Agri-Guard: IoT-Based Network for Agricultural Health Monitoring

with Fault Detection

Kushagra Singh, Kaﬁl Abbas Momin, M. Nishal, Chinmay Sultania and Madhav Rao

Dept. of Electronics and Communication Engineering,

International Institute of Information Technology Bangalore, Karnataka, India

Keywords:

Precision Agriculture, Gas Sensors, ESP8266, IoT Devices, Thermal Imaging, Sustainable Farming, Crop

Productivity.

Abstract:

Agricultural sector is increasingly adopting advanced technologies to enhance crop productivity and sustain-

ability. Precision agriculture leverages IoT devices, sensors, and data analytics to monitor and manage various

environmental parameters, addressing challenges such as global food demand, climate change, and resource

optimization. Previous research has demonstrated the efﬁcacy of wireless sensor networks (WSNs) and remote

sensing technologies in improving irrigation efﬁciency and early disease detection. However, these systems

often assume that all components continue to operate, thereby offering an incomplete view. This study presents

an advanced agricultural monitoring system referred to as Agri-Guard that integrates a wide array of sensors

to measure temperature, humidity, soil moisture, and gases like CO

, methane and ammonia. By utilizing an

ESP8266 microcontroller and IoT connectivity, the system ensures seamless data transmission and real-time

processing. Additionally, a centralized hub, equipped with a Raspberry Pi 5 and a thermal camera, enhances

the detection of crop anomalies, and an inoperative sensor hub. The sensor hub in the form of a cone is

optimally designed to detect environmental parameters besides being rainproof. The proposed Agri-Gaurd

setup clearly demonstrated the lack of manure and water from the sensors’ data, whereas thermal imaging

showcased the classiﬁcation of 92.7% between a dead and alive plant. The anomaly between an operating and

non-operating Agri-cone was found to be in complete agreement (100%). The proposed system represents a

signiﬁcant improvement over existing solutions, empowering farmers with precise data and faulty hub detec-

tion, leading to quick recovery and more sustainable farming practices.

1 INTRODUCTION

The agricultural sector has undergone a signiﬁcant

transformation in recent years, driven by the advent

of precision agriculture and the integration of ad-

vanced technologies (Condran et al., 2022), (Patil

et al., 2023), (Liakos et al., 2018), (Taghizadeh-

Mehrjardi et al., 2020), (Sharma et al., 2021). Preci-

sion agriculture involves the use of IoT devices, sen-

sors, and data analytics to monitor and manage criti-

cal agricultural parameters, aiming to optimize crop

yields and resource use while minimizing environ-

mental impact (Vitali et al., 2021), (Villa-Henriksen

et al., 2020), (Talavera et al., 2017), (Farooq et al.,

2019), (Naseer et al., 2024), (AlZubi and Galyna,

2023). This technological evolution has been cru-

cial in addressing the challenges posed by increasing

global food demand, climate change, and the need

for sustainable farming practices (Wu et al., 2010),

(Roux et al., 2018), (N

obrega et al., 2019), (Bru-

insma, 2009). Previous research has highlighted the

Thermal Camera

(FLIR C3-X)

Soil Sensor

(RS485-A)

Input Pins

Motor

Driver

Image from

Thermal Camera

ML Model

Micro-Controller

Centralised Camera stand

Controller and

preprocessing

Stepper Motor

Humidity Sensor

(DHT-11)

Soil Moisture

Sensor

Output Pins

Input Pins

Micro-Controller

Agri-Cone 1

Controller and

preprocessing

Gas Sensor (MQ-

9 & MQ-135)

Humidity Sensor

(DHT-11)

Soil Moisture

Sensor

Output Pins

Input Pins

Micro-Controller

Agri-Cone 2

Controller and

preprocessing

Gas Sensor (MQ-

9 & MQ-135)

To User's Device

Figure 1: Schematic showing the overview of Agri-Guard

and ﬂow of data.

potential of IoT-based solutions in enhancing agricul-

tural productivity. Wireless sensor networks (WSNs)

have been extensively utilized to monitor soil mois-

ture and temperature, providing real-time data that

Singh, K., Momin, K. A., Nishal, M., Sultania, C. and Rao, M.

Agri-Guard: IoT-Based Network for Agricultural Health Monitoring with Fault Detection.

