Artificial Intelligence in Agriculture

II. EXISTING SYSTEM

Abstract: We all know that in our contemporary day, Agriculture, or farming, is our country's largest
technology plays a very vital part in every work. In every industry, and it plays a critical part in the country's
field technology is employed to boost work efficiency, for
social and economic development. Farming is done
less time consumption and for perfect and better output.
using traditional techniques and ways before adopting
Human beings require three basic things: food, shelter, and
a technology in agriculture. Farming is influenced by a
clothing. Increasing the population leads to better farming
variety of elements, or to put it another way, farming
results. Nowadays, technology such as artificial intelligence
is used to improve farming processes, resulting in higher is influenced by a variety of things. In India, we
yield quality. This study presents a picture of how discovered a variety of habitats, weather patterns, and
technology might be used to fuel the outcomes of various soil types that are ideal for various crops in various
agricultural sectors. locations. Farmers used to harvest a certain type of
Keywords: Artificial Intelligence, Technology, Farming, crop or a crop cycle. In most parts of India,
Machines agricultural is dependent on rain or water supply. Crop
selection, land preparation, seed selection, seed
I. INTRODUCTION sowing, irrigation, crop development, fertilizing, and
harvesting are the eight primary tasks farmers do. And
Because to technological advancements, everything has gotten all of these stages are done by hand, which
faster and easier. When the industrial revolution occurred in the
necessitates extra human and animal work.
nineteenth century, machines were used to replace humans. It
minimizes human labor while also improving the quality of the
output. Artificial intelligence is now enabling human work, III. LITERATURE REVIEW
which is a fact. Artificial Intelligence is quickly adapting to its
varied tactics in agriculture. The current scenario necessitates Crop management systems, in general, provide an
more effective farming techniques utilizing technical interface for total crop management, encompassing all
breakthroughs and solutions. Artificial Intelligence (AI) will aspects of farming. McKinion and Lemmon originally
help farmers do more with less by improving the quality of advocated utilizing AI techniques in agricultural
their crops. Manpower was one of the most important aspects management in their article "Expert Systems for
of farming prior to the industrial revolution. Different Agriculture" [5] in 1985. Boulanger presented another
approaches are utilized nowadays, such as disease detection, corn crop protection expert system in his PhD thesis
determining crop readiness, determining which fertilizer to use [6]. Roach et al. presented the POMME expert system
based on soil type[11], field management, and so on. Climate for apple plantation management in 1987 [7].
change, population increase, and food security concerns are all COTFLEX [8] is an expert system developed by
factors that are causing the sector to create more inventive Stone and Toman for cotton crop management.
techniques to safeguard and boost crop productivity. [9] As a Another expert system based on rules.
consequence, the farming business is utilized. [5] Microsoft
Corporation is now operating in Andhra Pradesh, India, with Artificial intelligence (AI) should be used to improve
175 farmers providing various services. [12] Now comes the agricultural areas at minimal cost and with ease of
question of what Artificial Intelligence is. Artificial intelligence processing. Various agricultural difficulties are
refers to the replication of human intellect in computers that managed in a short period of time using Artificial
have been trained to think and act like humans. Intelligence. Artificial intelligence uses tactics such as
improving harvesting quality and introducing indoor
Artificial Intelligence (AI) is one of the most important fields farming to increase agricultural output rates. There
in computer science study. AI is fast becoming prevalent due to are numerous AI applications that will actually assist
its robust applicability in situations that cannot be handled farmers, such as analyzing farm data to improve crop
adequately by people or traditional computer architectures [1], quality and precision, detecting target weeds using AI
as well as its rapid technical improvement and broad range of sensors, and detecting illnesses in plants, pests, and so
applications. Agriculture is one such critical area, with 30.7 on. AI addresses labor issues. As we all know, less
percent of the world's population directly working on 2781 people are entering this field, thus farmers are
million hectares of agricultural land. From the seeding through experiencing a labor scarcity and a lack of personnel.
the harvest, such an endeavor is not without its difficulties. Pest The solution to this is agriculture bots, who will work
and disease infestation, inappropriate chemical application, alongside farmers. This bot harvests crops in more
incorrect drainage and irrigation, weed management, yield quantity and at a quicker rate. In Blue River
forecast, and other challenges are key concerns. The first report Technology, there are agricultural robots that are used
on the use of computers in agriculture was published in 1983 to manage weeds. Harvest CROO Robotics, which
[2]. To overcome the current difficulties in agriculture, several leads to agricultural harvesting, has developed a robot
techniques have been proposed, ranging from databases [3] to for farmers to pick and pack crops.
decision support systems [4]. Systems that use AI have been
determined to be the most excellent performers in terms of
accuracy and robustness among these solutions.
