Available online at www.sciencedirect.

com

ScienceDirect
Procedia Computer Science 260 (2025) 552–560

Seventh International Conference on Recent Trends in Image Processing and Pattern Recognition (RTIP2R-2024)

IoT-Driven Smart Farming with Machine Learning for Sustainable

Food Systems
Sanjana Murgoda, Tanushree Kabbura, Bibijan Mattea, Vaibhav Mujumdara, Meenaxi M
Raikara*
a
Department of Computer Science and Engineering, KLE Technological University Hubli, Karnataka, India

Abstract

Integrating the Internet of Things (IoT) and machine learning (ML) technologies in agriculture, commonly called smart farming,
is revolutionizing the sector by enhancing productivity, efficiency, and sustainability. This paper explores the application of IoT-
driven smart farming using machine learning for sustainable agricultural practices. The system introduces an efficient Soil
Moisture Detection System utilizing IoT Technology, revolutionizing modern farming practices. By continuously monitoring
crucial parameters such as soil moisture, temperature, and humidity in real-time, the system ensures seamless data transmission to
a centralized server. Additionally, integrating motion detection capabilities enhances security measures and promptly alerts
farmers to environmental changes. The dataset consisting of 100,000 rows is generated to facilitate the development and training
of five ML models to predict soil moisture trends. Decision Trees achieved an accuracy rate of 99.98%, while Random Forests
achieved 99.99%. The integration of these predictive models empowers farmers with actionable insights for precise irrigation
scheduling and optimal crop yield optimization. These models provide actionable insights for precise irrigation scheduling and
optimal crop yield optimization. Field tests have confirmed the efficacy of this approach, demonstrating significant improvements
in irrigation efficiency and subsequent crop yields. Thus, the proposed system represents a substantial advancement in leveraging
the synergistic potential of IoT and ML technologies to foster sustainable agricultural practices.
© 2025 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0)
Peer-review under responsibility of the scientific committee of the Seventh International Conference on Recent Trends in Image
Processing and Pattern Recognition.
Keywords: Precision Agriculture, Crop Management, Soil Moisture Detection, Blynk Integration, Motion Detection (PIR Sensor).

* Corresponding author. Tel.: +0-000-000-0000 ; fax: +0-000-000-0000 .

E-mail address: [email protected]

1877-0509 © 2025 The Authors. Published by Elsevier B.V.

This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0)
Peer-review under responsibility of the scientific committee of the Seventh International Conference on Recent Trends
in Image Processing and Pattern Recognition.
10.1016/j.procs.2025.03.233
 Sanjana Murgod et al. / Procedia Computer Science 260 (2025) 552–560 553

1. Introduction

The increasing demand for food due to population growth necessitates the adoption of innovative technologies in
agriculture to enhance productivity and sustainability. IoT and machine learning have emerged as pivotal
technologies in smart farming, enabling precise monitoring and management of agricultural processes. This paper
delves into the use of IoT and ML to foster sustainable farming practices. Soil moisture monitoring is essential in
agricultural contexts, profoundly influencing crop growth and yield. Optimizing irrigation strategies demands
precise monitoring, driving the integration of Internet of Things (IoT) Technology and machine learning (ML)
techniques into agricultural practices. This study introduces an IoT-based Soil Moisture Detection System
coupled with ML algorithms to reform soil moisture monitoring and irrigation management in agriculture. This
system offers real-time monitoring of soil moisture levels, temperature, and humidity, providing farmers with
actionable insights for irrigation optimization. Continuous data streams are collected and transmitted to a
centralized server for analysis by deploying IoT sensors in agricultural fields. ML techniques optimize the use of
water, fertilizers, and pesticides, minimizing waste and environmental impact. Precision irrigation systems
powered by ML ensure water is supplied based on the actual needs of the crops. This study represents a
significant advancement in leveraging IoT and ML technologies for sustainable agriculture, addressing the critical
need for precise soil moisture monitoring and resource management in modern farming practices. The synergy
between IoT and ML promotes sustainable agricultural practices by enabling precise and efficient resource
utilization. This integration reduces the carbon footprint of farming activities and conserves natural resources.
Precision farming in rice cultivation exemplifies this potential, where IoT sensors monitor critical parameters
such as soil moisture and nutrient levels. Machine learning models predict optimal irrigation schedules, resulting
in a notable 20% reduction in water usage. Similarly, pest management in vineyards benefits from IoT and ML
integration. IoT devices detect early signs of pest activity, and ML algorithms analyze the data to recommend
targeted pesticide applications, thereby reducing chemical use by 30%. These case studies highlight the
significant impact of IoT and ML in optimizing resource utilization and improving agricultural outcomes,
demonstrating their vital role in advancing sustainable farming practices. Despite these advancements, several
challenges must be addressed to fully realize IoT and ML's potential in agriculture. Data privacy and security
remain critical concerns, as the extensive use of IoT devices increases the risk of data breaches and cyber threats
[15]. Additionally, the high initial costs associated with implementing IoT and ML technologies pose a barrier,
particularly for small-scale farmers who may find such investments prohibitive. The following are the
contributions of the proposed system:
1. The IoT-based smart farming system employs machine learning to predict irrigation needs, optimize
water usage, and reduce waste by activating the pump only when necessary.
2. The proposed system implements continuous adaptation with dynamic adjustments based on ongoing
sensor readings and machine learning predictions to ensure the system remains responsive to changing
environmental conditions.

