REPORT ON:

APPLICATION OF MACHINE LEARNING AND

ARTIFICIAL INTELLIGENCE IN THE
AGRICULTURE SYSTEM.

SUBMITTED BY: SUBMITTED TO:

ANKA KUMARI LIMBU DIKSHANT KAFLE

ROLL NO:7 Asst. Prof. OF GAASC

8TH SEM DEPARTMENT OF ECONOMICS

GAASC, BAITADI
 ACKNOWLEDGEMENTS
I would like to express my sincere gratitude and appreciation to all those who have contributed to
the successful completion of this report. First and foremost, I would like to thank Mr. Dikshant
Kafle, Assistant Professor of GAASC, for providing a grand opportunity to write this report and
for valuable guidance and support throughout the entire process. Secondly, I would like to thank
my dear friend Arati Dhami, for helping me while reviewing various papers during report
writing. This report would not have been possible without the collective efforts and support of all
those mentioned above. Thank you for your invaluable contributions.

TABLE OF CONTENTS

Table of Contents
TITL
E.......................................................................................................................................................0
ACKNOWLEDGEMENTS...........................................................................................................................1
TABLE OF CONTENTS...............................................................................................................................1
ABSTRACT...................................................................................................................................................2
1. INTRODUCTION.................................................................................................................................2
2. MATERIALS AND METHODOLOGY:...............................................................................................5
3. RESULT AND DISCUSSION...............................................................................................................5
APPLICATION OF MACHINE LEARNING...............................................................................................5
3.1 Soil Properties and weather prediction................................................................................................5
3.2 Crop yield prediction...........................................................................................................................7
3.3 DISEASE AND WEED DETECTION................................................................................................9
3.4 Livestock Production and Management.............................................................................................10
3.5 Harvesting..........................................................................................................................................12
Application of A. I in Agriculture............................................................................................................13
4. CONCLUSION....................................................................................................................................15
5. REFERENCES:...................................................................................................................................15
ABSTRACT
Agriculture is a vital business with a rapidly expanding population, leading to the development
of new automated methods to address food and employment needs. The Internet of Things, Big
Data and Analytics, Artificial Intelligence (AI), and Machine Learning (ML) are transforming
agriculture by increasing product quality and quantity through smart farming. AI plays a
significant role in daily life, with applications in seed dealers, pest control companies, and supply
chain systems. ML algorithms are used to solve complex problems, such as weather prediction,
spam filtering, plant disease identification, and pattern recognition. ML has shown promising
results in crop yield prediction, crop yield prediction, and weather forecasting. ELM-based
regression models have been used to estimate soil surface humidity, nutrient solutions, and soil
moisture using data from MODIS. ML algorithms have also been used to forecast rainfall in
Tenerife, Spain, based on atmospheric synoptic patterns. Crop yield prediction is crucial for
farmers, and factors like pH value, soil type, and weather patterns significantly impact crop
production. ML models have been used to identify yield gap hotspots in wheat production,
predict rice crop yield, and predict hybrid maize yield using deep neural networks (DNN). AI
technology aids in agriculture by predicting the weather, land quality, disease detection,
monitoring crop health, and reducing crop loss. It revolutionizes agriculture by enabling farmers
to produce more with less work, supporting traditional farming methods, and improving yield
and fertilizer preparation. AI can also help increase agricultural yield and productivity by
assisting farmers with soil monitoring, pest control, healthier crops, supply chain management,
and farm data analysis. This report has reviewed the applications of machine learning and
artificial intelligence in the agriculture sector such as Disease identification and monitoring,
weed identification, livestock production and management, Crop yield, soil properties as well as
weather predictions.

APPLICATION OF MACHINE LEARNING AND ARTIFICIAL

INTELLIGENCE IN AGRICULTURE SYSTEM

1. INTRODUCTION
Agriculture is one of the world's oldest and most vital businesses. The world's population is
quickly expanding, increasing the demand for food as well as employment. Because farmers' old
techniques are unable to satisfy these requirements while providing billions of people worldwide
employment opportunities, new automated methods are being developed to address the world's
food needs (Zhang et al., 2021). Due to workforce scarcity, harsher laws, an expanding global
population, and a decline in farmers, farmers are compelled to look for new alternatives. The
Internet of Things, Big Data & Analytics, Artificial Intelligence (AI), and Machine Learning
(ML) are making inroads into practically every business. Efforts and research are underway to
increase agricultural product quality and quantity by making them "connected" and "intelligent"
via "smart farming." (2018) Dharmaraj et al. In 1955 at the Dartmouth Conference, John
McCarthy proposed a study based on the idea that "every aspect of learning or any other feature
of intelligence can, in principle, be so precisely described that a machine can be made to simulate
it" (McCarthy et al., 1995). This is when the term "Artificial Intelligence" was first used.
According to the European Commission's Independent High-level Expert Group on Artificial
Intelligence (AI HLEG, 2018), "Artificial intelligence (AI) refers to systems designed by humans
that, given a complex goal, act in the physical or digital world by perceiving their environment,
interpreting the collected structured or unstructured data, reasoning on the knowledge derived
from this data, and deciding the best action(s) to take (according to pre-defined parameters) to
achieve the given goal. ML approaches often incorporate a learning process to learn from
"experience" (training data) to complete a job. In ML, the input is a collection of examples. An
individual example is often characterized by a set of qualities, sometimes known as features or
variables. Feature types include nominal (enumeration), binary (i.e., 0 or 1), ordinal (such as A+
or B), and numeric (integer, real number, etc.). Liakos et al. (2018), writers analyzed several ML
applications in agriculture and talked about how the industry will profit from digital technology.
With the power to expand our perceptions and change the environment around us, artificial
intelligence (AI) has started to play a significant part in daily life (Kundalia et al., 2020; Gandhi
et al., 2020; Ahir et al., 2020). Plessen (2019) provided a system for harvest planning that relies
on the integration of crop assignment and truck routing. The fundamental idea behind artificial
intelligence is to create technology that works similarly to the human brain (Parekh et al., 2020;
Jani et al., 2019). The 2019 study by Jha et al. encloses the associated interactions between
various embedded systems and AI technology compatible with the agricultural area, and it
presented a short about the numerous uses of neural networks, ML, in this sector for precision
farming (Yang et al.,2007). Computers may learn without being explicitly programmed thanks to
a branch of computer science called machine learning (Arthur Samuel, 1959). In 1950, Alan
Turing published a research paper titled "The Turing Test for Machine Intelligence" in which he
put out the idea of learning machines. He tested the machine's capacity to exhibit intelligent
behavior akin to that of people. Today, thanks to the availability of novel algorithms and vast
data sets via internet resources, enterprises, and academic groups are widely employing ML
algorithms to solve a wide range of issues. Machine learning is used in the agriculture industry to
improve crop yield and quality. Seed dealers utilize this agriculture technology to generate data
to build better crops. They are used by pest control companies to identify various germs, bugs,
and animals. An agriculture supply chain system is made up of businesses or cooperatives that
are in charge of growing and distributing products made from plants, animals, grains, pulses, and
other items. In a variety of supply networks, agricultural products are used as raw materials to
make higher-value consumer goods (Zhou et al., 2021). The ML process is divided into three key
parts, i.e. data input, model building, and generalization as shown in Figure 2. Generalization is
the process of predicting the output for the inputs with which the algorithm has not been trained
before. ML algorithms are mainly used to solve complex problems where human expertise fails
such as weather prediction, spam filtering, disease identification in plants, and pattern
recognition (Sharma et al., 2020).
 Fuzy logic