AI can also do diagnostic analyses, such as Satellites for Farmers' expertise was utilized by Sicat et al. [6] to
Weather PredictionCrop Sustainability, which would be develop a fuzzy based system that recommended crops
extremely beneficial to farmers who had a prior understanding based on land suitability maps created by the fuzzy
of weather variations. Driverless Tractors are one of the AI system. Si et al. [6] and Tremblay et al. [7] are two such
approaches, and because they run without the presence of a fuzzy-based systems. Based on meteorological and soil
person inside the tractor, they will save farmers a lot of time water content data, Valdes-Vela et al. employed a Takagi
and effort. Sugeno Kang fuzzy inference system to predict a plant's
stem water potential [8]. Arif et al. [9] developed an
One of the most intriguing technologies to mention is Farmer's artificial neural network-based approach for estimating
Alexa, which will be able to speak with farmers in the same soil moisture in paddy.
way as Chabot’s do to solve difficult problems. Aerial spraying
is five times faster with drones than with regular technology, IV. APPLICATIONS OF AI IN AGRICULTURE
which benefits farmers. Agri-ECalculator is one of the smart
The most widely used AI applications in
apps launched in AI for farming, which assists farmers in
selecting good and inexpensive crops and calculating their agriculture seems to fall into three categories:
prices. There are many more programs on the market, but the
1. Agriculture robotics:
problem is that they are expensive and difficult to use. In basic
terms, the application of AI in agriculture allows farmers all Companies are creating and programming
over the world to operate more effectively. Insect pest self-driving robots to undertake farm jobs in
infestation is one of the most concerning issues in agriculture, greater volume and at a faster rate than
resulting in significant financial losses. people.[8]
2. Crop and soil monitoring:
Researchers have spent decades trying to lessen this threat by Drones or software-based technologies are
developing computerized systems that can identify active bugs utilized to collect/capture data for crop and
and recommend management methods. Pasqual and Mansfield
soil health monitoring, and computer vision
[13], Batchelor et alSMARTSOY .'s [14-15], Mozny et
alCORAC .'s [6], Knight and Cammell [13], Mahaman et al. and deep-learning algorithms are utilized to
[8], Li et al. [9], Chakraborty et al. [10], and Ghosh's Ghosh's process the data.[2]
Ghosh's Ghosh's Ghosh's Ghosh' 3. Predictive analytics:
Models based on artificial intelligence are used to
Artificial Intelligence (AI) may be used in a variety of fields, track and anticipate many elements that impact
and it has the potential to change the way we think about agricultural output, such as weather change. [2].
farming. As more high-tech machines are used in everyday
life, such as education, hospitals, and even government, AI-
Disease detection:
powered solutions will not only enable farmers to accomplish
more with less, but they will also assist farmers in obtaining a Image sensing and analysis guarantee that plant leaf
higher output. Agriculture is the most reverberated of all, with pictures are divided into surface areas such as
artificial intelligence focusing on ease of use and efficiency. infected, non-diseased, and background regions.
Artificial intelligence (AI) should be used to improve Cropped photos of unhealthy or infected areas are
agricultural areas at minimal cost and with ease of processing. forwarded to the laboratory for further examination,
Various agricultural difficulties are managed in a short period which leads to pest identification and nutritional
of time using Artificial Intelligence. Artificial intelligence uses
insufficiency.[1]
tactics such as improving harvesting quality and introducing
indoor farming to increase agricultural output rates.
Methods:
Because agricultural management information is frequently 1.Data description
incomplete, unclear, and imprecise, the rule-based expert 2.Measurement of performance
system may cause ambiguity. Several Fuzzy logic based expert 3.Approach
systems, such as Saini et al. [2], Siraj and Arbaiy [3], Peixoto
et al. [4], IPEST by Hayo et al. [5], Roussel et al. [6], Shi et al. Total 60 experimental configuration depend od
[7], Jesus et al. [8], have been suggested to capture this
following parameters:
uncertainty. Ghosh et al. used an object-oriented approach to
design a rule basis for creating TEAPEST, an expert system Choice of deep learning architecture:
for tea pest control [9]. A phase-by-phase identification and 1.AlexNet,
consultation approach was also used here. Later, Samanta and 2. GoogLeNet.
Ghosh modified the system using a multi-layered back
propagation neural network. In agriculture, soil and irrigation Choice of training mechanism:
management are extremely important. Crop loss and quality 1.Learning Transfer
degradation are caused by poor irrigation and soil 2.Training from the ground up.
management. This section covers some soil and irrigation
Choice of dataset type:
management research that has been aided by artificial
intelligence tools. Brats et al. [4] developed a rule-based expert Color,
system for evaluating micro irrigation system design and 1.Gray scale,
performance. 2.Leaf Segmented.