2. Literature survey

In [1], the authors present the design and development of a system for measuring and monitoring soil moisture
using a low-cost soil moisture sensor interfaced with a CC3200 LaunchPad, leveraging the Internet of Things (IoT),
Cloud computing, and Mobile computing technologies. In [2], the authors present the design and validation of a
novel soil moisture sensor to enhance agricultural irrigation efficiency through precise moisture measurement.
Experimental data revealed a strong linear negative correlation between the sensor's output voltage and soil
volumetric moisture content, with a coefficient of determination (R²) greater than 0.986. Accuracy tests indicated
that the sensor's average absolute error was below 2.5%, demonstrating its reliable performance, high accuracy, and
suitability for various soil types. In [3], the authors examine the evolution and principles of soil moisture sensors
over the past 20-30 years, highlighting their critical role in agriculture for optimizing irrigation and crop yield. It
summarizes the various applications of these sensors and compares their advantages, disadvantages, and influencing
factors. In [4], the study addresses the challenge of Technology adoption among new farmers, particularly in the
 554 Sanjana Murgod et al. / Procedia Computer Science 260 (2025) 552–560

context of an aging agricultural workforce and a shortage of successors. Recognizing that effective water
management is a critical aspect of farming practice, the research focuses on developing a soil moisture measurement
system using moisture sensors. In [5], the authors highlight the crucial role of soil moisture content, a vital nutrient
influenced by soil physical properties such as color, texture, structure, and bulk density. It underscores the
importance of soil moisture for various applications in hydrology, horticulture, agriculture, and meteorology,
impacting plant growth, percolation, evaporation, and heat exchange. In [6], the authors comprehensively review
soil moisture sensors and explore the potential for their development using nanomaterials. It discusses the
importance of soil moisture detection, various methods for its measurement, and the role of nanotechnology in
enhancing agricultural practices. In [7], the authors evaluate the accuracy of Insentek soil moisture sensors through
both laboratory and field lysimetric experiments, considering various soil conditions and irrigation levels. Using
standard methods like microweighing lysimeters and oven-drying, the researchers found that in laboratory
conditions, the Insentek sensors demonstrated high accuracy, with root mean square error (RMSE) values of 0.89–
1.04% vol and relative prediction deviation (RPD) values above the reliable threshold of 4.0.
In [8], the authors comprehensively evaluate soil moisture sensing technologies suitable for field applications,
addressing the increasing demand for smart irrigation and water-saving practices in agriculture. In [9], the authors
address the pressing issue of declining water content on Earth, leading to the desertification of once-fertile lands and
threatening agricultural productivity. Leveraging advancements in Internet of Things (IoT) Technology, particularly
using the FC-28 soil moisture sensor, the study proposes a solution to measure soil moisture levels accurately.
Integrating this sensor with a microcontroller and ESP8266 Wi-Fi module enables real-time monitoring and data
storage in the cloud. In [10], the authors address the critical issue of water scarcity in agricultural regions,
particularly in third-world countries where traditional practices often overlook efficient water management. The
study offers a practical solution to control water supply in water-deficient areas. In [11], the authors address the need
for efficient monitoring of environmental factors in agricultural settings, emphasizing the significance of
technologies to streamline data collection and analysis processes. The sensor, engineered with affordable materials
and methods, detects moisture levels in the soil through resistance changes. Leveraging XBEE wireless
communication modules, moisture data collected by sensor nodes is transmitted to a centralized server for water
supply control to streamline data collection and analysis processes. In [12], the authors present the development of
low-power, cost-effective dual-probe heat-pulse (DPHP) soil moisture sensors designed to address the specific needs
of developing countries. The sensor utilizes two probes - a heater and a temperature sensor spaced 3 mm apart - to
measure soil moisture content efficiently. With a supply voltage of 3.3 V and a low power consumption of 330 mW,
the sensor offers an energy-efficient solution suitable for resource-constrained environments. In [13], a
comprehensive review of the pivotal role of machine learning (ML) algorithms in soil moisture assessment across
various research domains is elucidated. Highlighting their versatility, ML algorithms such as artificial neural
networks, support vector machines, decision trees, and random forests are showcased for their adeptness in
modeling intricate relationships within soil moisture data. In [14], the authors evaluate the efficacy of three machine
learning (ML) methods for estimating soil moisture. In [16], authors apply YOLO model to detect and count the
tomatoes.