DOI: 10.5220/0013209500003944

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 10th International Conference on Internet of Things, Big Data and Security (IoTBDS 2025), pages 223-230

ISBN: 978-989-758-750-4; ISSN: 2184-4976

223

helps farmers optimize irrigation schedules, thereby

conserving water and improving crop health (Patil

et al., 2023), (Marwa et al., 2020). Additionally, re-

mote sensing technologies, including thermal imag-

ing and multispectral cameras, have been employed

to detect early signs of crop diseases and stress (Pal-

lagani et al., 2019), (Barjaktarovic et al., 2024), (Wil-

son et al., 2023), (Cui et al., 2021), (Hassan et al.,

2021), (Fevgas et al., 2023), (Toscano et al., 2024),

(Zhu et al., 2023). These technologies allow for

timely interventions, reducing crop loss, maintain-

ing soil moisture, and improving overall yield (Bai

et al., 2019). However, these systems often op-

erate in isolation, focusing on speciﬁc parameters

without providing a holistic view of the agricul-

tural environment. Several related works have at-

tempted to bridge this gap by integrating multiple sen-

sors and data sources into a uniﬁed monitoring sys-

tem. For example, systems combining soil sensors

with weather stations have been developed to offer

comprehensive environmental monitoring, aiding in

more accurate decision-making (Marwa et al., 2020),

(Hashmi et al., 2024), (Shaikh et al., 2022a), (Ahmed

et al., 2024), (Lin et al., 2023), (Ibraiwish et al.,

2024), (Caruso et al., 2021a), (Jani and Chaubey,

2022). Another notable advancement is the use

of drone technology equipped with various sensors

to survey large agricultural ﬁelds, providing high-

resolution data on crop conditions and facilitating

precision farming (Caruso et al., 2021b), (Panjaitan

et al., 2022), (Mohyuddin et al., 2024), (Shaikh et al.,

2022b), (Reddy Maddikunta et al., 2021), (Mukhame-

diev et al., 2023), (Verma et al., 2020), (Jasim et al.,

2023). Despite these advancements, most of these

works do not cater to the on-ﬁeld problems where few

of the sensors are at fault leading to an incomprehen-

sive view. Hence besides on-ﬁeld sensory informa-

tion, the operative status of installed components is

equally important.

The current work seeks to build upon these foun-

dations by developing a more sophisticated and com-

prehensive agricultural monitoring system. In ad-

dition to the array of sensors that measure temper-

ature, humidity, soil moisture, and gases such as

, methane, and ammonia, this system also offers

a mechanism to detect whether the sensor hubs are

manipulated from the housed setup. In general, the

proposed two-level Physical-security-enabled sensing

provides a holistic view of the agricultural environ-

ment. Advanced microcontrollers and IoT connectiv-

ity ensure that data from these sensors is seamlessly

transmitted and processed, offering real-time insights

and alerts to farmers. The incorporation of thermal

imaging enhances the system’s capability to not only

detect crop anomalies such as pest infestations and

diseases at an early stage but also alert any faulty sen-

sory hubs installed. This work improves upon pre-

vious systems by offering a fully integrated solution

that not only monitors a wide range of environmental

parameters but also processes and analyzes the data

to provide meaningful insights. The use of a central-

ized data processing hub allows for the aggregation

and analysis of data from multiple sensors, ensuring

that farmers receive comprehensive and actionable in-

formation. Furthermore, the automation of data col-

lection and analysis reduces the need for manual in-

tervention, allowing farmers to focus on other critical

tasks. The proposed system enables more precise and

timely interventions, ultimately leading to improved

crop yields and more sustainable farming practices.

(i)

(ii)

(iii)

(iv) (v)

(vi)

Figure 2: 3-D CAD ﬁles showing the, (i) overall structure

of the Agri-cone, (ii),(iii) the structure of the conical body

and (iv),(v),(vi) the hemispherical top.

2 PROPOSED AGRI-GUARD

DESIGN

Agri-Guard consists of two sets of devices: The IoT-

based Agri-cones and a Centralized Camera stand as

shown in Figure 1. The Agri-cones consist of an ar-

ray of sensors including temperature, humidity, mois-

ture, CO

and methane gas sensors. These host of

sensors provide crucial details about the plant health.

