Press “A” for Artificial Intelligence in Agriculture: A Review

Abstract

Agriculture is the backbone of the developing country. In old era agriculture was based on the experience which was shared by people to people but in this digital era technology play a very important and significant role in agriculture. Now agriculture become a business hub therefore farmers are focusing on precision farming. They introduced the technology in agriculture to define the accurate information about seed, soil, weather, disease and all factors which affecting the farming. Artificial Intelligence uses predictive analysis, image analysis, learning techniques and Pattern analysis to declare the best cost effective and maximum gain for the agriculturist. The aim of this paper is to provide the crucial information with the help of technology which a farmers can use to harvest the variety of crops as per the demand in world so that they can get maximum benefits.

Full text

112
Press “A” for Artificial Intelligence in Agriculture: A Review
Yogesh Awasthi
#
#
Department of Computer and Computer Information Systems, Africa University, Mutare, Zimbabwe
E-mail: awasthiy@africau.edu
Abstract— Agriculture is the backbone of the developing country. In old era agriculture was based on the experience which was
shared by people to people but in this digital era technology play a very important and significant role in agriculture. Now agriculture
become a business hub therefore farmers are focusing on precision farming. They introduced the technology in agriculture to define
the accurate information about seed, soil, weather, disease and all factors which affecting the farming. Artificial Intelligence uses
predictive analysis, image analysis, learning techniques and Pattern analysis to declare the best cost effective and maximum gain for
the agriculturist. The aim of this paper is to provide the crucial information with the help of technology which a farmers can use to
harvest the variety of crops as per the demand in world so that they can get maximum benefits.
Keywords— artificial intelligence; agriculture; analysis; precision farming; information technology.
I. I
NTRODUCTION
Agriculture and Technology are complementary to each
other in today’s world. Current cultivating and country
exercises work far extraordinarily rather than those a few
decades earlier, essentially by virtue of progress in
development, including sensors, devices, machines, and
information advancement. The current agribusiness routinely
uses refined advancements, for instance, robots, temperature
and sogginess sensors, flying pictures, and GPS
development. These pushed devices and exactness
agribusiness and mechanical systems grant associations to be
continuously gainful, powerful, progressively secure, and
even more earth neighbourly. The ascent of electronic
cultivating and its related headways has opened a wealth of
new data openings. Remote sensors, satellites, and UAVs
can collect information 24 hours out of every day over an
entire field [1][2] These can screen plant prosperity, soil
condition, temperature, tenacity, etc. The proportion of data
these sensors can make is overwhelming, and the criticalness
of the numbers is concealed in the uncontrolled slide of that
data. The idea is to allow ranchers to build an unrivalled
appreciation of the condition on the ground through front
line development, (for instance, remote identifying) that can
reveal to them more about their situation than they can see
with the independent eye. Furthermore, more absolutely just
as more quickly than seeing it walking or going through the
fields. Since we can see that these innovations are
profoundly received and will keep on being embraced by
farmers, we can investigate a portion of the advantages that
accompany accuracy agribusiness innovation. The main idea
behind the utilizing these cutting-edge innovations
incorporates: Efficiency being used of assets like synthetic
substances, composts, water, fuel, and so on., Improving
amount and nature of produce, Mellower yield in same size
of land, Reducing ecological impression, Risks mitigation
[3][4].
II. L
ITERATURE
R
EVIEW
Artificial intelligence innovation is supporting various
divisions to help profitability and productivity. Computer
based intelligence arrangements are helping to beat the
conventional difficulties in each field. Similarly. Computer
based intelligence in horticulture is helping farmers to
improve their proficiency and lessen natural antagonistic
effects.[5][6]. The horticulture business emphatically and
transparently grasped AI into their training to change the
general result. Computer based intelligence is moving the
manner in which our food is delivered where the farming
area's outflows have diminished by 20%. Adjusting AI
innovation is assisting with controlling and deal with any
excluded characteristic condition [7].