3. Methodology

In the proposed system, integration of IoT and ML is performed to automate the irrigation process by capturing the
different sensor data presented in Fig 1. IoT in smart farming involves interconnected devices such as sensors, actuators,
and communication networks. These devices collect real-time data, which are transmitted to a central system for analysis.
The methodology includes three phases, which are presented in the following sections.

3.1 Design and Implementation of IoT-Based Soil Moisture Detection System

This stage involves meticulous sensor selection, system architecture design, and network configuration for seamless
data acquisition from the agricultural field. The IoT sensors continuously collect real-time data and transmit it to a
centralized server for preprocessing and analysis. The components used for the prototype development are presented in
the next section.
• NodeMCU ESP8266: Handles data collection, processing, and transmission.
• Soil Moisture Sensor: Measures soil moisture levels for irrigation management.
 Sanjana Murgod et al. / Procedia Computer Science 260 (2025) 552–560 555

• PIR Motion Sensor: Enhances security by detecting field motion.

• Relay Module: Controls water pump activation based on sensor data.
• Breadboard and Jumpers: Facilitate circuit creation for easy connections.
• 18650 Battery and Push Button: Power system and provide manual control.
• Water Pump: Executes irrigation based on microcontroller commands.
• Mobile App for Remote Control and Database Storage: The Blynk app is integrated for real-time monitoring and
visualization.
The sensors, NodeMCU ESP8266, and the Blynk app are integrated for remote monitoring and control, as presented in
Fig 2 and 3. Sensors and actuators are interconnected to create a cohesive real-time monitoring system.

Fig 1. Proposed IoT based smart farming system activity diagram.

Fig. 2. Prototype development of IoT-based smart farming.

Fig. 3. Blynk app for remote monitoring of parameters and control.
556 Sanjana Murgod et al. / Procedia Computer Science 260 (2025) 552–560

• Implementation

The integration of the sensors, NodeMCU ESP8266, and the Blynk app are performed for remote monitoring and
control, as presented in Fig 2 and Fig 3. Sensors and actuators are interconnected to create a cohesive real-time
monitoring system.

• Data Collection and Communication

Sensors deployed across the farm gather extensive data, which is communicated via wireless networks to cloud-
based platforms. This data forms the foundation for informed decision-making and efficient farm management. Data
collected from sensors is transmitted wirelessly to the central NodeMCU ESP8266, then to the Blynk cloud
platform. Blynk Cloud stores and provides real-time monitoring via the app, supporting advanced applications such
as ML model training.
Algorithm-1: IoT-driven smart framing using machine learning for sustainability.
Inputs:
Sensor data, user interface commands
Output: The water pump is ON/OFF manually or automatically.
Steps:
1: Initialization
1.1 Set up the hardware components:
- LCD
- DHT11 sensor
- Soil moisture sensor
- PIR motion sensor
- Relay module
1.2 Establish connections to the Blynk cloud platform for data transmission and remote monitoring.
1.3 Initialize the Blynk library and associated timers for scheduled tasks.
2: Temperature and Humidity Data Collection
2.1 Read temperature and humidity values from the DHT11 sensor.
2.2 Validate the sensor readings for accuracy and consistency.
3: Soil Moisture Data Collection
3.1 Read soil moisture levels from the soil moisture sensor.
3.2 Validate the sensor readings to ensure reliable data acquisition.
4: Motion Detection
4.1 Read the PIR motion sensor to detect any movement in the vicinity.
4.2 Validate motion detection to avoid false positives.
5: Manual Irrigation Control
5.1 Monitor physical button presses to toggle the relay state, which controls the water pump.
5.2 Update the system status based on the button press actions.
6: Data Transmission to Blynk
6.1 Transmit the collected sensor data (temperature, humidity, soil moisture, motion status) to the Blynk cloud
platform for real-time monitoring.
6.2 Ensure data integrity during transmission.
7: Motion Event Handling
7.1 If motion is detected, activate the motion LED indicator.
7.2 Log the motion event in the Blynk cloud for historical data analysis.
8: Automated Irrigation Control
8.1 Use the ML model to predict irrigation needs based on historical soil moisture data and current readings.
8.2 If the ML model predicts a need for irrigation:
- Activate the water pump via the relay.
- Display the watering status on the LCD display.
9: Water Conservation
9.1 If soil moisture levels are sufficient, deactivate the water pump to conserve water resources.
9.2 Update the LCD display to reflect the current irrigation status.
10: Continuous Operation
10.1 Run the Blynk library and associated timers to ensure seamless communication with the cloud platform.
10.2 Execute scheduled tasks, including data synchronization and ML model predictions.
11: Monitoring Loop
11.1 Continuously loop through steps 2 to 10 to maintain real-time monitoring and control.
11.2 Adapt irrigation practices dynamically based on sensor data and ML predictions.

