water level, users will be notified by email
application. This proposed system has shown several advantages such as cost reduction,
wastewater reduction, and minimizing the physical interface of human.
An interactive cultivation sensing system is designed to ensure the continuous growth of
crops daily. With this, progress will be made in determining the efficient cultivation conditions
for machine learning, and in finding solutions to future problems ofagriculture.
Currently, robots are developed with autonomous navigation and used in paddy fields for
weed growth suppression. A robot that utilizes a laser range finder for autonomous mobility, and
a robot arm for weed suppression is developed in paper. It travels along the rice rows without
damaging the rice plants in a paddy field. In paper, ―Paddy Weeding Robot‖ is proposed. It
functions by agitating the soil and creating an environment that makes weeds difficult to
photosynthesize, therefore inhibiting their growth. Nissan has also developed for cultivating
purpose.
CHAPTER 2
RELATED WORK
To the extent of our knowledge, there has not been a proposal for an interactive
cultivation sensing system via instant messaging applications. In this section, we explain in brief
some of the IoT automation systems which we used as a reference while creating and executing
our proposed system. In paper, a monitoring system for indoor vertical farming has been
designed and implemented. Sensors were used to gather data such as temperature and light
intensity and sentto Beagle Bone Black (BBB) microcontroller. Then, the data was processed
and uploaded to the Thing speak Cloud.
The system also provides basic remote function where users could control the watering
system, and the Light Emitting Diodes (LEDs) via a web-based application. The analyzed data
are then gathered and displayed in the form of graphs, charts or figures for the user’s
understanding. A smart home garden irrigation system is developed, where all sensor devices are
connected to Raspberry Pi to obtain the data in paper . These data are then monitored and
analyzes. The watering activity is done using Android application as a Graphical User Interface
(GUI). At certain critical condition such as low water level, users will be notified by email
application. This proposed system has shown several advantages such as cost reduction,
wastewater reduction, and minimizing the physical interface of human.
A monitoring system for indoor plants through an Android mobile application on the
integration of urban agriculture is developed in paper . The mobile application gives information
and updates such as light and water plant needs through the collection of data by microcontroller
unit sensors and other components. The analysis of the plant condition scan be determined and
known by the plant owners through a connection with the cloud.
The authors used ―Cayenne‖, third-party software, to communicate with the application
programming interface (API) of Arduino and relay any sensor data that the device would retrieve
from its sensors. Paper introduced a smartagriculture method where the efficient use of IoT for
traditional agriculture is demonstrated. It utilizes Arduino and ESP8266 Wi-Fi module to
monitor and control the smart irrigation system. This system consists of sensors such as pH
sensor, water flow sensor, temperature sensor andsoil moisture sensor collected by Arduino.
This cultivation kit is mounted with multiple sensors used to retrieves data which affects
the growth of crops. These data include water level, soil moisture, temperature, humidity, pH
level, and carbon dioxide level. Images of the crops are also obtained using a camera. Both data
and images would then be analyzed to determine the degree of growth. The cultivation kit is
equipped with a network function, and sensor data is collected by the server. Based on sensor
data collected in various home gardens and other outdoor environments, machine learning can be
performed as big data, together with image data. Thus, it is possible to associate the growth
status of the crops with sensor information and by the crop earlier data which is stored in the
cloud.
Besides that, we also proposed an automatic control system for farm irrigations with
Paddy Drone, the robot is able to sense weather conditions and growth conditions of rice plant
via sensors. Data such as water depth would be retrieved to automate the opening and closing of
water gates wirelessly. Data and images taken would be analyzed and conveyed to farmers in
natural language for better understanding to their interactive cultivation purpose.
The proposed system can also automatically control the sprinkler, fog pumps, and the
functional status (alternating the on and off switches occasionally) through an instant messaging
application, namely LINE Application. In our proposed system, we combine the IoT-based
technologies, namely Raspberry Pi microcontroller, a camera module, and various sensors, to set
up an interactive cultivation sensing system for home gardens and paddy fields. We plan to
integrate image processing techniques and instant messaging applications in our proposed
system.