These sensors are interfaced using an ESP8266 mi-

crocontroller, which is Wi-Fi and Bluetooth enabled

to transfer these vital data to the hub of the IoT net-

work placed under the Centralized Camera stand. The

Centralized Camera stand receives data from all the

Agri-cones placed around it in its vicinity and relays

the data to the control unit, i.e. the user’s device. The

Centralized Camera stand is a height-adjustable metal

stand mounted with a thermal camera. This thermal

camera is allowed to rotate through two degrees of

freedom (DOF), allowing it a 360° view of its sur-

roundings. The base of the stand consists of an 8-in-1

IoTBDS 2025 - 10th International Conference on Internet of Things, Big Data and Security

224

(i) (ii)

Gas Sensor

(MQ-9)

Humidity Sensor

(DHT-11)

Soil Moisture

Sensor

Vent

Figure 3: The Agri-cone, (i) Bottom view and (ii) Front

view.

soil parameter sensor which enables it to monitor the

soil conductivity, pH, and levels of nitrogen, phospho-

rous and sodium (NPK) among others. The base also

consists of an IoT network hub that relays data from

all the Agri-cones in its vicinity to the base station for

efﬁcient monitoring. The thermal camera on top of

the stand serves two key purposes. First, it monitors

the health of the plants within its vicinity, detecting

whether any plants are showing signs of stress, have

died, or are under threat from parasitic attacks. This

allows the system to issue timely warnings, enabling

users to take preventive or corrective actions to main-

tain plant health.

In addition, the activity of the sensors within each

Agri-cone is continuously monitored to predict any

anomalies or malfunctions. If sensor data indicates an

issue, the thermal camera is used to conﬁrm the prob-

lem by visually checking the operational status of the

Agri-cone. This combination ensures that both plant

health and device functionality are accurately moni-

tored, providing a more reliable and comprehensive

system for agricultural management.

2.1 Agri-Cones

The Agri-cone features a conical body topped with a

hemispherical cover (Figure 2), combining aesthetic

design with functional utility. The cone-shaped body

ensures stable and secure placement in the soil, with a

wide base that helps anchor it ﬁrmly even in varying

soil conditions. The hemispherical top cover serves

as a critical element, providing dual functionalities:

protection and gas distribution. This cover shields the

internal components from rain, dust, and other envi-

ronmental elements, thereby enhancing the durabil-

ity and longevity of the device. Additionally, its de-

sign allows gases emanating from the soil to spread

evenly to the gas sensors mounted beneath it and es-

cape from the vent present at the top, ensuring accu-

rate gas readings and continuous ﬂow of gases (Fig-

ure 3). The entire structure of the Agri-cone is fab-

ricated from PLA Generic White material and 3D

printed using the Ultimaker 3 Extended, ensuring pre-

cision and consistency in the build. To further protect

the device, the entire structure is sealed and coated

with a water-repellent layer. This coating prevents

moisture ingress, safeguarding the internal electron-

ics from potential damage caused by water seepage.

At the heart of the Agri-cone is the ESP8266 micro-

controller, equipped with both Wi-Fi and Bluetooth

capabilities, allowing for seamless data transfer to the

IoT network hub located at the base of the Central-

ized Camera Stand. Inside the cone, the microcon-

troller is accompanied by a battery and sensors for

measuring soil moisture and temperature. These sen-

sors are housed within the conical body, providing

crucial data on the soil’s physical conditions. The

moisture sensor detects the water content in the soil,

which is vital for maintaining optimal plant hydration

levels. The temperature sensor monitors the soil tem-

perature, ensuring that it remains within a suitable

range for plant growth. Under the hood of the top

cover, the gas sensors and humidity sensors are strate-

gically placed. The gas sensors are designed to de-

tect key gases such as carbon dioxide (CO

), methane,

and ammonia, which are indicative of soil respiration,

microbial activity, and nutrient cycles. Monitoring

these gases provides insights into the biological and

chemical processes occurring within the soil. Ammo-

nia levels, in particular, can indicate the presence of

nitrogen-ﬁxing bacteria and the decomposition of or-

ganic matter. The DHT-11 humidity sensors, on the

other hand, measure the air moisture levels around the

plants, which is crucial for preventing diseases and

ensuring healthy plant growth.

(i) (ii)

(iii) (iv)

Thermal camera

Motor

Motor driver

Metallic stand

Micro-controller

Geared shaft

Figure 4: Structure of the Centralised Camera stand along

with the various components mounted on it.