3.2 Development and Evaluation of Machine Learning Models for Soil Moisture Prediction
 Sanjana Murgod et al. / Procedia Computer Science 260 (2025) 552–560 557

The second phase aims to develop ML models to predict soil moisture trends and leverage IoT sensor data
accurately. Through exhaustive experimentation with algorithms like Logistic Regression (71.77%accuracy),
Decision Trees (99.98%), Random Forests (99.99%), Support Vector Machines (91.92%), and Neural Networks
(95.01%), the optimal approach is determined. The models are trained on preprocessed data and evaluated using
performance metrics for efficacy in predicting soil moisture dynamics.
Training Process:
• Data Collection and Preprocessing: Historical data from sensors measuring soil moisture, temperature,
and humidity are collected and cleaned to handle missing values and remove outliers.
• Feature Engineering and Standardization: Feature engineering techniques are applied to create new
features, and the data is standardized or normalized to ensure consistency.
• Model Evaluation and Deployment: Model performance is assessed using metrics like Mean Absolute
Error (MAE), Root Mean Squared Error (RMSE), and R-squared (R²) on the test dataset. The model
demonstrating the highest accuracy and reliability is selected for deployment. It is integrated into the IoT-
based system for real-time soil moisture predictions, followed by validation through field tests to ensure
effectiveness.
Additionally, it's noteworthy that the dataset utilized in this study comprises 100,000 rows of input, providing a
robust foundation for training and evaluating the predictive models.

3.3 Integration, Testing, and Refinement of the Integrated System

The integration, testing, and refinement phase is pivotal, seamlessly incorporating trained machine learning
models into the Soil Moisture Detection System and integrating hardware and software components to forge a
cohesive and responsive ecosystem. The IoT-based smart farming system sequence diagram is presented in Fig 4.
The prototype soil moisture detection System leverages the Blynk platform, integrating features for monitoring
temperature, humidity, soil moisture, and motion detection.

Fig. 4. IoT-based smart farming system sequence diagram.

The cost analysis of the system includes high initial investments in hardware, software, and infrastructure, as well
as ongoing maintenance and training expenses. However, the benefits are increased efficiency, higher yields, and
sustainable practices. Funding options like government grants, agricultural loans, and public-private partnerships can
alleviate financial burdens, while phased implementation allows for gradual technology adoption and risk
management.

4. Result Analysis

Machine learning algorithms analyze real-time data to predict crop yields, pest infestations, and disease
outbreaks. These predictive models help farmers make proactive decisions to mitigate risks.
558 Sanjana Murgod et al. / Procedia Computer Science 260 (2025) 552–560

4.1 Data visualization

Fig. 5. Pair plot - Visualizing relationships between Fig. 6. Correlation Heatmap - Highlighting feature correlations
soil parameters and status classifications. for IoT sensor data and machine learning.