Swarm expert
intelligence systems

Artificial
intelligence

Natural
Machine
language
learning
processing

Computer r
vision

Fig1. Artificial Intelligence Techniques

Model building/
Data pre- machine
Raw data set Generatization
processing learning
algorithm

Fig2. A Machine learning process

Table no.1 Nomenclature

AI Artificial Intelligence
ML Machine Learning
DL Deep Learning
NLP Natural Language Processing
ANN Artificial Neural Network
ANFIS Adaptive Neuro-Fuzzy Inference System
MLP Multi-Layer Perceptron
NN Neural Network
kNN K-Nearest Neighbor
SVM Support Vector Machines
RF Random forest
 CDL Cropland Data Layer
GBM Gradient Boosting Model
GRNN Generalized Regression Neural Networks
BPNN Back Propagation Neural Network
RMSE Root Mean Square Error
SOM Soil Organic Matter
CEC Cation Exchange Capacity
MAE Mean Absolute Error
Mg Magnesium
K Potassium

2. MATERIALS AND METHODOLOGY:

To prepare this report, different articles and journals from the ResearchGate site, google scholar,
and the Internet were reviewed and secondary data were gathered from these websites and the
Internet.
Secondary source of data: Research papers, articles, journals, internet, etc.

3. RESULT AND DISCUSSION

While reviewing the different secondary sources many results were found. The machine learning
and A. I have various applications in the agriculture system. Various applications were found
from the different literature reviews of secondary sources. Some of them are as follows:

APPLICATION OF MACHINE LEARNING

3.1 Soil Properties and weather prediction
Soil properties prediction is crucial for crop selection, land preparation, seed selection, yield, and
fertilizer/manure selection. It involves predicting soil nutrients, surface humidity, and weather
conditions during the crop's lifecycle. Acar et al. (2019) used an extreme learning machine
(ELM)-based regression model to estimate soil surface humidity. Wang et al. (2018) used soft
sensors based on ELM to measure the composition of nutrient solutions in soilless farming. An
emerging planting technique is a soilless cultivation. It is critical to monitor pH, temperature, and
concentration variations in nutrient solution composition since the effectiveness of soilless
cultivation is largely reliant on these factors. Park et al. (2015) utilized ML algorithms to predict
soil moisture using data from MODIS. The authors used random forest (RF) and Cubist
algorithms to downscale the AMSR2 soil moisture to 1KM. To gather soil moisture data, an
ensemble of these methods was applied. The findings acquired using ML approaches were
compared to those obtained using the statistical ordinary least squares methodology. The ML
model had an R 2 (coefficient of determination) of 0.96 and an RMSE of 0.06, whereas the
statistical ordinary least squares model had an R 2 of 0.47 and an RMSE of 0.16. Khanal et al.
 (2018) developed an alternate strategy that uses remotely sensed aerial pictures of agricultural
land to produce the dataset for predicting soil attributes and crop output. To anticipate day-wise
mean air temperature at field level for 90 days, Ju-Young et al. (2020) studied a seasonal climatic
forecasting model employing regularized ELM. In the Chinese Loess Plateau, Feng et al. (2019)
measure soil temperature at various soil depths. To forecast the soil temperature, four distinct
machine learning (ML) algorithms—ELM, GRNN, BPNN, and RF—were examined. The
simulations show that ELM outperforms other ML algorithms in terms of RMSE, MAE, Nash-
Sutcliffe coefficient (NS), and concordance correlation coefficient. ML algorithms were trained
with air temperature, wind speed, relative humidity, vapor pressure, and solar radiation as input
parameters. Different ML algorithms were used by Sierra and Jesus (2020) to forecast rainfall in
Tenerife, a Spanish island, based on atmospheric synoptic patterns, and they discovered that NN
performed the best of the ML methods. Table 2 presents a comparative study of different ML
algorithms for the prediction of soil properties and weather prediction.