4. Choice of training-testing set
distribution: Train: 80%, Test: 20%,
Train: 60%, Test: 40%,
Train: 50%, Test: 50%, 1.A grayscale,
 2.Segmented leaves. V. CHALLENGES IN AI ADOPTION IN
AGRICULTURE
When something new enters the system, it is tough for
b. Identify readiness of crop: everyone to adapt it, and artificial intelligence in
It's not as simple as it sounds to recognize a ripe green agriculture presents several obstacles. Every new thing
apple. It necessitates a high level of competence and is is simple to utilize once we understand it and are
nevertheless subject to human mistake. We can familiar with it. Many farmers who are technologically
achieve reliable findings without the need for human savvy can benefit from it, but what about those who
interaction when we use AI. We use a camera to live in remote areas, One of the most difficult issues is
capture photos in the presence of white and ultraviolet persuading them that using technology may boost field
light, which are subsequently analyzed on a computer. production. [7]
The computer's result will be deemed the definitive
outcome. [3]
VI.CONCLUSION
Artificial intelligence has a place in the farming sector
since it enhances asset utilization and efficiency. It
explains a great deal about the lack of assets and work.
The use of AI in agriculture is really beneficial.
Artificial intelligence has the potential to be a game-
changer in agribusiness, helping to feed the world's
growing population. Farmer decision-making will be
aided and hampered by artificial thinking. There are
just too many pieces to discuss in this little area. To
Fig 2: Detection of Fruit Ripening cover the various methods, just a few representatives
are chosen to be discussed.
c. Field management:
Using various AI approaches to manage the field to This analysis was written to be as helpful as possible,
increase profit, it is always preferable to take safeguards including details on various AI approaches used in
for everything. Using AI, we can foresee weather changes, agriculture. Rule-based expert systems were widely
employed in the early 1980s and 1990s, but from 1990
future crop and soil needs, and therefore avoid future
onwards, artificial neural network models and fuzzy
failure.[4][11] inference systems have taken over. Hybrid systems,
such as neuro-fuzzy or image processing combined
with artificial neural networks, have become
increasingly popular in recent years. It progresses
toward more automated, precise, and real-time
systems. More research is being carried out using
more modern equipment in order for conventional
agriculture to transition into precision agriculture.

Fig 3: Management of the Field

VII. FUTURE SCOPE
Artificial Intelligence's future is difficult to forecast.
In the 1990s, artificial intelligence was primarily
d. Crop Health monitoring: focused on improving research and development; but,
is this the sole aim in the future? Research is now
For a good yield of produced crops, health monitoring is focusing on creating human-like machines or robots.
critical. Crop matrices may be built across acres of land The role of humans will undoubtedly alter if machines
using remote sensing and 3D laser technology.[5] begin to do tasks for humans. Researchers' efforts may
be rewarded in the future, when we will discover our
e. Precision Farming: work being done by machines and a robot strolling
Precision farming is a methodology or practice that aids alongside us. In the future, we will see robots laboring
farmers in increasing productivity and improving overall in agricultural fields, resulting in higher yields with
better quality.
performance. This is the morden farm strategy for
increasing profit, lowering costs, and reducing pesticide
and fertilizer use[6]. Precision farming's objectives:
a. Profitability
Recognize crops and market strategies
b. Sustainability
Different operations for improvement based on
seasons.
c. Efficiency
Allow to use all resources effectively.[5]
VIII. REFERENCES
[7]Joseph Byrum,”The Challenges for Artificial
[1] Using Deep Learning for Image-Based Plant Disease
Detection, by Sharada Prasanna Mohanty, David Peter Intelligence in Agriculture”, February 20, 2017
Hughes, and Marcel Salathe, was published on April 15,
2016.
[8]David Ball,Patric Ross,”Robotics for Sustainable
[2] Kumba Senar,”AI in Agriculture-Present Applications and Broad-Acre Agriculture”, January 2015
Impacts”, November 21,2019
[9]Parul Sharma,Wiqas Ghai,”Performance analysis of deep
[3] Using machine learning and image processing, a learning CNN models for disease detection in plants using
computer vision system for flaw recognition and grading image segmentation”, 18 November 2019
in tomatoes, says DavidIreri. 17th of June, 2019
[10]Verónica Saiz-Rubio,”From Smart Farming towards
Agriculture 5.0: A Review on Crop Data Management”, 3
[4] Monitoring and Control Systems in Agriculture Using
Intelligent Sensor Techniques: A Review of the Aeroponic February 2020
System, Imran Ali Lakhiar, Gao Jianmin, Tabinda Naz Syed, [11]Alexander M. Petrov, Mikhail A. Kanaev,”Technical
and Farman Ali Chandio, 19 December 2018. devices used for differential fertilizer application”, January
2020
[5]Paras M. Khandelwal and Himanshu
[12]Kasmin Fernandes,”Artificial Intelligence in
Chavhan,”Artificial Intelligence in Agriculture: An
Emerging Era of Research”,September 2019 Agriculture in India”, January 10 2020
[13]Victor Mokaya,”Future of Precision Agriculture in India
[6]K. Mufeedha,E. Abhilash Joseph,V. M. Abdul using Machine learning and Artificial Intelligence”, March
Hakkim,”Precision Farming: The Future of Indian
Agriculture”, November 2016 2019