The pair plot, correlation heatmap, and boxplot of the live data collected with the aid of sensors are presented for
visualization of the data collected.
• Pair plot - The Pair plot visualization provides an insightful depiction of the relationships between soil moisture,
temperature, and humidity and their respective status classifications predicted by machine learning models. Each
scatterplot within the grid illustrates the correlation between two variables, with colors representing different
classes or statuses (ON and OFF), as presented in Fig 5. Distinct patterns emerge from the pair plot, revealing
correlations between soil moisture, temperature, and humidity and their influence on the predicted status
classifications. For instance, higher soil moisture levels often coincide with lower temperatures, suggesting an
inverse relationship between these parameters. These observations suggest that specific combinations of soil
conditions are more conducive to particular status classifications, offering valuable insights for irrigation
management.
• Correlation - The correlation heatmap displays the correlations among different features derived from IoT
sensor data. Visualizing correlation coefficients offers insights into the relationships between soil moisture,
temperature, humidity, and other environmental factors, as presented in Fig 6. The heatmap indicates strong
correlations between soil moisture and temperature, soil moisture and humidity. These correlations imply that
changes in temperature and humidity levels are likely to influence soil moisture dynamics. Additionally, the
moderate negative correlation between time and soil moisture suggests temporal fluctuations in moisture, which
could inform irrigation scheduling.
• Boxplot - The boxplot analysis provides an overview of the distribution of soil parameters, facilitating anomaly
detection and preprocessing. Visualizing statistical metrics such as median, quartiles, and outliers reveals soil
moisture, temperature, and humidity variability and central tendencies. The boxplots highlight variations in soil
moisture distribution over different time periods, with some instances of outliers indicating anomalous soil
moisture levels, as presented in Fig 7. These outliers may signify irregular environmental conditions,
necessitating further investigation and preprocessing to ensure data quality and model robustness.
• Histogram - The histograms depict the frequency distribution of soil parameters, offering insights into feature
engineering and model accuracy assessment. Examining the frequency of values within each parameter reveals
distribution patterns and potential clusters in the data. The histograms show non-normal distributions in soil
moisture, temperature, and humidity, suggesting the presence of underlying patterns or clusters within the dataset,
as presented in Fig 8.
 Sanjana Murgod et al. / Procedia Computer Science 260 (2025) 552–560 559

Sanjana Murgod Computer

Fig. 7. Boxplot - Illustrating soil parameter distribution for Fig. 8. Histograms - Analyzing frequency distribution of the sensor
anomaly detection and preprocessing. data.
Various visualization techniques, including pair plots, correlation heatmaps, boxplots, and histograms, facilitated
a comprehensive understanding of the sensor data, aiding in anomaly detection and feature engineering. The
machine learning model - Logistic Regression (71.77% accuracy), Decision Trees (99.98%), Random Forests
(99.99%), Support Vector Machines (91.92%), and Neural Networks (95.01%) - exhibited varying levels of
accuracy in predicting soil moisture trends. The comparative performance analysis of the models is presented in
Fig 9. Based on these results, the Random Forest model, which demonstrated the highest accuracy, was selected
for prediction, further validating the system's predictive capabilities.

Fig. 9. Comparative performance analysis of the ML models.

5. Conclusion and Future Work

IoT-driven smart farming using machine learning offers a promising pathway to achieve sustainable
agriculture. The IoT-based Soil Moisture Detection System represents a significant technological advancement in
precision agriculture. By integrating real-time monitoring of soil moisture, temperature, and humidity with
advanced machine learning models, the system provides critical insights for optimizing irrigation schedules and
improving crop yields. This integrated approach underscores the synergistic potential of IoT and machine
learning technologies in fostering sustainable agricultural practices and enhancing resource management.
The future scope of this research includes several potential enhancements and expansions. Firstly, the
system can be scaled to include additional environmental parameters such as soil nutrient levels, pH, and weather
conditions, providing a more holistic view of the agricultural environment. Integrating advanced IoT technologies
like edge computing could improve data processing speed and reduce latency, enabling timely decision-making.
Furthermore, incorporating adaptive machine learning techniques that continuously learn from new data could
enhance the system's predictive accuracy over time. Another promising direction is the development of a user-
friendly mobile application, allowing users to receive real-time alerts and recommendations directly on their
devices. The limitation of the proposed system is data privacy, and security parameters are not incorporated in the
IoT-driven smart farming. The plan is to enhance the system by including federated learning techniques to
provide data privacy in IoT-based agricultural applications.
560 Sanjana Murgod et al. / Procedia Computer Science 260 (2025) 552–560