Table no.2 Different ML algorithms for prediction of soil properties and weather conditions

Attributes Experimental place ML algorithm Accuracy measures References

Soil Surface Polarimetric Extreme learning RMSE of 2.19% Acar et
Humidity measurements using Machine based al.,2019
Radarsat-2 for a field Regression
located at Dicle (ELMR) model

𝑆𝑂4 2−,𝐻2𝑃𝑂4 − Auxiliary

University

𝑆𝑂4
ELM RMSE of 1.2414 Wang et
measurement of for 2− al.,2018
conductivity, prediction and

𝐻2𝑃𝑂4
temperature, pH, and 0.8892 for
flow rate −
prediction

𝑅 2 of 0.96
Soil moisture AMSR2 soil moisture RF and Cubist RMSE of 0.06 and Park et
data from MODIS algorithm al.,2015
(satellite)
SOM, Mg, CEC, K, The northwestern part RF, SVM, Cubist, NN best predicts Khanal et
pH of Madison County, NN, GBM SOM with R2 of al.,2018
Ohio, USA 0.64, RMSE of 0.44
and CEC with R2
of 0.67, RMSE of
2.35; SVM for K
with R 2 of 0.21,
RMSE of 0.49 and
Mg with R 2 of
0.22, RMSE of
4.57); and GBM for
pH with R2 of 0.15,
RMSE of 0.62
Mean air Korea climatic model Regularized RMSE in the range Ju-Young
temperature from GloSea5GC2 Extreme Learning of 1.02 to 3.35 was et al.,2020
Machine (RELM) reported
Soil temperature Maize field, located ELM, GRNN, ELM best predicts Feng et
in Shouyang County, BPNN, and RF with RMSE of al.,2019
Shanxi Province, 2.26-2.95oC, MAE
northern P. R. China of 1.76–2.26oC, NS
of 0.856–0.930, and
concordance
correlation
coefficient of
0.925–0.965
Rainfall Prediction Tenerife Spain SVM, kNN, RF, NN best predicts Sierra and
K-means the rainfall Jesus 2020
clustering, NN occurrence and
intensity with an
average F score
close to 0.4 and R
0.1-0.8 depending
on the month and
gauge analyzed

3.2 Crop yield prediction

The estimation of crop production and suggestions for boosting it are important pieces of
knowledge for each farmer. In estimating crop production, factors such as pH value, soil type and
quality, and weather patterns such as temperature, rainfall, humidity, sunlight hours, fertilizers,
and harvesting schedules all have a significant impact (Liakos et al., 2018). Kamir et al. (2020)
used ML models to identify the yield gap hotspots in wheat production. The authors generated
very high-resolution yield maps using data from various sources between 2009 and 2015.
Kulkarni et al. (2018) utilized DL models to predict rice crop yield. The authors utilized soil
properties and nutrient measurements recorded over 31 years and historic rainfall data. The input
data was fed to recurrent neural network models for crop yield prediction. For effective
prediction, the authors explored different activation functions viz. sigmoid, reLu, and linear in
the neural network. Khaki and Wang (2019) predicted the hybrid maize yield with a dataset of
2,267 locations in the United States and Canada between the years 2008 to 2016 using deep
neural networks (DNN). Genotype, weather, and soil properties were the three components used
to train DNN. The proposed model accurately predicts the maize yield with RMSE of 12% of the
average yield for the predicted weather dataset and 11% of the average yield for the perfect
weather dataset and outperforms LASSO, shallow neural network (SNN), and regression tree
(RT). Leroux et al. (2019) explored the ML algorithm for predicting the maize yield in Burkina
Faso with a remotely sensed dataset. A process-based crop model SARRO which is designed to

outperforms MLR in maize yield prediction with 𝑅 2 of 0.59 at the end of the season and 0.49
simulate attainable agricultural yields under tropical conditions is used in this study. RF

before two months of harvest. Maimaitijiang et al. (2020) explored the potential of UAV with
 DNN for soybean yield prediction from the fields of Columbia, Missouri, USA. Multi-modal
information such as canopy spectral, structural, and thermal features extracted from images
obtained from the sensors installed on UAV is used as the input dataset for training DNN. The

square regression (PLSR), RF, and SVR algorithms with 𝑅2 of 0.720 and RMSE of 15.9%.
simulation result shows that DNN accurately predicts the crop yield and outperforms partial least

Table 3 presents a comparative study of different ML algorithms for crop yield prediction.

Table No. 3 Different ML Algorithms for Crop Yield Prediction

Attributes Experimental ML algorithm Accuracy measures References

place
Determine yield High-resolution Ensemble-based Support vector regression Kamir et al.
for wheat crop yield maps learners using 9 with radial basis function [2020]
using: weather ML algorithms displayed the lowest RMSE
data, NDVI of 0.59 and R2 score of
time-series data, 0.73
yield data from
harvesting
equipment, crop
type, and
geolocation data
Determine crop Historic soil Recurrent Neural RMSE of 41.497 and MAE Kulkarni et al.
yield for rice properties and Networks of 41.6 [2018]
crops rainfall data (31
years period)
Hybrid maize U.S. and Canada Deep NN, Shallow Deep NN best predicts with Khaki and
yield prediction NN, Regression RMSE of 12% of average Wang [2019]
Tree yield and 50% of the
standard deviation for the
predicted weather dataset.
RMSE reduced to 11% of
the average yield and 46%
of the standard deviation for
the perfect weather dataset
Maize and cotton Burkina Faso, RF, MLR RF best predicts (an R2 of Leroux et al.
yield prediction West Africa USA 0.59) at the end of the (2019)
season and an R2 of 0.49
approximately two months
before harvest
Soybean yield RGB, Partial Least DNN-F2 best predicts with Maimaitijiang
prediction multispectral, Squares Regression an R2 of 0.720 and an et al. (2020)
and thermal (PLSR), RF, SVR, RMSE of 15.9%
images were input-level feature
collected through fusion-based DNN
UAV in (DNN-F1), and
 Columbia, intermediate-level
Missouri, USA feature fusion-
based DNN (DNN-
F2)

