International Journal of Trend in Scientific Research and Development (IJTSRD)

Volume 8 Issue 5, Sep-Oct 2024 Available Online: www.ijtsrd.com e-ISSN: 2456 – 6470

Farm Care Using Machine Learning & AI

Akshad Korle1, Devanshu Doijod2, Ibrar Ansari3, Ujjwal Lanjewar4, Dr. Suman Sen Gupta5
1,2,3,4
School of Science, G H Raisoni University, Amravati, Maharashtra, India
5
G H Raisoni University, Amravati, Maharashtra, India

ABSTRACT How to cite this paper: Akshad Korle |

The “Farm Care” is an AI-based project for user assistance. It refers Devanshu Doijod | Ibrar Ansari | Ujjwal
to a wide range of actions and procedures used to keep agricultural Lanjewar | Dr. Suman Sen Gupta "Farm
operations sustainable, productive, and healthy. To guarantee the best Care Using Machine Learning & AI"
Published in
possible growth and yield, it entails managing the environment,
International
livestock, and crops. Soil management, disease and insect prevention, Journal of Trend in
irrigation, nutrient management, are important facets of farm care. It Scientific Research
also entails putting best practices for resource efficiency, and Development
conservation, and compliance with regulations into action. In (ijtsrd), ISSN:
addition to increasing output and profitability, good farm 2456-6470, IJTSRD69450
management promotes environmental conservation and the welfare of Volume-8 | Issue-5,
rural people. It provides users with an intuitive interface to developed October 2024, pp.777-784, URL:
using Java for robust backend support, CSS for engaging and www.ijtsrd.com/papers/ijtsrd69450.pdf
responsive front-end design, and Python for advanced data analysis
and recommendation systems. With features like Soil management, Copyright © 2024 by author (s) and
International Journal of Trend in
disease and insect prevention, irrigation, nutrient management, “Farm Scientific Research and Development
Care” enhances the user experience. Journal. This is an
KEYWORDS: Farm care, Soil management, Data analysis, Open Access article
distributed under the
Recommendation system, Artificial intelligence, Irrigation, Welfare of
terms of the Creative Commons
rural people, Resource efficiency Attribution License (CC BY 4.0)
Technologies (http://creativecommons.org/licenses/by/4.0)
 PYTHON - Python is an interpreter, object-oriented, high level,
dynamically semantic programming language.
 Machine Learning - Decisions on which crop species to plant and
what tasks to complete during the growing season are influenced
by machine learning in agriculture.

I. INTRODUCTION
Welcome to "Farm Care", an inventive artificial It has a user-friendly interface that makes
intelligence project that aims to transform agricultural complicated agricultural operations easier. It was
operations through all-encompassing user support. created with Python for powerful data analysis and
Promoting profitable, healthy, and sustainable recommendation algorithms, CSS for an interesting
farming methods is the main goal of this state-of-the- and responsive front-end design, and Java for strong
art platform. "Farm Care" enables farmers to optimize backend support. "Farm Care" improves the user
their operations and achieve the highest potential experience by providing tools like disease detection,
growth and productivity by leveraging the power of soil testing, and customized guidance, empowering
technology. farmers to make informed decisions.
Fundamentally, "Farm Care" refers to a wide range of It also help farmers that use "Farm Care" can improve
actions and processes that address the complex productivity and profitability while simultaneously
requirements of agricultural management. This covers promoting environmental preservation and the
fertiliser management, irrigation, disease and pest welfare of rural communities. Our goal is to provide
control, and more. With its integration of everyone with access to sustainable farming methods
conservation, resource efficiency, and regulatory so that future generations can benefit from a healthy
compliance best practices, "Farm Care" elevates the planet. Come celebrate with us the "Farm Care" of
bar for farm management. farming's future.