Agri-Guard: IoT-Based Network for Agricultural Health Monitoring with Fault Detection

225

2.2 Centralised Camera Stand

The Centralised Camera Stand, an essential part of

the Agri-Guard system, is designed to integrate and

process data from multiple sources for comprehen-

sive ﬁeld monitoring. It features a height-adjustable

metallic stand, ensuring stability and durability in var-

ious environmental conditions. At its base, a Rasp-

berry Pi 5 acts as the central hub, receiving data from

all neighbouring Agri-cones via the IoT network. The

Raspberry Pi processes this data, ﬂagging any irreg-

ularities and transforming it into meaningful infor-

mation for the user. Mounted atop the stand is a

PLA Generic White structure housing the FLIR C3-

X Compact thermal camera. This camera, capable of

detecting thermal anomalies such as pest infestations

and crop diseases, interfaces with the Raspberry Pi

to send thermal images to the user’s device for de-

tailed analysis (Figure 5). To achieve a comprehen-

sive 360° ﬁeld view, the thermal camera is mounted

on a motorized platform that allows it to rotate freely

along two degrees of freedom (DOF) as shown in Fig-

ure 4. Programmed to survey the ﬁeld every six hours,

the camera captures 12 images per rotation, each cov-

ering a 30-degree segment, ensuring thorough mon-

itoring. Embedded in the base of the stand is a 7-

in-1 RS485 JXCT soil sensor, providing critical pa-

rameters such as pH, electrical conductivity (EC), and

NPK levels. These parameters are vital for long-term

soil health monitoring. The integration of these ad-

vanced technologies in the Centralized Camera Stand

enables efﬁcient data management and provides farm-

ers with actionable insights, enhancing productivity

and sustainability in agricultural practices.

(i)

(iii)

(ii) (iv)

Figure 5: Images obtained by the FLIR C3-X Thermal

Camera mounted on the Stand along with their original

color image.

2.3 Advantages of Wi-Fi

Communication

In this experimental setup, Wi-Fi was chosen as the

primary communication protocol for several reasons:

Higher Data Throughput: Wi-Fi provides the

bandwidth necessary for transmitting real-time sensor

data, including high-resolution thermal images from

the centralized stand, which is not feasible with low-

power, long-range alternatives like LoRa or Zigbee.

Ease of Integration: The ESP8266 Wi-Fi module

is easy to set up and integrates seamlessly with cloud

platforms for remote data access. This simpliﬁes de-

velopment and allows for real-time data monitoring.

Sufﬁcient Range for Small Deployments: For

the current setup, the 5-meter communication radius

is well within Wi-Fi’s range, ensuring reliable con-

nectivity between Agri-cones and the centralized hub

without requiring additional infrastructure.

Experimental Efﬁciency: Wi-Fi enables rapid

prototyping and testing, ideal for laboratory research

and small-scale ﬁeld trials, where quick setup and

easy access to data are more important than long-

range communication.

(i)

(ii)

Figure 6: (i) Thermal image of a dead plant alongside an

alive plant and (ii) the corresponding color image.

2.4 System Architecture and

Communication

This system is speciﬁcally designed for laboratory re-

search and experimental purposes, as well as small-

scale ﬁeld deployments. The current conﬁguration,

with two Agri-cones per square meter and a 5-meter

IoTBDS 2025 - 10th International Conference on Internet of Things, Big Data and Security

226

communication radius from the centralized hub, al-

lows for precise environmental monitoring in smaller

areas. The setup enables researchers to validate the

system’s performance before scaling it to larger ﬁelds

or more complex deployments. The use of Wi-Fi

provides easy integration for real-time data collection

and simpliﬁes the testing process in these controlled

settings.

Sensor Data Collection: Each Agri-cone collects

environmental data from its attached sensors. The

DHT11 measures humidity, the soil moisture sensor

detects water levels, and the MQ9 and MQ135 gas

sensors monitor CO

, methane, and ammonia levels.

The ESP8266 microcontroller processes the sensor

readings locally, performing initial ﬁltering and ag-

gregation.

Data Transmission: Using Wi-Fi, the ESP8266

sends the pre-processed data to the Centralized Cam-

era Stand at intervals. The communication occurs

within a 5-meter radius, ensuring reliable transmis-

sion from all Agri-cones in the area. This distance is

ideal for small-scale experiments and offers sufﬁcient

coverage for small agricultural plots or lab setups.

Centralized Processing: The Raspberry Pi 5 at

the stand collects and processes the incoming data

from all Agri-cones in the network. It aggregates sen-

sor readings, performs anomaly detection (e.g., de-

tecting abnormal gas levels), and integrates the ther-

mal camera’s data to monitor plant health. The ther-

mal camera captures infrared images, helping to de-

tect unhealthy plants or non-functional Agri-cones.