3.3 DISEASE AND WEED DETECTION

Diseases, microorganisms, and bacteria damage plants, affecting crop yield and causing
economic losses. Farmers use pesticides to combat diseases, but excessive use harms the
water and soil cycle. AI systems can reduce disease risk, minimize economic impact, and
minimize environmental damage during crop growth. Researchers have used CNN-based
models to detect diseases in cassava crops, Ramcharan et al. (2017), Ferreira et al.
(2017), Waheed et al. (2020), Olsen et al. (2019), Sethy et al. (2020), and Kerkech et al.
(2020) to identify diseases and weeds in various crops. These models have shown high
accuracy in detecting diseases and weeds, outperforming other CNN models in terms of
parameters, accuracy, computation complexity, and computation time. Olsen et al. (2019)
explored robust deep-learning models Inception-v3 and ResNet-50 for weed species
identification and classification from a dataset of images collected in Australian
rangeland. Sethy et al. (2020) identified rice leaf disease using a hybrid CNN and SVM,
with resnet50 with the SVM classification model best classifying with an F1 score of
0.9838. Kerkech et al. (2020) identified vine diseases from visible and infrared UAV
images, with a CNN model identifying with an accuracy of 92% at the grapevine level
and 87% at the leaf level. Overall, AI systems can help reduce disease risk, minimize
economic impact, and minimize environmental damage during crop growth. Table 4
presents a comparative study of different ML algorithms for disease and weed
identification

Table No. 4 Different ML Algorithms for Disease and Weed Identification

Attributes Experimental place ML Accuracy References

algorithm measures
Disease 10,000 labeled high-resolution CNN-based Best case Sambasivan
detection in image databases collected from DL model accuracy is and Opiyo
cassava Uganda 99.30% (2020)
Crops
Disease and Digital image dataset collected CNN Best case Ramcharan
pest from Tanzania Inception accuracy is et al.
detection in v3 based 93.0% (2017)
cassava crops on
GoogLeNet
Weed A dataset of high-resolution field CNN based Accuracy is Ferreira et
detection in images collected through a CaffeNet higher than al. (2017)
Soybean phantom DJI3 drone in Brazil 98% for the
crops classification
of all
classes.
Corn Leaf A dataset of 12,332 images of 250 CNN An accuracy Waheed et
disease X250 dimension is collected from (DenseNet) of 98.06% is al. (2020)
identification different sources Dense achieved in
and recognizing
Classification three corn
disease
Weed A dataset of 17,509 labeled images Two deep Average Olsen et al.
identification is obtained of eight weed species learning classification (2019)
across 8 locations in northern models accuracy of
Australia Inception- 95.1% and
v3 and 95.7% is
ResNet-50 obtained for
Inception-v3
and ResNet-
50
Rice leaf A dataset of 5932 diseased rice Deep The F1 score Sethy et al.
disease leaf images is obtained from fields features of the (2020)
identification of western Odisha and database ResNet50 proposed
http://bcch.ahnw.gov.cn/Right.asp plus SVM model is
x model 0.9838
Vine disease A dataset of visible and infrared CNN Accuracy of Kerkech et
detection UAV images obtained in the 92% at the al. (2020)
Center Val de Loire region in grapevine
France level and
87% at the
leaf level is
achieved

3.4 Livestock Production and Management

Livestock production involves the management and consumption of cattle for human
consumption. The optimization of cattle's health, food, nutrition, and behavior is based on
farming parameters, such as health, food, nutrition, and behavior. Artificial intelligence,
IoT, and Blockchain technologies are being explored to improve livestock sustainability
and analyze their behavior, such as chewing, eating, and movement patterns. ML
techniques have been used to classify cattle behavior, analyzing events like oestrus and
dietary changes for well-nutrition. Durtta et al. (2015) used machine learning techniques
to classify cattle behavior using collar-based sensors, analyzing events like oestrus and
dietary changes for well-nutrition. Ebrahimie et al. (2019) proposed an ML predictive
model for estimating Sub-Clinical Mastitis (SCM) from milking parameters in dairy
 herds, using decision trees, stump decision trees, parallel decision trees, and random
forests. Hansen et al. (2018) used deep learning techniques to identify pig faces from
digital images, achieving an accuracy of 96.7%. Fenlon et al. (2016) analyzed calving
difficulties in Irish dairy herds using ML algorithms, finding that logistic regression
outperforms RF, decision trees, and Naive Bayes in predicting conception using artificial
insemination. Table 5 presents a comparative study of different ML algorithms for
livestock production and management.

Table No. 5 Different ML Algorithms for Livestock Production and Management

Attributes Experimental ML algorithm Accuracy References

place measures
Cattle behavior The dataset is Binary Tree, Bagging Durtta et al.
classification collected from Linear ensemble (2015)
the Tasmanian Discriminant classification
Institute of Analysis, Naive with Tree learner
Agriculture Bayes, k-NN, best classify
Dairy Research ANFIS classifier with an accuracy
Facility at Elliott of 96%, a
sensitivity of
97%, specificity
of 89%, an F1
score of 89%,
and a false
discovery rate of
9%
Prediction of Data is collected DL, Naïve GBT best Ebrahimie et al.
subclinical from a Bayes, predicts with an (2019)
bovine mastitis commercial New Generalized accuracy of
Zealand dairy Linear Model, 84.9%
farm in Logistic
Ongaonga, Regression,
Hawkes Bay, Decision Tree,
from July 2011 GBT, and RF
to June 2013 and
then it is
transformed by
ZStandardization
Identification of A dataset of Conventional Pigs’ faces are Hansen et al.
livestock such as digital images of neural networks recognized with (2018)
pigs and cows pig faces is an overall
created accuracy of
96.7%
Prediction of Dataset of 2,723 Logistic Logistic Fenlon et al.
insemination artificial regression, regression best (2016)
 outcome in Irish insemination Naive Bayes, predicts with a
dairy cows records from Decision Tree, precision of
Irish research and RF 57.89%, Recall
farms and 4,205 of 44.82%,
breeding events Fscore of
from 50.52%, and
commercial Matthews
dairy farms correlation
coefficient of
0.16