Today, most of new businesses in farming are adjusting
AI-empowered way to deal with increment the proficiency
of agrarian creation. The Market study report expressed that
the worldwide Artificial Intelligence (AI) in Agriculture
showcase size is required to arrive at 1800 million
US$ before the finish of 2030. Actualizing AI-engaged
methodologies could identify sicknesses or atmosphere
changes sooner and react sagaciously. The organizations in
farming with the assistance of AI are handling the agrarian
information to lessen the unfavourable results [8].
INTERNATIONAL JOURNAL ON INFORMATICS VISUALIZATION
VOL 4 (2020) NO 3
e-ISSN : 2549-9904
ISSN : 2549-9610

113
Computer based intelligence in a propelled manner is
helping the farmers to remain refreshed with the information
identified with climate anticipating. The assessed/foreseen
data help farmers with growing returns and advantages
without taking a risk with the reap. The examination of the
data made makes the ranchers avoid any and all risks by
cognizance and learning with AI. By actualizing such
practice assists with settling on a savvy choice on schedule.
Using AI is a productive method to lead or screen
distinguishes potential imperfections and supplement lacks
in the dirt. With image recognition approach, AI
distinguishes potential deformities through pictures caught
by the camera. With the assistance of Al profound learning
application are created to examination verdure designs in
farming. Such AI-empowered applications are strong in
understanding soil deserts, plant bugs, and un-wellness
[9][10].
Farmers can utilize AI to oversee weeds by actualizing
computer vision, apply autonomy, and AI. With the
assistance of the AI, information are accumulated to keep a
mind the weed which causes the farmers to shower synthetic
concoctions just where the weeds are. This legitimately
decreased the use of the substance splashing a whole field.
Therefore, AI decreases herbicide use in the field relatively
the volume of synthetic substances ordinarily splashed [11].
Computer based intelligence empowered agribusiness
bots help farmers to discover progressively effective
approaches to shield their yields from weeds. This is
likewise assisting with beating the work challenge.
Computer based intelligence bots in the farming field can
collect harvests at a higher volume and quicker pace than
human workers. By utilizing PC vision assists with
observing the weed and splash them. Consequently,
Artificial Intelligence is helping ranchers discover
increasingly proficient approaches to shield their yields from
weeds [12].
III. I
MPACT OF
A
RTIFICIAL
I
NTELLIGENCE
M
ETHODS ON
A
GRICULTURE
The man-made consciousness (AI) innovation is
supporting various segments to help their profitability. The
AI arrangements have overcome the difficulties looked by a
few ventures and now it is consistently making its place in
the farming segment as well. Computer based intelligence
advances sizably affect the horticulture segment. This
section focuses on the few AI methods or techniques which
majorly related with the Agriculture:
A. Autonomous Tractor
With the substantial interest in creating self-sufficient
vehicles for different requirements, the horticulture area will
be likewise getting benefits with self-driving or you can say
driverless tractors.
With greater quality AI and machine learning information
for horticulture, the farm segment will be reformed by the
enormous scope utilization of autonomous tractors for
playing out different undertakings.
These self-driving or driverless tractors are modified to
freely identify their furrowing position into the fields or
choose the speed and keep away from impediments like
water system items, people and creatures while performing
different assignments.
B. Agriculture Robotics
So also, AI organizations are creating robots that can
without much of a stretch play out numerous assignments in
the cultivating field. Such apply autonomy machines are
prepared to control weeds and collect the yields at a lot
quicker pace with higher volume contrast with humans.
These robots are all around prepared to help for checking the
nature of harvest and recognize undesirable plants or weeds
with picking and pressing of harvests simultaneously fit to
battle with different difficulties looked by the rural work
force. Companies like Blue River Technology and Harvest
CROO Robotics are making such mechanical technology
machines that can control undesirable yields or weeds and
help ranchers in picking or pressing of yields with higher
volumes.
C. Pest Infestation Control
Pests are one of the most exceedingly terrible foes of the
farmers harming the yields universally before it is gathered
and put away for human utilization. Mainstream bugs like
beetles, grasshoppers, and different creepy crawlies are
eating the benefits of farmers and eating the grains implied
for people. Be that as it may, presently AI in cultivating
gives producers a weapon against such bugs.