@ IJTSRD | Unique Paper ID – IJTSRD69450 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 777
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
Smart farming uses cutting-edge, contemporary II. RELATED WORK:
technology to support precision agriculture and gives The study area's location:
farmers the ability to remotely monitor their plants. The study area lies between 20° 35' to 21° 44' N and
The automation of sensors and machinery has 78° 15' to 79° 40' E, eastern part of Maharashtra state,
increased the efficiency of the farming staff, which close to the geographical centre of India. It is a
benefits agricultural activities including crop yields portion of the clayey soil in western southern part,
and harvesting [16]. Agriculture undergoes a clay loam and sandy clay found in eastern part and
technological revolution as a result of the sandy clay loam found in north and south-east region
technologies that replace manual farming practices of Nagur. With a mean annual maximum of 27°C,
with automated machinery. Today's agricultural also the average yearly temperature is 27°C. Because
technology has changed farming practices, and the of the high annual rainfall of roughly 1064.1 mm,
Internet of Things has revolutionized traditional there is higher precipitation.
methods.
Soil samples collection and analysis:
Objectives : A broad range of soil textures, including clayey soil,
 To develop an AI-based platform that provides clay loam, and sandy clay loam, are present in the
comprehensive user assistance for sustainable, research area due to the consecutive layers of profile.
productive, and healthy agricultural operations. The research area was split into multiple sectors: The
 Improve farmer’s crop production. percentage of alluivial soil present in Nagpur is about
5%, percentage of basalt soil (Deccan trap) is 49%,
 Give farmers professional advice on managing percentage of lamta soil is 1%, percentage of
their soil, preventing diseases and insects, Gondwana soil is 10% and percentage of archeans
managing nutrients, and watering. soil is 35%. In order to cover the various soil types in
 To create an engaging and responsive front-end the study region, one hundred surface soil samples
design using CSS to enhance user experience. (0–10 cm) have been collected.
 Resource Optimization and risk management.

@ IJTSRD | Unique Paper ID – IJTSRD69450 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 778
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470

Remote sensing data collection & pre-processing:

On the same day, two different types of satellite photos were captured. Instead of using Landsat 8, the 30 m
spatial resolution Landsat 7 ETM+ image is utilized since it shows no clouds on the day of the field visit.
Another reason is that the research area is in the middle of the scene, meaning that no issues with faulty lines
coexist. To determine the soil moisture content, Sentinel-1 (ESA SAR) was utilized, which has a spatial
resolution of 10 m. A number of procedures are involved in pre-processing, such as picture calibration and
filtering (using Lee speckle filtering to eliminate dark spots and speckles; Lee et al., 1999).
Farmers can assess crop health by gathering information on vegetation indices, chlorophyll content, and water
stress. By using this information, agricultural producers can maximize crop output and reduce resource
consumption by making well-informed decisions about irrigation, fertilization, and pest control.

@ IJTSRD | Unique Paper ID – IJTSRD69450 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 779
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
III. PRAPOSED WORK:
A. Global Positioning System (GPS):
Latitude, longitude, and elevation data are precisely recorded by GPS. In order for GPS receivers to calculate
their location in real time and offer continuous positions while moving, Global Positioning System satellites
must emit signals. Farmers can use the precise location information to find field data, like weeds, pest
occurrence, soil type, and other obstacles, at the exact place. In order to apply the required inputs (seed,
fertilizer, herbicide, pesticide, and water) to a specific field, the system makes it easier to identify different field
locations.
B. Sensor Technology:
Numerous elements, including soil moisture, nutrient availability, light exposure, humidity, rainfall totals, leaf
colour, and more, all have an impact on a plant's health. By keeping the ideal temperature and light level, as well
as by micro irrigating to save water and energy, the plants are kept under observation. Numerous characteristics
are detected by various sensors. The sensor detects changes if they go over a threshold and sends the information
to the microcontroller, which then takes the necessary steps to bring the parameter back to its ideal level.

C. Geographic Information System (GIS):

In order to create maps and study characteristics and geography for statistical and spatial approaches, the GIS
consists of hardware and software that are designed to provide compilation, storage, retrieval, attributes analysis,
and location data [62]. The GIS database establishes the relationship between elements that affect a crop on a
specific farming field and gives information on field soil types, nutrient status, topography, irrigation, surface
and subsurface drainage, quantity of chemical applications, and crop yield [63]. The GIS is used for more than
just storing and displaying data; it also evaluates current and alternative management by combining and
modifying data layers to aid in decision-making.
D. Crop Management:
Data on soil changes and crop performance impacted by field topography can be obtained from satellite imagery.
Consequently, farmers are able to precisely monitor the production factors seeds, fertilizer, and pesticides that
lead to increased yields and efficiency.
At the regional level, the information is provided almost instantly thanks to the satellite pictures' temporal revisit
frequency and spatial coverage. Particularly in red and near-infrared combinations (vegetation indices) to
monitor green foliage, the spectral reflectance properties of vegetation predict the relationship between the
spectral features of crops and their biomass/yield tests. Because leaf area and crop productivity are strongly
associated, the normalized difference vegetation index (NDVI) is the most often used indicator of these two
parameters.