3.5 Harvesting
Smart harvesting systems are a growing approach that reduces human labor and improves crop
yields. These systems utilize technologies like smart sensors, robotics, UAVs, IoT devices, AI,
and ML-based computer vision techniques to harvest crops intelligently. Over the years, various
robots have been developed for harvesting fruits and vegetables, offering better insights into
crops and increased productivity. Smart harvesting systems have numerous advantages over
traditional harvesting methods, such as reduced labor, optimized crop yield, maximum
probability, better crop insight, reduced costs of harvesting, and cost-efficient production. One
significant issue in the Japanese agriculture industry is labor shortages. Sapasaki et al. (2013)
utilized machine vision for asparagus robot harvesting in Nagasaki prefecture. Horng et al.
(2019) proposed a smart harvesting system using IoT and image recognition systems to detect
mature crops using object detection features trained on an MLP neural network. Zhang et al.
(2020) explored Regions-CNN (RCNN) for multi-class canopy object detection in the shake-
and-catch apple harvesting system, using RGB images and pre-trained RCNN for real-time
detection of apples, branches, and trunks. Zhang et al. (2020) investigated eleven canopy
parameters using principal component analysis (PCA) and classified the removal status of apples
into mechanically harvested and mechanically unharvested. Pise and Upadhye (2018) explored
Naive Bayes and SVM ML techniques for grading harvested mangoes based on color, size,
features, quality, and maturity. Table 6 presents a comparative study of different ML algorithms
for intelligent harvesting.

Table No. 6 Different ML Algorithms for Intelligent Harvesting

Attributes Experimental ML algorithm Accuracy References

place measures
Intelligent Dataset of digital MLP, CNN The object Horng et al.
harvesting images from the detection model (2019)
system using IoT agricultural land has a mean
and smart image average
recognition precision of 84%
and the arm
 movement
prediction model
has a mean
picking accuracy
of 89%
Multi-class Dataset of color CNN based Average Zhang et al.
canopy object images and AlexNet, Precision of (2020)
detection and corresponding VGG16, and 82.4%, average
estimation of 3D point cloud VGG19 VGG19 computation
shaking location data of apple best predicts time is 0.45s per
in apple trees were with mean image
harvesting acquired in
system commercial
orchards
Determination of The dataset is k-NN, PCA
In training Zhang et al.
key canopy collected from prediction (2020)
parameters for the two-year accuracies are
mass mechanical field trials in two 76-92% and 62-
apple harvesting commercial 74% for
apple orchards "Scifresh" and
"Envy" in the
testing dataset,
the overall test
accuracies were
81-91% on
"Scifresh" but
only 36-79% on
"Envy"
Grading of Dataset of Naive Byes and Results are more Pise and
harvested mango images SVM accurate Upadhye (2018)
mangoes based
on their quality
and maturity

Application of A. I in agriculture

Table no 7 Artificial intelligence applications in Agriculture

S.N. Application Description References

1 Prediction of AI technology aids in agriculture by (Rizvi et al., 2021), (Singh and
weather predicting the weather, land quality, Kaur, 2022), (M.
and disease detection, monitoring Barenkamp,2020), (Tzachor et
crop health through soil and plant al., 2022) and (Al-bayati and
sensors, and reducing crop loss Üstündag, 2020)
through complex algorithms.
2 Allowing AI revolutionizes agriculture by (Sujata et al., 2021), (Alam et
farmers to enabling farmers to produce more al.,2020), (Mumtaz and Nazar,
achieve with less work, supporting traditional 2022)
efficient farming methods through cognitive
results IoT, growth-driven crops, and data
analysis for improved yield and
fertilizer preparation.
3 Self-driving AI, self-driving tractors, and IoT can (Songol et al., 2021), (Jiayu et
tractors address labor shortages in farming by al.,2015), (Sharma et al.,2022),
providing more precise, cost- and (Santangeli et al., 2020)
effective, and accurate technologies.
Robotics, like harvesting fruits and
vegetables, offer farmworkers more
accuracy, reduced errors, and
extended work periods. AI
advancements can transform
agriculture by providing data insight
and predictive analytics, enabling
businesses to make better decisions
and navigate complex environments.
4 Crop and soil AI is used in crop and soil (Qazi et al., 2022), (Lowe et
monitoring monitoring, using drones, IoT, and al., 2022), (Weng et al., 2019),
satellite images for data collection (Shelake et.al 2021) and
and analysis. Mobile applications (Marcu et al., 2019)
help farmers assess crop data,
manage operations, and optimize
herbicide and pesticide use, ensuring
food safety and reducing pest attacks.
5 Identifying AI can help farm workers plant crops (Sane TU,2021), (Bharti and
locations for based on geographical features, soil Bhan, 2018), (Quinn et al.,
sowing composition, and other criteria. It can 2014), and (Fuentes et al.,
specific crops also analyze soil flaws and nutrient 2020)
deficits, control weeds, and automate
complex processes.
6 Increase AI techniques in agriculture increase (Sujawat, 2021), (Kushkhova
agricultural yield and productivity by assisting et al., 2019), (Kun W. 2020),
yield and farmers with soil monitoring, pest (Abreu and Deventer, 2021).
productivity control, healthier crops, supply chain
management, and farm data analysis.
However, ML can reduce yield
quality and increase production costs.
7 Helps to solve The agriculture industry is using AI (Mir et al., 2021), (Katyayan et
the problem to address food shortages and waste al. 2021), (Garrett et al., 2022),
of food caused by locust swarms, climate and (Hyunjin and Sainan,
shortage change, droughts, and floods. AI
 recommends crops and seeds based 2021)
on factors like soil conditions,
weather forecasts, and market trends.
8 Detect the AI algorithms detect insects, alert (Sarkar et al., 2022), (Patil and
smallest insects farmers to pests, and improve crop Kumar, 2020), (Upadhyay and
health and yields. AI-powered Gupta, 2021), (Singh and Jain,
farming approaches respond 2022) and (Ashima et al.,
intelligently to climate change, 2021)
reduce labor challenges, and speed
up food transformation. AI startups
monitor crop health using weather
and satellite images and develop real-
time models for crop health and price
forecasting.