We proposed and integrated IoT-based technologies to create an interactive cultivation
sensing system to assist farmers in growing their crops in optimum condition and also overcome
the problem of the labor shortage. Currently, we have created an environment in which data such
as temperature, humidity and camera images can be obtained and uploaded to a server for
Kaiwalingual. Also, the function to convey the current temperature interactively using Line is
operated. For Paddy Drone, we have gained camera images of the surrounding environment of a
paddy field wirelessly and can be viewed through a smartphone. In the future, we will realize the
device. By adding different types of sensors, storing the data collected in a server, and
performing machine learning, the conditions of the radish sprouts and rice plants can be
estimated more accurately. More complex speech patterns would be generated .
Japan has been relying heavily on food imports to support its people’s needs, and
currently is strongly urged to fully open its agricultural market and deregulate its agricultural
policies via negotiations with Trans-Pacific Partnership (TPP), Free Trade Agreement (FTA) and
Economic Partnership Agreement (EPA) . These actions could lead to the degradation of Japan’s
agricultural and rural economies, and hinder the future of Japanese food security.
Operations such as crop cultivations are often only managed by skilled workers as they
need long experience and instinct. Thus, a great deal of time and effort is required to guide new
farmers to master the know-how. To overcome these problems and to make agriculture a
progressive industry, MAFF has been gearing towards the idea of ―smart agriculture‖ . Smart
agriculture utilizes information communications technologies (ICTs) which incorporate Artificial
Intelligence (AI) and Internet of Things (IOT) to allow ultra-labor-saving and high-quality
production. Machinery manufacturers and ICT vendors are now working with farmers to actively
develop new technologies that can be used for smart agriculture.
The Internet of Things (IoT) is a network of devices for communicating machine to machine
(M2M) based on wired and wireless Internet. IoT in agriculture is a revolutionary technology that
can be applied to agricultural production year-round. The aim of this study is to summarize cases of
IoT being applied to agricultural automation in the agricultural sector and to discuss the limitations
and prospects for expanding the application of IoT technology in Korean nations from their citizens.
The average area of cultivated land per business unit has steadily expanded from 1.6 hectares
in 1995 to 2.5 hectares in 2015. It is possible to manage plant cultivations in their optimum
environmental conditions for both smart agriculture and urban agriculture, provided that there would
be some constraints. The actual number of skilled farmers who can contribute to gathering plant
growth data required for machine learning in AI is limited.
According to the latest data by the Ministry of Agriculture, Forestry and Fisheries (MAFF),
the number of agricultural workers in 1995 is 4.14 million, and it has then decreased by half to 2.1
million by 2015. The average area of cultivated land per business unit has steadily expanded from
1.6 hectares in 1995 to 2.5 hectares in 2015. The average age of the population engaged in
agriculture in 2016 was 66.4 years and more than 60% were above the age of 65. Operations such as
crop cultivations are often only managed by skilled workers as they need long experience and
instinct. Thus, a great deal of time and effort is required to guide new farmers to master the know-
how.
To overcome these problems and to make agriculture a progressive industry, MAFF has been
gearing towards the idea of ―smart agriculture‖. Smart agriculture utilizes information
communications technologies (ICTs) which incorporate Artificial Intelligence (AI) and Internet of
Things (IoT) to allow ultra-labor-saving and high-quality production. Machinery manufacturers and
ICT vendors are now working with farmers to actively develop new technologies that can be used
for smart agriculture. By incorporating ICTs, it is expected that soil productivity would be improved,
and operations such as seed planting, weeding and harvesting would be done efficiently. Farm works
could also be recorded via cloud systems.
A smart home garden irrigation system is developed, where all sensor devices are connected
to Raspberry Pi to obtain the data in paper. These data are then monitored and analyzes. The
watering activity is done using Android application as a Graphical User Interface (GUI). At certain
critical condition such as low water level, users will be notified by email application.