Data Visualization: The processed data is dis-

played in real-time on a custom dashboard, allowing

the user to monitor environmental conditions and re-

ceive alerts for anomalies. This system supports real-

time decision-making, vital for applications like pre-

cision farming or experimental trials.

3 DATA COLLECTION AND

RESULTS

3.1 Agri-Cones

The Agri-Guard system’s sensors provide environ-

mental data, including temperature, soil moisture, hu-

midity, and gas concentrations (CO

, CH

, & NH

These readings are plotted on a graph to visualize

changes over time (Figure 8). Upon adding manure

to the soil, notable changes in sensor readings are ob-

served. The gas sensor MQ9, which detects CO

and

methane, shows a signiﬁcant increase due to the or-

ganic matter decomposition in the manure. Concur-

(i)

(iii)

(ii)

Figure 7: Thermal image of the Agri-Cone when it is (i)

switched off, (ii) switched on and (iii) the corresponding

original color image.

rently, the added water raises the humidity and soil

moisture levels, which is reﬂected in their respective

sensor readings. Similarly, over time, as microbial de-

cay progresses, the MQ135 sensor, which measures

ammonia levels, also shows a slight increase, indicat-

ing the breakdown of organic nitrogen compounds in

the manure.These dynamic changes are plotted and

compared with those from the control setup (Figure

8), where the control lacks manure addition, unlike

the experimental setup.

3.2 Centralized Camera Stand

The Agri-Guard system employs advanced thermal

imaging technology to monitor plant health and the

functionality of Agri-cones. The thermal images cap-

tured by the FLIR C3-X Compact thermal camera

provide a detailed heatmap of the monitored area,

which is then analyzed to determine the status of

plants and Agri-cones. This section explains the pro-

cess and effectiveness of using thermal imaging in

conjunction with a classiﬁcation model to achieve ac-

curate monitoring results. The thermal images ob-

tained from the camera display variations in tempera-

ture across the monitored ﬁeld. Living plants exhibit

distinct thermal heatmaps characterized by larger ar-

eas of red, indicating higher temperatures due to ac-

tive biological processes such as photosynthesis and

respiration. These processes generate heat, which is

captured by the thermal camera, resulting in a promi-

nent red coloration on the heatmap (Figure 6). Con-

versely, dead plants lack these active processes, lead-

ing to a cooler temperature proﬁle that appears in the

Agri-Guard: IoT-Based Network for Agricultural Health Monitoring with Fault Detection

227

Figure 8: Agri-cone sensor measurements over time showing graphs before addition of manure and water (Control), and after

adding the organic manure and water (Experimental).

lower spectrum of the heatmap, typically represented

by blue and green colors. To classify whether a plant

is dead or alive, the thermal images are fed into a

classiﬁcation model. Numerous binary classiﬁcation

models were evaluated for this task, including logis-

tic regression, support vector machines (SVM), and

Decision trees. However, it was found that a region-

based Convolutional Neural Network (R-CNN) pro-

vided the best performance. The R-CNN model was

particularly effective due to its ability to analyze spe-

ciﬁc regions within the thermal images, focusing on

areas of interest and providing a more accurate clas-

siﬁcation. The R-CNN model achieved an overall

accuracy of 92.7%, signiﬁcantly outperforming other

models investigated. The accuracy of all the models

investigated is listed in Table 1. In addition to plant

Table 1: Accuracy of models investigated for classifying

between dead and alive plant for a thermal image.

Model Name Accuracy

R-CNN 92.7%

Decision Tree 89.9%

SVM 82.4%

Logistic Regression 80.1%

health monitoring, the thermal images are also used to

assess the functionality of the Agri-cones. Functional

Agri-cones generate heat due to the active circuitry

inside, which is reﬂected in their thermal heatmap

as a higher concentration of red and green colors.

These colors indicate the heat produced by the elec-

tronic components in operation. On the other hand,

Agri-cones that are switched off or malfunctioning

show a lack of heat generation, appearing in the lower

spectrum of the thermal heatmap with cooler colors

such as blue and green (Figure 7). The classiﬁcation

model used for Agri-cone functionality also lever-

ages the region-based CNN approach giving 100%

classiﬁcation accuracy. This model effectively dis-

tinguishes between operational and non-operational

Agri-cones by analyzing the thermal signature of each

device. The higher concentration of warm colors in

the heatmap corresponds to active, powered-on Agri-

cones, while the absence of such colors indicates a

non-functional state.