@ IJTSRD | Unique Paper ID – IJTSRD69450 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 780
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470

Because of its strong relationship to the leaf area index (LAI) and photosynthetic activity of green vegetation,
the normalized difference vegetation index (NDVI) is the most often used indicator to evaluate the health of the
vegetation and crop productivity. The interpretation of indications obtained from remote sensing by contrasting
the current crop status with that of prior or typical seasons is the basis of crop monitoring techniques. In some
cases, crop production can be estimated ahead of harvest time thanks to the correlation between biomass and
vegetation indices [66]. The more complex functionalities made possible by automated field management
include the automated data acquisition, processing, monitoring, decision-making, and management of farm
operations including the fundamental tasks of crop production (yields), profits and losses, farm weather
prediction, field mapping, and soil nutrient tracking.
FUT estimated depending on the best wavelengths. Using
A. Soil and plant sensors: NIR spectrophotometry technology, the soil's
Sensor technology, a key component of precision nitrogen and phosphorus contents are anticipated.
agriculture, gives data on the fertility, water status, Since ECa is sensitive to variations in soil texture and
and soil qualities. As a result, new sensors have been salinity, the soil apparent electrical conductivity
created that differ from those that are currently on the (ECa) sensors continuously gather data on the field
market and are based on desirable properties. surface. Optoelectronic, acoustic, impedance, and
Plant wearables and soil sensors track physical and nanostructured biosensors are used to identify soil
chemical signals in the soil in real time, including insects and pests.
temperature, pH, moisture content, and pollutants. B. Yield Monitor:
The data they offer helps farmers improve crop The set of sensors and other parts that manage
development conditions, combat biotic and abiotic integration and interaction components—such as a
challenges, and boost yields. The most crucial computer, data storage device, and user interface—
elements for crop production are nitrogen (N), are called yield monitors. The force of mass or
potassium (K), phosphorus (P), and soil organic volume of grain flow is evaluated by the sensor to
matter (SOMs). continually measure yield. The idea behind the mass
The nitrogen content of both surface and deep soil is flow sensor was to measure the energy that rebounds
measured spatially using NIR reflectance-based after being struck by microwave energy beams. GPS
sensors. By evaluating the soil's spectral reflectance receivers in yield monitors use location yield data to
in the infrared and visible wavelength areas, SOM is generate yield maps.

@ IJTSRD | Unique Paper ID – IJTSRD69450 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 781
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
To display real-time harvest data, the yield monitor is AI Use Cases for the Farm Care Application:
mounted atop a harvester, connected to a mobile app, Image Recognition System:
and it automatically uploads data to a web platform. AI-based computer vision models trained on photos
Farmers can export more farm management data for of healthy and sick crops are used to detect pests and
examination, and the software can create and share diseases.
excellent yield maps with an agronomic. Fruit growth
Crop health, growth rates, and abnormalities can be
is regarded as one of the most important factors in the
tracked using drones and cameras to take pictures that
crop developing stage in horticulture crops, providing
AI models can process.
precise information about the quantity and quality of
food produced. Color photos are used to monitor fruit NLP, or natural language processing:
conditions in order to determine when to harvest, AI chatbots: Giving farmers suggestions or guidance
target the appropriate market, and estimate fruit on farm management.
maturity. One method for tracking agricultural yields Voice assistance: AI-driven voice interfaces that can
across large areas in real time is the use of satellite be used to manage Internet of Things equipment or to
pictures. deliver real-time farm status updates.
Tools for AI and Machine Learning: Automation and Assisting with Decisions:
AI/ML Frameworks: AI is being used to automate farm machinery,
TensorFlow: Used to create machine learning models including drones, autonomous tractors, and other
for tasks like soil analysis, yield prediction, and crop devices that run on AI algorithms.
disease detection.
AI decision support systems: Based on data analysis,
PyTorch: Used to train deep learning models for AI models indicate the best times to harvest crops,
things like crop health and pest identification through rotate crops, or apply fertilizer.
images.
Tools for Application Development:
Scikit-learn: For easier machine learning assignments Front-end Programming:
like yield prediction or best-practice planting For web-based interfaces, use Vue.js, Angular, or
schedule recommendations. React.
Keras: It is an approachable deep learning framework For mobile applications, use Swift/Kotlin, Flutter, or
that is frequently combined with TensorFlow. React Native
Pre-programmed AI Models: Development of Backend:
Google Cloud AutoML: Especially helpful for picture For creating server-side APIs, you can use Python
and object detection (e.g., identifying plant diseases) (with Django or Flask), Node.js (Express), or Java
for creating bespoke AI models. (Spring Boot).
Microsoft Azure Cognitive Services: pre-configured Database Administration:
AI models for tasks related to farm management, such SQL databases for storing organized agricultural data,
as computer vision and natural language processing. such as PostgreSQL or MySQL.
AWS SageMaker: For creating, honing, and MongoDB and other NoSQL databases are useful for
implementing cloud-based machine learning models. managing unstructured data, like sensor feeds and
Data Sources & Sensors (IoT Integration): drone footage.
Internet of Things Sensors: Deployment and Monitoring of AI Models:
Sensors for the soil (pH, moisture, etc.) weather Deployment Platform Models:
stations for monitoring the climate in real time AI models can be deployed and containerized in
Satellites or drones for crop imaging from the air AI scalable settings using Docker or Kubernetes.
models are linked to smart irrigation systems to TensorFlow Serving, also known as TorchServe: For
enable automated watering. providing farm care applications with trained models
Tools for Data Processing: as APIs.
For processing and broadcasting real-time data from Tools for Model Monitoring:
drones, Internet of Things sensors, and other data Grafana or Prometheus: For tracking the effectiveness
sources, use Apache Spark or Kafka. of AI models, especially those that incorporate real-
Time-series databases, such as InfluxDB, are used to time data from sensors or drones.
store and analyze trends over time in sensor data,
such as temperature and soil moisture.