This cultivation kit is mounted with multiple sensors used to retrieves data which affects the
growth of crops. These data include water level, soil moisture, temperature, humidity, pH level, and
carbon dioxide level. Images of the crops are also obtained using a camera and sensors. In addition,
the device has a built-in speaker, and it is possible to generate information such as lack of water,
high and low temperature, and growth status depending on the condition of crops. The data would be
presented in natural language instead of numerical values. Furthermore, by incorporating a
microphone, we aim to realize a system that responds more interactively to the farmer’s questions.
CHAPTER 3
SYSTEM ANALYSIS
3.1.1 DISADVANTAGES
Duration is more.
Waste of time.
Lack of Infrastructure.
High Cost.
Lack of Security.
lack of information.
Operations such as seed planting, weeding and harvesting would be done efficiently.
Faster Access
Dept. of CSE, BGSIT, B.G Nagar Page 10
Interactive cultivation system for the future IOT-based agriculture 2021-2022
The Figure Shows the example of IOT based tomato plucking machine which is using in the
agriculture land for plucking the good quality tomato. The essential feature of precision farming is a
crop monitoring apparatus. Similar to weather forecasting stations, these devices are installed in the
farm to collect data needed for farming, including soil temperature, leaf-water potential
precipitation, and overall crop production. Similar to crop monitoring, there are IoT system devices
that can be used to monitor the health of livestock and track their performance.
Large farm owners could use wireless sensors to gather data about the location and health of
their livestock, which can minimize labor costs such as ranchers to locate cattle. IoT devices may
warn farmers about certain disease-affected animals and isolate them from the herd, preventing
diseasespread.
IoT enables easy collection and management of large amounts of data collected from
used sensors with the help of joining experimental distribution such as cloud storage, agricultural
field maps, and other details that can found anywhere and everywhere that enables live
monitoring and end-to-end connections.
Smart Farming IOT Based Agriculture Stick for Live Monitoring of Temperature and
Soil Moisture has been proposed using Node MCU Chip, Wifi Module, and various other
Hardware Devices. The stick has high efficiency and accuracy in fetching the live data of
temperature, humidity, and soil moisture. The IoT based Agriculture stick being developed
through this paper will help farmers in increasing the agriculture yield and take efficient care of
food production as the stick will always provide a helping hand to farmers for getting accurate
live feed of environmental temperature and soil moisture with accurate results. With the help of
these systems, various problems faced by farmers in daily life are being solved to a greater
extent.
Using the Internet of Things in agriculture is undeniably the smartest way to feed
humanity. But IoT promises even more. It's one of the few realistic ways to sustainably feed a
growing population. IoT enhances crop monitoring and produces crops on the maximum power
of implant. The impact of the Internet of Things (IoT) and connected devices in this modern-day
world is undeniable. Today it has reached almost everywhere, from home to the health sector,
smart cities, fitness, to the industrial sector.
Expansion:- By the time we have 9 billion people on the planet, 70% of them will live in
urban areas. IoT-based greenhouses and hydroponic systems enable short food supply
chains and should be able to feed the people. Smart closed-cycle agricultural systems
allow growing food everywhere—in supermarkets, on skyscrapers' walls and rooftops, in
shipping containers, and, of course, in the comfort of everyone's home.
Reduced resources:- Plenty of IoT solutions are focused on optimizing the use of
resources—water, energy, land. Precision farming using IoT relies on the data collected
from diverse sensors in the field which helps farmers accurately allocate just enough
resources within one plant.
Clean process:- Not only do IOT-based systems for precision farming help producers
save water and energy and, thus, make farming greener, but also significantly scale down
on the use of pesticides and fertilizer. This approach allows getting a cleaner and more
organic final product compared to traditional agricultural methods.