3.3 Power Management and Battery

Life Analysis

The Agri-cone system is powered by two 3.7V,

4000mAh Li-Po batteries in series, providing a total

of 7.4V with a combined capacity of 4000mAh. The

following section provides a detailed analysis of the

power consumption and battery life based on the com-

ponents used and the operating conditions. The active

components of each Agri-cone include the Wemos D1

Mini (ESP8266), DHT11 humidity sensor, soil mois-

ture sensor, and gas sensors (MQ9, MQ135). The We-

mos D1 Mini ESP8266 is characterized for operating

current of 170 mA at 3.3V. Hence for a higher volt-

age of 7.4V, equivalent current of 75.8mA is drawn.

IoTBDS 2025 - 10th International Conference on Internet of Things, Big Data and Security

228

The other components including DHT11 Humidity

Sensor, Soil Moisture Sensor, MQ9 Gas Sensor, and

MQ135 Gas Sensor draws current of 0.3mA, 35mA,

150mA, and 56mA respectively. The total current

consumption during active operation is estimated to

be 317.1 mA. The total power consumption in active

mode is estimated to be 2.3475 W.

With a 4000mAh battery, the device would last

approximately for Battery Life (Active) =

4000mAh

317.1mA

≈

12.61hours To extend the battery life, a duty cycle

is implemented where the device remains active for

10% of the time and in deep sleep (where power con-

sumption is minimal) for the remaining 90%, where

the current drawn in sleep state is 0.02mA. The aver-

age time weighted current drawn by the device is thus

computed as per the Equation 1.

Average Current =

(0.1 × 317.1 mA)+

(0.9 × 0.02 mA) = 31.73 mA

(1)

The corresponding power consumption with the

duty cycle is: P

avg

= 31.73 mA × 7.4V = 0.2348 W

The battery life with deep sleep mode enabled is thus

given by: Battery Life (Sleep Mode) ≈

4000mAh

31.73mA

≈

126hours This battery optimization allows the sys-

tem to operate for approximately 126 hours (over 5

days) between charging, making it ideal for small-

scale ﬁeld experiments where continuous monitoring

is required without frequent battery changes.

4 CONCLUSION

The Agri-Guard system represents a signiﬁcant leap

in precision agriculture by combining thermal imag-

ing with R-CNN classiﬁcation to provide accurate,

real-time monitoring of plant health and the op-

erational status of agricultural equipment. Lever-

aging the FLIR C3-X Compact thermal camera,

Agri-Guard effectively differentiates between healthy

and distressed plants through distinct thermal signa-

tures, while identifying functional and non-functional

equipment based on heat patterns. Its modular design,

optimized power management, and Wi-Fi-enabled

data transmission to enhance adaptability for labora-

tory and ﬁeld applications.

Future advancements will improve scalability, re-

ﬁne long-term deployment strategies, and integrate

additional functionalities, enabling broader applica-

tions across diverse agricultural contexts. By ad-

dressing these areas, Agri-Guard has the potential to

drive substantial improvements in resource manage-

ment, crop yield monitoring, and sustainable agricul-

tural practices.

REFERENCES

Ahmed, A., Parveen, I., Abdullah, S., Ahmad, I., Alturki,

N., and Jamel, L. (2024). Optimized data fusion with

scheduled rest periods for enhanced smart agriculture

via blockchain integration. IEEE Access, 12:15171–

15193.

AlZubi, A. A. and Galyna, K. (2023). Artiﬁcial intelli-

gence and internet of things for sustainable farming

and smart agriculture. IEEE Access, 11:78686–78692.

Bai, X., Huang, Y., Ren, W., Coyne, M., Jacinthe, P.-A.,

Tao, B., Hui, D., Yang, J., and Matocha, C. (2019).

Responses of soil carbon sequestration to climate-

smart agriculture practices: A meta-analysis. Global

Change Biology, 25(8):2591–2606.

Barjaktarovic, M., Santoni, M., and Bruzzone, L. (2024).

Design and veriﬁcation of a low-cost multispectral

camera for precision agriculture application. IEEE

Journal of Selected Topics in Applied Earth Observa-

tions and Remote Sensing, 17:6945–6957.

Bruinsma, J. (2009). The resource outlook to 2050: by how

much do land, water use and crop yields need to in-

crease by 2050? fao expert meeting on how to feed

the world in 2050, 24–26 june.

Caruso, A., Chessa, S., Escolar, S., Barba, J., and L

opez,

J. C. (2021a). Collection of data with drones in preci-

sion agriculture: Analytical model and lora case study.

IEEE Internet of Things Journal, 8(22):16692–16704.

