APPLE PESTICIDE DETECTION AND ALERT

SYSTEM
Swetha Amireddy Kuruva Vaishnavi
Mrs. Ch. Anusha
dept. of ECE dept. of ECE
dept. of ECE
GNITS GNITS
GNITS
Hyderabad,India Hyderabad,India
Hyderabad, India
[email protected] [email protected]
[email protected]
Vulli Varshini
Pantham Apuruva
dept. of ECE
dept. of ECE
GNITS
GNITS
Hyderabad, India
Hyderabad, India
[email protected]
[email protected]

Abstract: Many modern techniques were developed to

produce more quantity of food for growing population. Now a II. LITERATURE SURVEY
day fruits and vegetables have become the major source of
nutrients and energy. Many chemicals are used in the A. Title 1 : Artificial Intelligence and IoT based detection
production of fruits and vegetables, which are dangerous for of pesticide in organic fruits and vegetables.
consumers. To identify pesticides in organic fruits and The paper discusses the experimental methodology,
vegetables, it is necessary to build low-cost, portable, sensitive,
sensor calibration techniques, and data processing algorithms
and selective biosensing platform. Many sensors like nano-
employed to ensure the reliability and accuracy of the
sensors, pH sensors, humidity and temperature sensors and gas
sensors are used to detect the pesticides on the fruits and detection system. Through comprehensive field trials and
vegetables. The input data is taken from sensors and send to the validation studies, the effectiveness of the IoT
microcontroller where the program is written in such a way
that it calculates the amount of pesticide present on fruits and B. Title-2: Identification of Chemicals in Fruits and
display theoutput. Vegetables using IOT
Focus is on the detection of leaf diseases by calculating
The software for this project is written in Arduino c
leaf area through pixel number statistics, aiming to increase
program. The safe value of pesticides in fruits or vegetables
throughput and reduce subjectivity arising from human
which can be consumed by humans and animals are mentioned
in the Arduino c program. If the pesticide content is above or
experts. Additionally, a proposed system involves image
below the threshold level then the sample contains pesticides. acquisition, pre-processing, feature extraction, artificial
Developing an alert system that triggers messages when neural network-based training, classification, and diagnosis
pesticide levels surpass predefined thresholds. The message is for disease identification and treatment
sent through proper communication channels such as SMS to
the stakeholders such as farmers, distributors, retailers, and III. PROPOSED SYSTEM
consumers by using GSMmodem.
To develop a system for detecting pesticides in apples
Keywords: Pesticides, nano-sensor, pH sensor, GSM and creating an alert message, the process involves collecting
Modem, SMS, Arduino C a dataset of images of apples with and without pesticide
residue, preprocessing the images to enhance features,
extracting features using techniques like Histogram of
I. INTRODUCTION
Oriented Gradients (HOG) or Convolutional Neural
The population is growing very fast but the land for Networks (CNNs), training a machine learning model (e.g.,
growing food is same. Hence many technologies are SVM, Random Forests, or CNNs) on the extracted features,
developed to produce more food for growing population. implementing a detection algorithm using the trained model,
Chemical fertilizers, insecticides and pesticides are used to generating an alert message upon detection of pesticide
enhance the yield production. But consuming these residue, testing and validating the system, deploying it in a
pesticides and chemical though food can lead to many real-world setting, and monitoring and maintaining its
dangerous diseases like cancer and even death in worst Performance. [1] For the detection of pesticides in apples, a
conditions. But many are adopting the organic forming, system can be developed that utilizes image processing and
integrated forming system and many newer technologies to machine learning techniques. Initially, images of apples are
improve the fruits and vegetables quality. The fruits and captured using digital cameras or smartphones. These images
vegetables grown using the organic methods coast double the undergo preprocessing, where they are resized, filtered, and
amount of the normal vegetables and fruits. Producers selling noise is removed to enhance their quality, the model is used
fertilized fruits and vegetables by labelling as organic fruits as part of a detection algorithm that analyses the features of
and vegetables. This can be dangerous to the consumers. new images to determine the presence of pesticide residues.
Detection of the pesticides on fruits and vegetable plays a When residues are detected, an alert message is generated,
major role in agriculture sectors. indicating the presence of pesticides and potential health
risks. This system aims to automate and improve the
accuracy of pesticide detection in apples, thereby enhancing
food safety and consumer trust.
 IV. BLOCK DIAGRAM calibrate the sensors. The system could include storage
capabilities to log data for future analysis or reference.
Power supply provides regulated power to the entire system,
ensuring stable and reliable operation.