Agility:- One of the benefits of using IoT in agriculture is the increased agility of the
processes. Thanks to real-time monitoring and prediction systems, farmers can quickly
respond to any significant change in weather, humidity, air quality as well as the health of
each crop or soil in the field. In the conditions of extreme weather changes, new
capabilities help agriculture professionals save the crops.
Improved product quality:- Data-driven agriculture helps both grow more and better
products. Using soil and crop sensors, aerial drone monitoring, and farm mapping,
farmers better understand detailed dependencies between the conditions and the quality of
the crops. Using connected systems, they can recreate the best conditions and increase the
nutrition.
3.4 APPLICATIONS
CHAPTER 4
SYSTEM CONFIGURATION
Raspberry Pi Camera Module v2: Used to acquire the camera images of the
white radishsprouts for growth analysis.
Temperature and Humidity module (DHT 11): Used to determine the
temperature andhumidity of the surrounding air simultaneously.
Light sensor (GL5528): Used to detect the surrounding light intensity.
The software for controlling interactive devices consists of the following modules.
Sensor Module: The various sensors are activated based on the schedule set for each crop,
and the necessary data are formatted. By utilizing the time management function installed by
the operating system (OS), flash photography is performed. at night. An infrared camera
was used for crops that had a prohibition period of lighting. The program and the settings in
the moduleare switched depending on the type of crops.
Communication Module: The data formatted by the sensor module is sent to the server.
Besides that, based on the result of learning by the server, the estimated condition of the
crops is transferred to the speech module.
Speech Module: Based on the estimation condition from the server, a function to convey the
condition to the farmer as a voice is realized.
External database: Sensor data and images from sensor modules would be collected and
combined with cultivation advice from well-established farmers to produce appropriate
cultivation advice as output. This advice would be transmitted as reference for new farmers in
case they face difficulties in crop cultivations depending on the situation of crops. In
cooperation with many other interactive devices, big data for cultivation support according to
various regions, climates and environments are constructed.
A gas sensor is an electronic device to detect the concentration of air in atmosphere.In general a
normal human can able to identify one trillion types of odours. Similarly the different gas sensors
are used to identify odours. From that sensor we can obtain the voltage value to predict the
concentration of air. In our project gas sensors are used to identify the ripening of fruits.So that
the farmers can able to predict the time of fruits need to be plucked from field.
Level sensor: It is used to detect the level of water and other liquids that are available in
agriculture field. It can measure large changes in water level. It is one of the important
sensors and play important role in a variety of industrial application. The main purpose of a
liquid level sensor is to monitor and control the level of granular materials.
CHAPTER 5
SYSTEM ARCHITECTURE
Fig. 5.1. shows the interactive sensing system in the form of a cultivation kit proposedinthis
study.
This cultivation kit is mounted with multiple sensors used to retrieves data which affects the
growth of crops. These data include water level, soil moisture, temperature, humidity, pH level,
and carbon dioxide level. Images of the crops are also obtained using a camera. Both data and
images would then be analyzed to determine the degree of growth. The cultivation kit is
equipped with a network function and sensor data is collected by the server. Based on sensor
data collected in various home gardens and other outdoor environments, machine learning can be
performed as big data, together with image data. Thus, it is possible to associate the growth
status of the crops with sensor information.
Dept. of CSE, BGSIT, B.G Nagar Page 17
Interactive cultivation system for the future IOT-based agriculture 2021-2022
One of the most promising agritech advancements is the use of agricultural drones in smart
farming. Also known as UAVs (unmanned aerial vehicles), drones are better equipped than
airplanes and satellites to collect agricultural data. Apart from surveillance capabilities, drones
can also perform a vast number of tasks that previously required human labor: planting crops,
fighting pests and infections,agriculture spraying, crop monitoring droneseed for example,
builds drones for planting trees in deforested areas. The use of such drones is 6 times more
effective than human labor.