@ IJTSRD | Unique Paper ID – IJTSRD69450 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 782
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
Machine Learning flow:  Initial Costs: Implementing advanced
For monitoring model trials, installations, and technologies and systems can require significant
evaluations of performance. upfront investment, which may be a barrier for
IV. RESULT ANALYSIS: small-scale farmers.
The "Farm Care" project is a noteworthy  Training Requirements: Farmers may need
development in the use of machine learning and training to effectively use the platform, which
artificial intelligence in agricultural settings. The could pose an additional challenge for those less
platform encourages sustainable farming practices familiar with technology.
while also increasing farm output with its extensive
 Data Privacy Concerns: The collection and
user help. Agricultural parameters including crop
analysis of agricultural data raise potential
performance, soil health, and resource utilization can
privacy issues that need to be addressed.
now be precisely monitored and managed thanks to
the integration of technologies like GIS, sensor VI. REFERENCES:
systems, and IoT. [1] M.J. Carrer, H.M. de Souza Filho, M.D.M.B.
Vinholis, C.I. Mozambani “Precision
"Farm Care" provides a user-friendly interface with
agriculture adoption and technical efficiency:
sophisticated data analysis and recommendation
an analysis of sugarcane farms in Brazil
systems created in Python, responsive front-end
Technol. Forecast. Soc. Change”, 177 (2022)”,
design using CSS, and strong back-end support in
Article 121510 [View PDF] [View article]
Java. This gives farmers the information they need to
[View in Scopus] [Google Scholar]
maximize crop productivity while protecting the
environment and maintaining the welfare of rural [2] D.C. Rose, J. Chilvers “Agriculture 4.0:
communities. broadening responsible innovation in an era of
smart farming Front. Sustain. Food Syst”., 2
V. CONCLUSION:
(2018), p. 87 [View in Scopus] [Google
Farm Care" is a noteworthy development in
Scholar]
agricultural management that combines sustainable
farming methods with technology to improve farming [3] Srisruthi, S.; Swarna, N.; Ros, G.M.S.;
production and efficiency. It gives farmers the ability Elizabeth, E.” Sustainable agriculture using
to maximise their productivity while emphasising eco-friendly and energy efficient sensor
environmental stewardship by employing state-of-the- technology.” In Proceedings of the 2016 IEEE
art equipment and techniques. International Conference on Recent Trends in
Electronics, Information & Communication
Advantages:
Technology (RTEICT), Bangalore, India, 20–
 Enhanced Efficiency: Automation and smart
21 May 2016; IEEE: Bangalore, India, 2016;
farming technologies streamline operations,
pp. [Google Scholar] [CrossRef]
reducing labor costs and increasing crop yields.
[4] G. Sponchioni, M. Vezzoni, A. Bacchetti, M.
 Data-Driven Decision Making: Access to real-
Pavesi, F.M. Renga “The 4.0 revolution in
time data and analytics allows farmers to make
agriculture: a multi-perspective definition
informed choices about irrigation, fertilization,
Summer School F. “Turco-Industrial Systems
and pest control, leading to better resource
Engineering (2019)”, [View in Scopus]
management.
[Google Scholar]
 Sustainability: "Farm Care" promotes
[5] Erdoğan Assessing farmers' perception to
environmentally friendly practices, supporting
Agriculture 4.0 technologies: a new
soil health, biodiversity, and the conservation of
interval‐valued spherical fuzzy sets based
natural resources.
approach Int. J. Intell. Syst., [View at
 User-Friendly Interface: The platform's intuitive publisher] [Crossref] [View in Scopus] [Google
design ensures that farmers, regardless of Scholar]
technical expertise, can easily access and utilize
[6] Reddy, T.; Dutta, M. Impact of Agricultural
its features
Inputs on Agricultural GDP in Indian
Restriction: Economy. Theor. Econ. Lett. 2018, 8, 1840–
 Technology Dependence: Increased reliance on 1853. [Google Scholar] [CrossRef]
technology may lead to challenges if systems fail
[7] Patil, K.A.; Kale, N.R. A “model for smart
or if there is a lack of internet connectivity in
agriculture using IoT.” In Proceedings of the
rural areas.