Caruso, A., Chessa, S., Escolar, S., Barba, J., and L

opez,

J. C. (2021b). Collection of data with drones in preci-

sion agriculture: Analytical model and lora case study.

IEEE Internet of Things Journal, 8(22):16692–16704.

Condran, S., Bewong, M., Islam, M. Z., Maphosa, L., and

Zheng, L. (2022). Machine learning in precision agri-

culture: A survey on trends, applications and eval-

uations over two decades. IEEE Access, 10:73786–

73803.

Cui, J., Liu, M., Zhang, Z., Yang, S., and Ning, J.

(2021). Robust uav thermal infrared remote sensing

images stitching via overlap-prior-based global simi-

larity prior model. IEEE Journal of Selected Topics

in Applied Earth Observations and Remote Sensing,

14:270–282.

Farooq, M. S., Riaz, S., Abid, A., Abid, K., and Naeem,

M. A. (2019). A survey on the role of iot in agricul-

ture for the implementation of smart farming. IEEE

Access, 7:156237–156271.

Fevgas, G., Lagkas, T., Argyriou, V., and Sarigiannidis, P.

(2023). Detection of biotic or abiotic stress in vine-

yards using thermal and rgb images captured via iot

sensors. IEEE Access, 11:105902–105915.

Hashmi, A. U. H., Mir, G. U., Sattar, K., Ullah, S. S., Al-

roobaea, R., Iqbal, J., and Hussain, S. (2024). Ef-

fects of iot communication protocols for precision

agriculture in outdoor environments. IEEE Access,

12:46410–46421.

Hassan, S. I., Alam, M. M., Illahi, U., Al Ghamdi, M. A.,

Almotiri, S. H., and Su’ud, M. M. (2021). A sys-

tematic review on monitoring and advanced control

Agri-Guard: IoT-Based Network for Agricultural Health Monitoring with Fault Detection

229

strategies in smart agriculture. IEEE Access, 9:32517–

32548.

Ibraiwish, H., Eltokhey, M. W., and Alouini, M.-S. (2024).

Uav-assisted vlc using led-based grow lights in pre-

cision agriculture systems. IEEE Internet of Things

Magazine, 7(3):100–105.

Jani, K. A. and Chaubey, N. K. (2022). A novel model

for optimization of resource utilization in smart agri-

culture system using iot (smaiot). IEEE Internet of

Things Journal, 9(13):11275–11282.

Jasim, A. N., Fourati, L. C., and Albahri, O. S. (2023). Eval-

uation of unmanned aerial vehicles for precision agri-

culture based on integrated fuzzy decision-making ap-

proach. IEEE Access, 11:75037–75062.

Liakos, K. G., Busato, P., Moshou, D., Pearson, S., and

Bochtis, D. (2018). Machine learning in agriculture:

A review. Sensors, 18(8).

Lin, Y.-B., Chen, W.-E., and Chang, T. C.-Y. (2023).

Moving from cloud to fog/edge: The smart agricul-

ture experience. IEEE Communications Magazine,

61(12):86–92.

Marwa, C., Othman, S. B., and Sakli, H. (2020). Iot based

low-cost weather station and monitoring system for

smart agriculture. In 2020 20th International Confer-

ence on Sciences and Techniques of Automatic Control

and Computer Engineering (STA), pages 349–354.

Mohyuddin, G., Khan, M. A., Haseeb, A., Mahpara, S.,

Waseem, M., and Saleh, A. M. (2024). Evaluation

of machine learning approaches for precision farming

in smart agriculture system: A comprehensive review.

IEEE Access, 12:60155–60184.

Mukhamediev, R. I., Yakunin, K., Aubakirov, M., As-

sanov, I., Kuchin, Y., Symagulov, A., Levashenko,

V., Zaitseva, E., Sokolov, D., and Amirgaliyev, Y.

(2023). Coverage path planning optimization of het-

erogeneous uavs group for precision agriculture. IEEE

Access, 11:5789–5803.

Naseer, A., Shmoon, M., Shakeel, T., Ur Rehman, S., Ah-

mad, A., and Gruhn, V. (2024). A systematic litera-

ture review of the iot in agriculture—global adoption,

innovations, security, and privacy challenges. IEEE

Access, 12:60986–61021.

obrega, L., Gonc¸alves, P., Pedreiras, P., and Pereira, J.

(2019). An iot-based solution for intelligent farming.

Sensors, 19(3).