This cultivation kit is mounted with multiple sensors used to retrieves data which affects the
growth of crops. These data include water level, soil moisture, temperature, humidity, pH level,
and carbon dioxide level. Images of the crops are also obtained using a camera. Both data and
images would then be analyzed to determine the degree of growth. The cultivation kit is
equipped with a network function and sensor data is collected by the server. Based on sensor
data collected in various home gardens and other outdoor environments, machine learning can be
performed as big data, together with image data. Thus, it is possible to associate the growth
status of the crops with sensor information.
Currently, robots are developed with autonomous navigation and used in paddy fields for
weed growth suppression. A robot that utilizes a laser range finder for autonomous mobility, and
a robot arm for weed suppression is developed in paper [25]. It travels along the rice rows
without damaging the rice plants in a paddy field. In paper [26], ―Paddy Weeding Robot‖ is
proposed. It functions by agitating the soil and creating an environment that makes weeds
difficult to photosynthesize, therefore inhibiting their growth. Nissan has also developed
―Aigamo Robot‖ as a method of reducing the use of herbicide and pesticide on rice crops [27]. It
features two rotating brushes as paddles to stir up the water in the paddy field and thus
preventing weed growth, similar to the Paddy Weeding Robot. Aigamo Robot uses Wi-Fi,
batteries, solar power, and Global Positioning System (GPS) to navigate autonomously. With the
concept of being low-cost, low-maintenance and high-durability, we proposed and designed a
compact-sized interactive robot ―Paddy Drone‖ as support for rice cultivation. The prototype of
Paddy Drone is shown in Fig. 7. Using Wi-Fi and an interactive device similar to Fig. 2 mounted
on Paddy Drone, it can be programmed to travel across the water surface of the paddy field
automatically.
Fig 5.3 shows an overview of the interactive device. In the device, Raspberry Pi, a
single-board computer, was used to acquire multiple growth data and send them to an
external server. A variety of sensors are equipped to acquire growth- related data based on
temperature, humidity, light intensity, soil moisture, and pH level, as well as conditions of
the crops. Multiple camera modules are equipped to obtain camera images. The height and
the thickness of the stems of the crops are then analyzed the obtained stereo images.
In addition, the device has a built-in speaker, and it is possible to generate information such
as lack of water, high and low temperature, and growth status depending on the condition of
crops. The data would be presented in natural language instead of numerical values.
Furthermore, by incorporating a microphone, we aim to realize a system that responds more
interactively to the farmer’s questions.
FUTURE ENHANCEMENT
For development in the future, it can be enhanced by developing this system for
large acres of land. Also, the system can be integrated to check the quality of the soil and the
growth of the crop in each soil. The sensors and microcontroller are successfully interfaced
and wireless communication is achieved between various nodes. Future work also includes
focusing more on increasing sensors on this stick to fetch more data especially concerning
Pest Control and by also integrating GPS modules in this IOT Stick to enhance this
Agriculture IOT Technology to a full-fledged Agriculture Precision ready product. As they
speak, the number of companies that will help empower their IoT (Internet of Things) ideas.
And as a result, we hear of new ideas and solutions that are already solving business
challenges through M2M (Machine to Machine) communications. In a recent post, we
discussed some of our favorite IoT industrial applications. Measuring soil moisture is
important in agriculture to help farmers manage their irrigation systems more efficiently.
Not only are farmers able to generally use less water to grow a crop, but they can also
increase yields and the quality of the crop by better management of soil moisture during
critical plant growth stages.
The basic building blocks of an IoT System are Sensors, Processors, and applications. So the
block diagram below is the proposed model of our project which shows the interconnection
of these blocks. The sensors are interfaced with Microcontroller, data from the sensor is
displayed on the mobile app of the user. A mobile app provides an access to continuous data
from sensors and accordingly helps the farmer to take action to fulfill the requirements of the
soil. Farming is a labor-intensive task that requires lots of time and effort. Usually, these
tasks are repetitive and monotonous. Farmers can delegate these labor-intensive tasks to
robotics and automation-based solutions. Such solutions can perform tasks that range from
seeding and watering to harvesting and sorting.
CONCLUSION
REFERENCES