4. CONCLUSION
AI-enabled technologies use data from temperature, precipitation, wind speed, and solar
radiation to predict weather conditions, analyze crop sustainability, and evaluate farms for
diseases and pests. Farmers with Wi-Fi connectivity can receive AI-tailored farm plans,
enhancing output and income without reducing natural resources. AI-assisted farming practices,
automation, and yield management improve net output, addressing issues like climate change
and insect infestations. IoT-enabled smart sensors, actuators, satellite images, robots, and drones
have significantly boosted the agriculture industry, collecting real-time data and making
decisions without human intervention. Artificial intelligence, which automates intelligent
behavior, benefits the planet and humans in various aspects of life. This paper reviews machine
learning and applications for agriculture. Various papers were reviewed about the applications of
machine learning and artificial intelligence in the agriculture sector such as Disease identification
and monitoring, weed identification, livestock production and management, Crop yield, soil
properties as well as weather predictions.
5. REFERENCES:
Alam, M. A., Ahad, A., Zafar, S., & Tripathi, G. (2020). A neoteric smart and sustainable
farming environment incorporating a blockchain‐based artificial intelligence
approach. Cryptocurrencies and Blockchain Technology Applications, 197-213.

Al-bayati, J. S. H., & Üstündağ, B. B. (2020, October). Artificial intelligence in smart

agriculture: Modified evolutionary optimization approach for plant disease identification.
In 2020 4th International Symposium on Multidisciplinary Studies and Innovative
Technologies (ISMSIT) (pp. 1-6). IEEE.

Ashima, R., Haleem, A., Bahl, S., Javaid, M., Mahla, S. K., & Singh, S. (2021). Automation
and manufacturing of smart materials in Additive Manufacturing technologies using the
Internet of Things towards the adoption of Industry 4.0. Materials Today: Proceedings, 45,
5081-5088.

Barenkamp, M. (2020, June). A new IoT gateway for artificial intelligence in agriculture.
In 2020 International Conference on Electrical, Communication, and Computer
Engineering (ICECCE) (pp. 1-5). IEEE.

Bharti, V. K., & Bhan, S. (2018). Impact of artificial intelligence for agricultural
sustainability. Journal of Soil and Water Conservation, 17(4), 393-399.

Boursianis, A. D., Papadopoulou, M. S., Diamantoulakis, P., Liopa-Tsakalidi, A.,

Barouchas, P., Salahas, G., ... & Goudos, S. K. (2022). Internet of things (IoT) and
agricultural unmanned aerial vehicles (UAVs) in smart farming: A comprehensive
review. Internet of Things, 18, 100187.

De Abreu, C. L., & van Deventer, J. P. (2022). The application of artificial intelligence (AI)
and Internet of things (IoT) in agriculture: A systematic literature review. In Southern
African Conference for Artificial Intelligence Research (pp. 32-46). Springer, Cham.

Dharmaraj, V., & Vijayanand, C. (2018). Artificial intelligence (AI) in

agriculture. International Journal of Current Microbiology and Applied Sciences, 7(12),
2122-2128.

Diez-Sierra, J., & Del Jesus, M. (2020). Long-term rainfall prediction using atmospheric
synoptic patterns in semi-arid climates with statistical and machine learning
methods. Journal of Hydrology, 586, 124789.
dos Santos Ferreira, A., Freitas, D. M., da Silva, G. G., Pistori, H., & Folhes, M. T. (2017).
Weed detection in soybean crops using ConvNets. Computers and Electronics in
Agriculture, 143, 314-324.

Ebrahimi, M., Mohammadi-Dehcheshmeh, M., Ebrahimie, E., & Petrovski, K. R. (2019).

Comprehensive analysis of machine learning models for prediction of sub-clinical mastitis:
Deep Learning and Gradient-Boosted Trees outperform other models. Computers in
biology and medicine, 114, 103456.

Fadziso, T. (2019). Implementation of artificial intelligence in agriculture: a review for cms

optimization. Malaysian Journal of Medical and Biological Research, 6(2), 127-134.

Feng, Y., Cui, N., Hao, W., Gao, L., & Gong, D. (2019). Estimation of soil temperature
from meteorological data using different machine learning models. Geoderma, 338, 67-77.

Fenlon, C., O'Grady, L., Dunnion, J., Shalloo, L., Butler, S., & Doherty, M. L. (2016). A
comparison of machine learning techniques for predicting insemination outcome in Irish
dairy cows.

Fuentes, S., Gonzalez Viejo, C., Cullen, B., Tongson, E., Chauhan, S. S., & Dunshea, F. R.
(2020). Artificial intelligence applied to a robotic dairy farm to model milk productivity
and quality based on cow data and daily environmental parameters. Sensors, 20(10), 2975.

Garrett, K. A., Bebber, D. P., Etherton, B. A., Gold, K. M., Plex Sulá, A. I., & Selvaraj, M.
G. (2022). Climate change effects on pathogen emergence: Artificial intelligence to
translate big data for mitigation. Annual Review of Phytopathology, 60, 357-378.

Hansen, M. F., Smith, M. L., Smith, L. N., Salter, M. G., Baxter, E. M., Farish, M., &
Grieve, B. (2018). Towards on-farm pig face recognition using convolutional neural
networks. Computers in Industry, 98, 145-152.

Horng, G. J., Liu, M. X., & Chen, C. C. (2019). The smart image recognition mechanism for
crop harvesting system in intelligent agriculture. IEEE Sensors Journal, 20(5), 2766-2781.

Hyunjin, C., & Sainan, H. (2021). A study on the design and operation method of plant
factories using artificial intelligence. Nanotechnology for Environmental
Engineering, 6(3), 41.

Jiayu, Z., Shiwei, X., Zhemin, L., Wei, C., & Dongjie, W. (2015, May). Application of
intelligence information fusion technology in agriculture monitoring and early-warning
research. In 2015 International Conference on Control, Automation and Robotics (pp. 114-
117). IEEE.
Kamir, E., Waldner, F., & Hochman, Z. (2020). Estimating wheat yields in Australia using
climate records, satellite image time series, and machine learning methods. ISPRS Journal
of Photogrammetry and Remote Sensing, 160, 124-135.