References

[1] PalleDivyavani, Raghavender Rao, ‘Measurement and monitoring of soil moisture using cloud IOT and Android system’, Indian Journal
of Science and Technology, Vol 9(31), DOI: 10.17485/ijst/2016/v9i31/95340, August 2016.
[2] Xingchao Zhang, Chengzhi Yang, and Lu Wang, ‘Research and application of new soil moisture sensor’, MATEC Web of Conferences
175, 02010(2018). IFCAE-IOT-2018, DOI: org/10.1051/matecconf/201817502010.
[3] Limin Yu, Wanlin Gao, Redmond R. Shamshiri, Sha Tao, Yanzhao Ren, Yanjun Zhang, GuilianSu, ‘Review of research progress on soil
moisture sensor technology’, International Journal of Agricultural and Biological Engineering, July 2021, Vol 14 No.4, DOI:
10.25165/j.ijabe.20211404.6404, July 2021.
[4] Daisuke Hatanaka, Alireza Ahrary, Dennis Ludena, ‘Research on Soil moisture measurement using moisture sensor’, International
Congress on Advanced Applied Informatics, DOI: 10.1109/IIAI-AAI.2015.289, July 2015.
[5] Saima Ansari, Ratnadeep R. Deshmukh, ‘Estimation of Soil Moisture Content: A Review’, International Journal of Theoretical and
Applied Mechanics ISSN 0973-6085 Volume 12, Number 3(2017) pp. 571-577.
[6] AlmawAyeleAniley, Naveen Kumar S.K, Akshaya Kumar A, ‘Soil moisture sensors in agriculture and the possible application of
nanomaterials in soil moisture sensors fabrication’, International Journal of Advanced Engineering Research and Technology (IJAERT)
Volume 6 Issue 1, January 2018, ISS No: 2348-8190.
[7] Anzhen Qin, Dongfeng Ning, Zhandong Liu, and Aiwang Duan, ‘Analysis of the Accuracy of an FDR Sensor in Soil Moisture
measurement under laboratory and field conditions’, Journal of Sensors, Volume 2021, DOI: 10.1155/2021/6665829.
[8] Limin Yu, Sha Tao, Yanzhao Ren, Wanlin Gao, Xinliang Liu, Yongkang Hu, and Redmond R.Shamshiri, ‘Comprehensive evaluation of
soil moisture sensing technology applications based on Analytic Hierarchy process’, MDPI, DOI: 10.3390/agriculture11111116,
November 2021.
[9] Anusha R, Polepally Vamshi Krishna, Harsh Jain, Dosada Suresh, ‘Analysis of soil moisture data using IOT and KNN algorithm’, Journal
of Engineering Sciences, ISSN NO: 0377-9254, November-2020.
[10] Kumar A, Kamal K, Arshad MO, Mathavan S, Vadamala T., ‘Smart irrigation using low-cost moisture sensors and XBee-based
communication’. IEEE Global Humanitarian Technology Conference; 2014. pp. 333–7.
[11] Mathurkar SS, Lanjewar RB, Patel NR, Somkuwar RS. ‘Smart sensor-based monitoring system for agriculture using FPGA’.
International Conference on Circuit, Power and Computing Technologies [ICCPCT]; Nagercoil. 2014. pp. 339–44.
[12] Jorapur N, Palaparthy VS, Sarik S, John J, ShojaeiBaghini M, Ananthasuresh GK. ‘A low-power, low-cost soil-mois¬ture sensor using
dual-probe heat-pulse technique’. Sensors and Actuators A: Physical, DOI: 10.1016/j.sna.2015.06.026
[13] Kumar, Jitendra & Rani, Alka & Kumar, Nirmal & Sinha, Nishant. (2022). Machine learning for soil moisture assessment.
10.1016/B978-0-323-85214-2.00001-X.
[14] N. Jarray, A. B. Abbes and I. R. Farah, "An Evaluation of Soil Moisture Retrieval Using Machine Learning Methods: Application in
Arid Regions of Tunisia," 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, Brussels, Belgium, 2021, pp.
6331-6334, doi: 10.1109/IGARSS47720.2021.9554585.
[15] Gagandeep Lamani, T M Gagan, Shankaraling Halemani, Fayaz K, Meenaxi M Raikar, “Tomato Leaf Disease Detection using Federated
Learning”, 2024 IEEE International Conference for Intelligent Technologies.
[16] Meenaxi M Raikar, Tejas G V, Kishorkumar T, Dhanaraj K, Rohit Reddy, “Tomato detection and counting using deep learning models”,
IEEE, International Technology Conference (OTCON 3.0) on Smart Computing for Innovation and Advancement in Industry 4.0, 2024.