Computer based intelligence and information
organizations are helping farmers to get alert on his
Smartphones about the grasshoppers liable to slip towards a
specific homestead or developed harvest field.
Simulated intelligence organizations utilizing the new
satellite pictures against photos of a similar utilizing
authentic information and AI calculation distinguishes that
the bugs had arrived at another area and farmers utilize such
data after affirmation and convenient expel the exorbitant
nuisances from their fields.
D. Health Monitoring of soil and crop
Proceeds with deforestation and debasement of soil
quality are turning into a major test for food delivering
nations. In any case, presently a German-based tech startup
PEAT has built up a profound learning-based application
called Plan tix that can recognize the expected imperfections
and supplement insufficiencies in the dirt including plant
bugs and ailments.
This application is chipping away at picture
acknowledgment-based innovation and you can utilize you
your cell phone to catch the plant's picture and distinguish
the deformities into the plants. You will likewise get soil
reclamation procedures with tips and different arrangements
on short recordings on this app. Similarly, Trace Genomics
is another AI based organization gives soil investigation
administrations to reformers. Such applications help farmers
to screen the dirt and harvest's wellbeing conditions and
produce a solid yield with a more elevated level of
productivity. Sky Squirrel Technologies procured by another
comparable organization Vine View brought drone-based
ethereal imaging answers for checking crops wellbeing. An
automaton is utilized to make a series of catching the
information from the vineyards field and afterward all the

114
information is moved by means of a USB drive from the
automaton to a PC and dissected by the experts [13]. The
organization utilizes the calculations to examine the caught
pictures and gives a point by point report containing the
momentum soundness of the vineyard, for the most part the
state of grapevine leaves as these plants are profoundly
inclined to grapevine infections like moulds and microbes
helping ranchers to ideal control utilizing the nuisance
control and different strategies.
E. Predicting water levels using small data sets
This innovation executes ideas from power through
pressure utilized for structural designing estimation and
development work at streams and sea shores. We made a
capacity dependent on the tank model, utilizing AI to
process past precipitation and water level information.
Through this, we constructed a numerical model to infer
ideal boundaries. This permitted us to foresee water levels
dependent on as meagre as three days of precipitation, water
levels, and woodland precipitation information [14].
F. Weather Forecasting
Artificial intelligence in a propelled manner is helping the
farmers to remain refreshed with the information identified
with climate gauging. The estimated/anticipated information
assist ranchers with expanding returns and benefits without
taking a chance with the yield. The investigation of the
information produced causes the rancher to avoid potential
risk by comprehension and learning with AI. By actualizing
such practice assists with settling on a keen choice on
schedule. IBM's Deep Thunder and Monsanto's Climate
Corporation is utilized to give rural climate expectations.
G. Maximize ROI
Computer based intelligence can assist farmers with
picking the correct kind of yield. In light of information,
they can decide the correct blend of harvests that are
tweaked for different needs and climate. Computer based
intelligence advances can likewise give bits of knowledge on
how a specific kind of seed will respond to a specific sort of
soil profile, neighbourhood atmosphere conditions and
climate estimates. By relating and examining this data, the
year-to-year result can be advanced and thus, ROI can be
amplified.
H. Image Analysis
Picture acknowledgment is another headway that would
permit Farmers to screen their territory and harvests all the
more rapidly and effectively, and furthermore comprehend
past examples after some time. Computer based intelligence
is being prepared to perceive more than 5000 types of plants
and creatures, which would improve drone capacity to
distinguish bother ailment and harvest harm. Undesirable
plants developing in ranches can likewise be identified by
joining picture handling and AI methods. Picture handling
can likewise be utilized in natural product reviewing
frameworks to fragment and arrange with extraordinary
exactness. With right imaging strategies and calculations, the
grouping exactness of up to 96% can be acquired.
I. Manpower Challenge
Computer based intelligence empowered agribusiness
bots help farmers to discover increasingly proficient
approaches to shield their harvests from weeds. This is
likewise assisting with conquering the work challenge.