V. WORKING PRINCIPLE
The process of detecting pesticides in fruits and
vegetables using sensors like LDR, IR, MQ2 gas sensor, an
LCD display, and an Arduino Uno involves a comprehensive
approach that integrates sensor data, processing, and display
for identification and indication of pesticide residues. Each
sensor in this setup plays a specific role: the LDR sensor
monitors variations in light intensity, the IR sensor measures
infrared radiation, and the MQ2 gas sensor detects gases,
including volatile organic compounds found in pesticides.
Fig. 1. Block Diagram To initiate the detection process, the sensors require
calibration to establish baseline values and sensitivity levels
The Block diagram of the IoT based detection of to specific compounds present in pesticides.[6]When
pesticides in apple and potato is shown in the below fig.1. exposed to fruits or vegetables suspected of pesticide
The input Sensors of the Arduino UNO are LDR Sensor, contamination, these sensors collect data based on changes in
MQ2 Sensor, IR Sensor and the output is LCD Display, light intensity, infrared radiation, and gas concentrations.
while sending alerts via the GSM module when necessary. The Arduino Uno, acting as the central processing unit,
LDR Sensor (Light Dependent Resistor) Detects changes in reads, processes, and interprets the sensor data using
light intensity. It measures variations caused by the presence predefined algorithms and thresholds to determine the
of pesticides affecting light transmission.IR Sensor used for presence of pesticides. Calibration ensures the accuracy and
detecting infrared radiation changes caused by the presence specificity of the sensors to differentiate pesticide-related
of pesticides. Gas Sensors, these sensors specifically detect variations from background environmental factors. The
pesticide gases in the environment. Each sensor's output may decision-making process within the Arduino determines if
require signal conditioning, such as amplification or filtering, detected variations or concentrations exceed predefined
to ensure accurate readings. Microcontroller/Processor, unit limits, indicating the potential presence of pesticides.
processes the analog signals from the sensors. Analog-to-
Digital Converter (ADC) Converts analog sensor signals into Results of the detection process are then displayed on
digital data for the processor to handle. The processor the LCD screen, providing a visual output that indicates the
executes algorithms to interpret the sensor data. It could presence or absence of pesticides and potentially triggers alerts
involvepattern recognition, threshold analysis, or comparison if pesticide levels surpass acceptable limits. Challenges in this
against predefined values to identify the presence and process include ensuring sensor sensitivity and specificity,
concentration of pesticides. [2] accurate threshold setting to avoid false positives or negatives,
and accounting for real-world variations in environmental
The IR, MQ2 gas sensor, and LDR sensors interface with conditions and pesticide compositions. [7]7While this setup
the Arduino microcontroller. Analog outputs from these offers a fundamental model for pesticide detection, practical
sensors are connected to analog pins on the Arduino. The implementation often requires more sophisticated and
Arduino processes the analog signals from the sensors and specialized equipment for increased accuracy and reliability
converts them into meaningful data related to pesticide in real-world scenarios. Calibration precision, threshold
levels. The Arduino code includes calibration, threshold accuracy, and consideration of environmental factors remain
setting, and alert generation logic. crucial for effective pesticide detection using sensor-based
The Arduino sends the processed data to the LCD display systems.
for real-time visualization of pesticide levels. The LCD Here's a detailed breakdown of the process:
updates as the sensor readings change. The Arduino is
connected toa GSM module, which is responsible for A. Sensor Integration and Functionality:
sending alert messages when pesticide levels exceed
predefined thresholds. [4]The GSM module uses SMS (Short 1) LDR Sensor (Light Dependent Resistor):
Message Service) for communication. The alert system is LDRs measure light intensity. When fruits or
triggered by the Arduino when sensor readings surpass the vegetables are contaminated with certain pesticides, they
predefined thresholds. The alert system activates the GSM might exhibit altered light absorption or reflection properties
module to send alert messages. due to chemical changes. The LDR sensor detects these
variations in light intensity.
Based on the processed data, the processor makes
decisions about the presence or absence of pesticides and 2) IR Sensor (Infrared Sensor):
their concentration levels. [5] LCD displays the output IR sensors detect infrared radiation. Some
information such as pesticide presence, concentration levels, pesticides might have distinctive infrared absorption or
and potentially other relevant data for the user. Visual or emission patterns. The IR sensor is used to measure these
audible indicators can alert users to the presence of pesticides specific changes in infrared radiation caused by pesticide
beyond certain thresholds. Interfaces like buttons or switches residues on the produce.
can allow users to interact with the system, set thresholds, or
 3) MQ2 Gas Sensor: processing implemented by the onboard processor, the
The MQ2 sensor is capable of detecting various gases, system offers both qualitative and quantitative analysis of
including volatile organic compounds (VOCs) that are pesticide presence and concentration levels
emitted by certain pesticides. It identifies the presence of
these gases in the vicinity of the fruits or vegetables.