@ IJTSRD | Unique Paper ID – IJTSRD69450 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 783
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
2016 International Conference on Global [14] Albaum, G. (1997). The Likert scale revisited.
Trends in Signal Processing, Information Market Research Society Journal, 39(2), 1–21
Computing and Communication, Jalgaon, India, [View] [Google Scholar]
22–24 December 2016; IEEE: Jalgaon, India, [15] ACIAR. (2016). Management of nutrients for
2016;. [Google Scholar] [CrossRef] improved profitability and sustainability of
[8] Shi, X.; An, X.; Zhao, Q.; Liu, H.; Xia, L.; Sun, crop production in Central Myanmar.
X.; Guo, Y. State- of- the- Art Internet of Retrieved from
Things in Protected Agriculture. Sensors 2019,. https://www.aciar.gov.au/project/SMCN-2014-
[Google Scholar] [CrossRef] 044 Google Scholar
[9] King, T.; Cole, M.; Farber, J.M.; Eisenbrand, [16] V. Saiz-Rubio, F. Rovira-Más From smart
G.; Zabaras, D.; Fox, E.M.; Hill, J.P. Food farming towards agriculture 5.0: a review on
safety for food security: Relationship between crop data management Agronomy, 10 (2)
global megatrends and developments in food (2020), [View at publisher] [Crossref] [View in
safety. Trends Food Sci. Technol. 2017, 68, Scopus] [Google Scholar]
160–175. [Google Scholar] [CrossRef] [17] Siddharth, D.K. Saini, A. Kumar Precision
[10] Kong, Q.; Chen, H.; Mo, Y.L.; Song, G. Real- Agriculture with Technologies for Smart
time monitoring of water content in sandy soil Farming towards Agriculture 5.0. Unmanned
using shear mode piezoceramic transducers and Aerial Vehicles For Internet Of Things (IoT)
active sensing-A feasibility study. Sensors Concepts, Techniques, and Applications (2021),
2017, 17, 2395. [Google Scholar] [CrossRef] [View at publisher] [Crossref] [Google Scholar]
[11] Yunus, M.A.M.; Mukhopadhyay, S.C. Novel [18] Lajoie-O'Malley, K. Bronson, S. van der Burg,
Planar Electromagnetic Sensors for Detection L. Klerkx The future (s) of digital agriculture
of Nitrates and Contamination in Natural Water and sustainable food systems: an analysis of
Sources. IEEE Sens. J. 2011, 11, 1440–1447. high-level policy documents Ecosyst. Serv., 45
[Google Scholar] [CrossRef] (2020), Article 101183 [View PDF] [View
article] [View in Scopus] [Google Scholar]
[12] G. Gagliardi, M. Lupia, G. Cario, F. Cicchello
Gaccio, V. D'Angelo, A.I.M. Cosma, A. [19] Nuthall P (2011a) Common methods used in
Casavola An internet of things solution for the analysis of farming systems. CAB Rev
smart agriculture Agronomy, 11 (11) (2021), Perspect Agric Vet Sci Nutr Nat Resour
[View at publisher] [Crossref] [View in 6(040):1–7. doi:10.1079/pavsnnr20116040.
Scopus] [Google Scholar] https://gsda.maharashtra.gov.in/nagpur-district
[13] W. Liu, X.F. Shao, C.H. Wu, P. Qiao A [20] भूजल सव ण आिण िवकास यं णा, महारा शासन,
systematic literature review on applications of भारत यां चे हे अिधकृत संकेत ळ आहे. सव# अिधकार
information and communication technologies सुरि त. Powered By: TechBeats Software Pvt
and blockchain technologies for precision Ltd.
agriculture development J. Clean. Prod., 298
(2021), Article 126763 [View PDF] [View
article] [View in Scopus] [Google Scholar]

@ IJTSRD | Unique Paper ID – IJTSRD69450 | Volume – 8 | Issue – 5 | Sep-Oct 2024 Page 784