Simulated intelligence bots in the farming field can collect
yields at a higher volume and quicker pace than human
workers. By utilizing AI vision assists with observing the
weed and splash them. In this manner, Artificial Intelligence
is helping farmers to discover progressively effective
approaches to shield their yields from weeds [15[16].
Fig. 1 Adoption of Technology in Agriculture
IV. R
OLE
O
F
I
NFORMATION
T
ECHNOLOGY
I
N
A
GRICULTURE
Here figure 2 shows the how an information technology
can connect with the end users. It also depicts the areas
where IT may leave his impact. This figure 1 shows that
what the livelihood technology are connected with rural
population according to their skill set and requirement. Here
AI paly a very important role to collect the data set and
identify the requirements of the famers along with available
resources. Now the biggest question is that how the benefits
of these technologies communicate to the farmers. This is
the biggest challenge because the education for the villagers
or farmers are not too much tech savvy. To overcome this
problem, I have suggested a prototype model in figure 3.
Here the role coordinator in very important. He is the
interface between the farmers and Technology. He is directly
connected with the farmers and experts via internet in global
world.
Fig. 2 Relation of Information Technology with Farmers

115
V. P
ROPOSED
P
ROTOTYPE
M
ODEL FOR
F
ARMERS TO
U
SE
AI
This is the proposed model for the farmers to develop in
their areas. Here, Farmers register into the system by
sending their information about crop, soil and all other
related information. The coordinator sends the crop status
through the images and text data. Then the team Agriculture
expert collect the information and matches with artificial
Intelligence Information System and send back to the advice
to the farmer on weekly basis based on the crop status
supplied by the coordinator. The coordinator explains the
advices to the farmers. Farmers follow the advice take
appropriate steps and send the feedback. By this approach
this model will work. Here purchaser and investor may also
participate directly to the farmers so they can get proper
price of the crop.
VI. C
HALLENGES OF
A
RTIFICIAL
I
NTELLIGENCE IN
A
GRICULTURE
Man-made consciousness in horticulture division can be
actualized for different mechanical progressions. These
incorporate Machine Learning administrations, Artificial
Intelligence counselling administrations, information
investigation, web of things and accessibility of sensors and
cameras, and so on.
Fig. 3 A Prototype Model of Agriculture for Formers
Machine vision advances can possibly reform applications
in the agrarian segment. The utilization of some in
cultivating can be utilized in rural procedures like reaping,
utilization of exact weed-executing synthetic concoctions,
and so on but there are numerous difficulties to send the AI
in farming.
A. Skill Requirement
Usage of AI methods in agriculture requires a different
level of skill set. These technologies based on hardware,
software’s, sensors and various tools which requires a
training to operate properly. This prompt the proper
education should be incorporated at each level.
B. Response Time and Accuracy
In this instance, AI has a very important role. Sometime
farmers want to test small part of the soil for particular crop.
The behaviour of the crop should be analysed in the shortest
period of time with accurate information because to develop
a crop they have fixed season and if the time passes this will
be very difficult to carry on. Therefore, the response time
and accuracy play a vital role in agriculture.
C. Durability
Whenever farmers are going to deploy a technology that
must be durable. In this digital era technology is changing
very fast. This will not be economical for farmers to change
the devices and sensors in a short period of time because this
will not be cost effective for small scale farmers.
D. Initial Costing
As the government is planning to double the income of
farmer in next five years but still majority of farmers are not
so worthy to afford these type technologies in their fields.
The Main issue behind this is initial cost of installation. So
companies has to focuses on this issues how they can help to
farmers. They can start the usages of these machine or
technology on rental basis or percentage of crop basis.
E. Maintenance Cost
Maintenance of specialized hardware is also big issue.
Maintenance cost is also count as an investment of crop. If
the maintenance cost is higher, then price of your crop may
be higher than others. For small scale farmers this will not be
suitable and their crop may waste.
F. Regular Updates
Regular updates of machine and software’s are required.