Pallagani, V., Khandelwal, V., Chandra, B., Udutalapally,

V., Das, D., and P. Mohanty, S. (2019). dcrop: A deep-

learning based framework for accurate prediction of

diseases of crops in smart agriculture. In 2019 IEEE

International Symposium on Smart Electronic Systems

(iSES) (Formerly iNiS), pages 29–33.

Panjaitan, S. D., Dewi, Y. S. K., Hendri, M. I., Wicaksono,

R. A., and Priyatman, H. (2022). A drone technology

implementation approach to conventional paddy ﬁelds

application. IEEE Access, 10:120650–120658.

Patil, P., Kestur, R., Rao, M., and C†, A. (2023). Iot based

data sensing system for autogrow, an autonomous

greenhouse system for precision agriculture. In 2023

IEEE Applied Sensing Conference (APSCON), pages

1–3.

Reddy Maddikunta, P. K., Hakak, S., Alazab, M., Bhat-

tacharya, S., Gadekallu, T. R., Khan, W. Z., and Pham,

Q.-V. (2021). Unmanned aerial vehicles in smart agri-

culture: Applications, requirements, and challenges.

IEEE Sensors Journal, 21(16):17608–17619.

Roux, J., Escriba, C., Fourniols, J.-Y., and Soto-Romero, G.

(2018). A new bi-frequency soil smart sensing mois-

ture and salinity for connected sustainable agriculture.

Journal of Sensor Technology, 09.

Shaikh, F. K., Karim, S., Zeadally, S., and Nebhen, J.

(2022a). Recent trends in internet-of-things-enabled

sensor technologies for smart agriculture. IEEE Inter-

net of Things Journal, 9(23):23583–23598.

Shaikh, F. K., Karim, S., Zeadally, S., and Nebhen, J.

(2022b). Recent trends in internet-of-things-enabled

sensor technologies for smart agriculture. IEEE Inter-

net of Things Journal, 9(23):23583–23598.

Sharma, A., Jain, A., Gupta, P., and Chowdary, V. (2021).

Machine learning applications for precision agricul-

ture: A comprehensive review. IEEE Access, 9:4843–

4873.

Taghizadeh-Mehrjardi, R., Nabiollahi, K., Rasoli, L., Kerry,

R., and Scholten, T. (2020). Land suitability assess-

ment and agricultural production sustainability using

machine learning models. Agronomy, 10(4).

Talavera, J. M., Tob

on, L. E., G

omez, J. A., Culman, M. A.,

Aranda, J. M., Parra, D. T., Quiroz, L. A., Hoyos, A.,

and Garreta, L. E. (2017). Review of iot applications

in agro-industrial and environmental ﬁelds. Comput-

ers and Electronics in Agriculture, 142:283–297.

Toscano, F., Fiorentino, C., Capece, N., Erra, U., Travascia,

D., Scopa, A., Drosos, M., and D’Antonio, P. (2024).

Unmanned aerial vehicle for precision agriculture: A

review. IEEE Access, 12:69188–69205.

Verma, M., Lafarga, V., Dehaeze, T., and Collette, C.

(2020). Multi-degree of freedom isolation system with

high frequency roll-off for drone camera stabilization.

IEEE Access, 8:176188–176201.

Villa-Henriksen, A., Edwards, G. T., Pesonen, L. A., Green,

O., and Sørensen, C. A. G. (2020). Internet of

things in arable farming: Implementation, applica-

tions, challenges and potential. Biosystems Engineer-

ing, 191:60–84.

Vitali, G., Francia, M., Golfarelli, M., and Canavari, M.

(2021). Crop management with the iot: An interdisci-

plinary survey. Agronomy, 11(1).

Wilson, A. N., Gupta, K. A., Koduru, B. H., Kumar, A.,

Jha, A., and Cenkeramaddi, L. R. (2023). Recent ad-

vances in thermal imaging and its applications using

machine learning: A review. IEEE Sensors Journal,

23(4):3395–3407.

Wu, J., Ping, L., Ge, X., Wang, Y., and Fu, J. (2010). Cloud

storage as the infrastructure of cloud computing. In

2010 International Conference on Intelligent Comput-

ing and Cognitive Informatics, pages 380–383.

Zhu, D., Zhang, Y., Gao, Q., Lu, Y., and Sun, D. (2023).

Infrared and visible image fusion using threshold seg-

mentation and weight optimization. IEEE Sensors

Journal, 23(20):24970–24982.

IoTBDS 2025 - 10th International Conference on Internet of Things, Big Data and Security

230