Katyayan, A., Mashelkar, S., DC, A. G., & Morajkar, S. (2021, December). Design of smart
agriculture systems using artificial intelligence and big data analytics. In 2021 3rd
International Conference on Advances in Computing, Communication Control and
Networking (ICAC3N) (pp. 592-595). IEEE.

Kerkech, M., Hafiane, A., & Canals, R. (2020). Vine disease detection in UAV multispectral
images using optimized image registration and deep learning segmentation
approach. Computers and Electronics in Agriculture, 174, 105446.

Khaki, S., & Wang, L. (2019). Crop yield prediction using deep neural networks. Frontiers
in plant science, 10, 621.

Khanal, S., Fulton, J., Klopfenstein, A., Douridas, N., & Shearer, S. (2018). Integration of
high-resolution remotely sensed data and machine learning techniques for spatial
prediction of soil properties and corn yield. Computers and electronics in agriculture, 153,
213-225.

Kondo, N., Monta, M., & Noguchi, N. (2004). Agri-robot (I)-fundamentals and
theory. Chinese translation from the Japanese language) Corona Publishing Co., Ltd, 56-
58.

Kulkarni, S., Mandal, S. N., Sharma, G. S., & Mundada, M. R. (2018, September).
Predictive analysis to improve crop yield using a neural network model. In the 2018
international conference on advances in computing, communications, and informatics
(ICACCI) (pp. 74-79). IEEE.

Kun, W. (2020). Design of multi-parameter monitoring system for intelligent agriculture

greenhouse based on artificial intelligence. In Multimedia Technology and Enhanced
Learning: Second EAI International Conference, ICMTEL 2020, Leicester, UK, April 10-
11, 2020, Proceedings, Part I 2 (pp. 269-280). Springer International Publishing.

Kushkhova, B. A., Gazaeva, M. S., Gyatov, A. V., Ivanova, Z. M., & Eneeva, M. N. (2019,
August). Artificial intelligence in agriculture of Kabardino-Balkaria: current state,
problems, and prospects. In IOP Conference Series: Earth and Environmental
Science (Vol. 315, No. 2, p. 022013). IOP Publishing.
Leroux, L., Castets, M., Baron, C., Escorihuela, M. J., Bégué, A., & Seen, D. L. (2019).
Maize yield estimation in West Africa from crop process-induced combinations of multi-
domain remote sensing indices. European Journal of Agronomy, 108, 11-26.

Liakos, K. G., & PatriziaBusato, D. Simon Pearson, and DionysisBochtis.(2018).". Machine

learning in agriculture: A review." Sensors, 18.

Liakos, K. G., Busato, P., Moshou, D., Pearson, S., & Bochtis, D. (2018). Machine learning
in agriculture: A review. Sensors, 18(8), 2674.

Lowe, M., Qin, R., & Mao, X. (2022). A review on machine learning, artificial intelligence,
and smart technology in water treatment and monitoring. Water, 14(9), 1384.

Maimaitijiang, M., Sagan, V., Sidike, P., Hartling, S., Esposito, F., & Fritschi, F. B. (2020).
Soybean yield prediction from UAV using multimodal data fusion and deep
learning. Remote sensing of environment, 237, 111599.

Maraveas, C., Loukatos, D., Bartzanas, T., & Arvanitis, K. G. (2021). Applications of
artificial intelligence in fire safety of agricultural structures. Applied Sciences, 11(16),
7716.

Marcu, I. M., Suciu, G., Balaceanu, C. M., & Banaru, A. (2019, June). IoT-based system for
smart agriculture. In 2019 11th International Conference on Electronics, Computers and
Artificial Intelligence (ECAI) (pp. 1-4). IEEE.

McCarthy, J., Minsky, M. L., Rochester, N., & Shannon, C. E. (2006). A proposal for the
Dartmouth summer research project on artificial intelligence, August 31, 1955. AI
magazine, 27(4), 12-12.

Mir, H., Zaraatgari, R., & Sotoudeh, R. (2021). Improving the food and agriculture sector
Tehran stock exchange by using artificial intelligence. Agricultural Marketing and
Commercialization Journal, 5(2), 90-114.

Mumtaz, N., & Nazar, M. (2022). Artificial intelligence robotics in agriculture: See &
spray. Journal of Intelligent Pervasive and Soft Computing, 1(01), 21-24.

Olsen, A., Konovalov, D. A., Philippa, B., Ridd, P., Wood, J. C., Johns, J., ... & White, R.
D. (2019). DeepWeeds: A multiclass weed species image dataset for deep
learning. Scientific reports, 9(1), 2058.
Park, S., Im, J., Park, S., & Rhee, J. (2015, July). AMSR2 soil moisture downscaling using
multisensor products through a machine learning approach. In 2015 IEEE International
Geoscience and Remote Sensing Symposium (IGARSS) (pp. 1984-1987). IEEE.

Patil, R. R., & Kumar, S. (2020). A Bibliometric Survey on the Diagnosis of Plant Leaf
Diseases using Artificial Intelligence. Library Philosophy and Practice, 1-26.

Pise, D., & Upadhye, G. D. (2018, January). Grading of harvested mangoes' quality and
maturity based on machine learning techniques. In the 2018 international conference on
smart city and emerging technology (ICSCET) (pp. 1-6). IEEE.

Qazi, A. M., Mahmood, S. H., Haleem, A., Bahl, S., Javaid, M., & Gopal, K. (2022). The
impact of smart materials, digital twins (DTs), and Internet of things (IoT) in an Industry
4.0 integrated automation industry. Materials Today: Proceedings, 62, 18-25.

Quinn, J., Frias-Martinez, V., & Subramanian, L. (2014). Computational sustainability and
artificial intelligence in the developing world. AI Magazine, 35(3), 36-47.

Ramcharan, A., Baranowski, K., McCloskey, P., Ahmed, B., Legg, J., & Hughes, D. P.
(2017). Deep learning for image-based cassava disease detection. Frontiers in plant
science, 8, 1852.