As the technology updated farmer have to update their
system to get more accurate and fresh information. Some
system will not work properly with old version.
VII. C
ONCLUSION
Agribusiness is gradually getting advanced and AI in
farming is rising in three significant classifications, (i)
agrarian apply autonomy, (ii) soil and yield observing, and
(iii) prescient investigation. Farmers are progressively
utilizing sensors and soil inspecting to assemble information
and this information is put away on farm the executive’s
frameworks that takes into account better preparing and
examination. The accessibility of this information and other
related information is clearing an approach to send AI in
agribusiness.
Computer based intelligence System in agribusiness
helping farmers to computerize their cultivating as well as
moving to exact development for higher harvest yield and
better quality while utilizing less assets.
Artificial intelligence fuelled arrangements won't just
empower farmers to accomplish more with less, however it
will likewise improve quality and guarantee quicker go-to-
advertise for crops. While we may simply be at the
beginning phase of this change, here are some significant
ways AI is changing the agrarian part.
The AI-controlled innovations can help the agribusiness
division to yield more beneficial harvests, control bugs,
screen soil, and developing conditions, compose information
for ranchers, help with the remaining burden, and advance a
huge scope of horticulture related assignments in the whole
food gracefully chain. These developments to cultivating

116
have been significantly determined by environmental change,
populace development and food security concerns.
R
EFERENCES
[1] https://customerthink.com/the-role-of-artificial-intelligence-in-
agriculture-sector/
[2] https://emerj.com/ai-sector-overviews/ai-agriculture-present-
applications-impact/
[3] https://www.analyticsinsight.net/ai-in-agriculture-strengthening-the-
future-of-farming/
[4] https://www.analyticsinsight.net/ai-in-agriculture-strengthening-the-
future-of-farming/
[5] http://www.fao.org/e-agriculture/news/can-artificial-intelligence-
help-improve-agricultural-productivity
[6] https://www.plugandplaytechcenter.com/resources/artificial-
intelligence-agtech/
[7] Kumar, N., Y. Awasthi, and R. P. Agarwal. “Authenticating cloud
and data centre with iris,” International Journal of Engineering and
Research, vol 4(3), pp. 213-216, 2016.
[8] Deepak G. Panpatte. “Artificial Intelligence in Agriculture: An
Emerging Era of Research,” Intuitional Science, CANADA, 2018.
[9] R. Tomar and Y. Awasthi, "Prevention Techniques Employed in
Wireless Ad-Hoc Networks," 2019 International Conference on
Advanced Science and Engineering (ICOASE), Zakho - Duhok, Iraq,
pp. 192-197, 2019, doi: 10.1109/ICOASE.2019.8723725.
[10] N. Singh and Y. Awasthi, “WSN-AI based Cloud Computing
Architectures for Energy Efficient Climate Smart Agriculture with
Big Data analysis," 2019 International Journal of Advanced Trends
in Computer Science and Engineering, pp. 91-97, 2019.
[11] Thangadurai Natarajan and Niraj Prasad Bhatta," Utilization of IOT
and AI for Agriculture," International Journal of Advanced
Technology and Engineering Exploration,” vol. 8(5), pp. 2731-2735,
2019.
[12] Awasthi, Yogesh, R. P. Agarwal, and B. K. Sharma. “A Novel
Secure Watermarking Scheme for RDBMS,” International Journal of
Data Warehousing & Mining, vol. 4. 2014.
[13] Medar, Ramesh, Vijay S. Rajpurohit, and Shweta Shweta. "Crop
Yield Prediction using Machine Learning Techniques." 2019 IEEE
5th International Conference for Convergence in Technology (I2CT).
IEEE, 2019.
[14] Pathan, Misbah, et al. "Artificial cognition for applications in smart
agriculture: A comprehensive review." Artificial Intelligence in
Agriculture, 2020.
[15] Geli, Hatim, et al. "Climate Adaptive Smart Systems for Future
Agricultural and Rangeland Production.", 2019.
[16] Ruggieri, Sara. "Ai and robots: the impact on the labour market."
(2019).