Rizvi, A. T., Haleem, A., Bahl, S., & Javaid, M. (2021). Artificial intelligence (AI) and its
applications in Indian manufacturing: a review. Current Advances in Mechanical
Engineering: Select Proceedings of ICRAMERD 2020, 825-835.

Sakai, H., Shiigi, T., Kondo, N., Ogawa, Y., & Taguchi, N. (2013). Accurate position
detection during asparagus spear harvesting using a laser sensor. Engineering in
agriculture, environment, and food, 6(3), 105-110.

Sambasivam, G. A. O. G. D., & Opiyo, G. D. (2021). A predictive machine learning

application in agriculture: Cassava disease detection and classification with imbalanced
dataset using convolutional neural networks. Egyptian informatics journal, 22(1), 27-34.

Santangeli, A., Chen, Y., Kluen, E., Chirumamilla, R., Tiainen, J., & Loehr, J. (2020).
Integrating drone-borne thermal imaging with artificial intelligence to locate bird nests on
agricultural land. Scientific reports, 10(1), 10993.

Sarkar, M. R., Masud, S. R., Hossen, M. I., & Goh, M. (2022, May). A comprehensive study
on the emerging effect of artificial intelligence in agriculture automation. In 2022 IEEE
18th International Colloquium on Signal Processing & Applications (CSPA) (pp. 419-
424). IEEE.
Sharma, P., Vimal, A., Vishvakarma, R., Kumar, P., porto de Souza Vandenberghe, L.,
Gaur, V. K., & Varjani, S. (2022). Deciphering the black box of omics approaches and
artificial intelligence in food waste transformation and mitigation. International Journal of
Food Microbiology, 372, 109691.

Shelake, S., Sutar, S., Salunkher, A., Patil, S., Patil, R., Patil, V., & Tamboli, T. (2021).
Design and implementation of artificial intelligence-powered agriculture multipurpose
robot. International Journal of Research in Engineering, Science, and Management, 4(8),
165-167.

Shin, J. Y., Kim, K. R., & Ha, J. C. (2020). Seasonal forecasting of daily mean air
temperatures using a coupled global climate model and machine learning algorithm for
field-scale agricultural management. Agricultural and Forest Meteorology, 281, 107858.

Singh, P., & Kaur, A. (2022). A systematic review of artificial intelligence in

agriculture. Deep Learning for Sustainable Agriculture, 57-80.

Singh, P., & Kaur, A. (2022). A systematic review of artificial intelligence in

agriculture. Deep Learning for Sustainable Agriculture, 57-80.

Sudars, K., Jasko, J., Namatevs, I., Ozola, L., & Badaukis, N. (2020). Dataset of annotated
food crops and weed images for robotic computer vision control. Data in brief, 31,
105833.

Sujatha, K., Koti, M. S., & Supriya, R. (2021). Analysis of farm data using artificial
intelligence. In Innovative Data Communication Technologies and Application:
Proceedings of ICIDCA 2020 (pp. 203-211). Springer Singapore.

Sujawat, G. S. (2021). Application of Artificial Intelligence in the detection of diseases in

plants: A Survey. Turkish Journal of Computer and Mathematics Education
(TURCOMAT), 12(3), 3301-3305.

Susheel, K. S., & Rajkumar, R. (2022). Pests And Diseases Detection Of Cotton Farm Using
Artificial Intelligence Technologies: A Review. NVEO-NATURAL VOLATILES &
ESSENTIAL OILS Journal| NVEO, 423-443.

Turing, A. M. (2009). Computing machinery and intelligence (pp. 23-65). Springer

Netherlands.

Tzachor, A., Devare, M., King, B., Avin, S., & Ó hÉigeartaigh, S. (2022). Responsible
artificial intelligence in agriculture requires a systemic understanding of risks and
externalities. Nature Machine Intelligence, 4(2), 104-109.
Upadhyay, N., & Gupta, N. (2021, October). A survey on disease detection for agriculture
crops using artificial intelligence. In 2021 5th International Conference on information
systems and computer networks (ISCON) (pp. 1-8). IEEE.
Waheed, A., Goyal, M., Gupta, D., Khanna, A., Hassanien, A. E., & Pandey, H. M. (2020).
An optimized dense convolutional neural network model for disease recognition and
classification in corn leaf. Computers and Electronics in Agriculture, 175, 105456.
Wang, X., Hu, W., Li, K., Song, L., & Song, L. (2018, December). Modeling of soft sensor
based on DBN-ELM and its application in the measurement of nutrient solution
composition for soilless culture. In 2018 IEEE International Conference of Safety Produce
Informatization (IICSPI) (pp. 93-97). IEEE.
Weng, S., Zhu, W., Zhang, X., Yuan, H., Zheng, L., Zhao, J., ... & Han, P. (2019). Recent
advances in Raman technology with applications in agriculture, food, and biosystems: A
review. Artificial Intelligence in Agriculture, 3, 1-10.

Zhang, J., Karkee, M., Zhang, Q., Zhang, X., Yaqoob, M., Fu, L., & Wang, S. (2020). Multi-
class object detection using faster R-CNN and estimation of shaking locations for
automated shake-and-catch apple harvesting. Computers and Electronics in
Agriculture, 173, 105384.

Zhang, P., Guo, Z., Ullah, S., Melagraki, G., Afantitis, A., & Lynch, I. (2021).
Nanotechnology and artificial intelligence to enable sustainable and precision
agriculture. Nature Plants, 7(7), 864-876.

Zhang, X., He, L., Zhang, J., Whiting, M. D., Karkee, M., & Zhang, Q. (2020).
Determination of key canopy parameters for mass mechanical apple harvesting using
supervised machine learning and principal component analysis (PCA). Biosystems
Engineering, 193, 247-263.

Zhou, Y., Xia, Q., Zhang, Z., Quan, M., & Li, H. (2022). Artificial intelligence and machine
learning for the green development of agriculture in the emerging manufacturing industry
in the IoT platform. Acta Agriculturae Scandinavica, Section B—Soil & Plan t Science, 72(1),
284-299.