B. Arduino Uno Processing:

1) Data Interpretation:
Arduino Uno processes and interprets the collected
sensor data. It applies predefined algorithms or thresholds to
analyse the readings from the sensors.
2) Data Acquisition:
The sensor data, including readings from LDR, IR, and MQ2
sensors, arecollected by the Arduino Uno microcontroller.
3) Decision Making and Output Display:
Determine the likelihood of pesticide contamination
based on the changes detected in light intensity, infrared Fig. 2. Detection of pesticide and making an alert message
radiation, and gas concentrations.
Algorithmic analysis, real-time processing capabilities,
4) Threshold Comparison: and the intuitive LCD display ensure accurate and immediate
Arduino Uno compares the sensor readings with feedback to users, detailing pesticide levels and potential
predefined thresholds or established patterns that indicate the alerts as shown in fig.3.2. Evaluating sensor sensitivity,
presence of pesticides. If the readings exceed these compensating for environmental interferences, and exploring
thresholds, it signifies potential pesticide contamination. avenues for future improvements form critical aspects of this
innovative pesticide detection project. [9]The outputs of the
5) LCD Display Output: pesticide content monitoring system with IR, MQ2 gas
The results of the pesticide detection process are sensor, LDR sensors, LCD display, and GSM integration are
displayed on an LCD screen. This display visually indicates the actionable information and alerts generated by the
the presence or absence of pesticides in the fruits or vegetables system.
based on the interpretations made by the Arduino Uno.
6) Calibration and Sensitivity: A. LCD Display Output:
Proper calibration of sensors is crucial to ensure they The LCD display provides a visual representation of real-
are sensitive enough to detect specific changes caused by time pesticide levels. It enables stakeholders to monitor the
pesticides without generating false positives or negatives. current status and changes in pesticide concentrations.
7) Threshold Setting: The LCD display, as a crucial output component, serves
Establishing accurate thresholds or criteria based on as a direct interface between the pesticide content monitoring
sensor readings to distinguish between contaminated and system and its stakeholders, providing them with the
uncontaminated produce. necessary information for decision-making and proactive
response to changes in pesticide levels.
8) Environmental Factors:
Considering environmental conditions that might
influence sensor readings, such as temperature, humidity,
and external interference, to ensure accuratedetection.
This system provides a basic yet effective method for
detecting potential pesticide residues in fruits and vegetables
by analysing variations in light intensity, infrared radiation,
and gas emissions associated with pesticide presence.
Calibration, [1]precise threshold setting, and accounting for
environmental factors are essential for reliable and accurate
pesticide detection using this sensor-based approach.

VI. RESULTS
This project aims to revolutionize pesticide detection by
integrating Light Dependent Resistor (LDR), Infrared (IR),
and gas sensors into a comprehensive detection system. The Fig. 3. Displaying pesticide value on LCD
combined sensor array offers a multi-dimensional approach
to identifying pesticides in varying environmental B. Alert Messages via GSM:
conditions. The LDR sensor captures changes in light
transmission caused by pesticide particles, while the IR The message initiation starts when pesticide levels
sensor detects alterations in infrared radiation due to pesticide exceed predefined thresholds, the system triggers an alert.
interactions. Gas sensors add specificity by detecting The GSM module sends alert messages via SMS to
particular pesticide gases. Through sophisticated data
predefined recipients, such as farmers, distributors, or and policymakers will be essential to overcome these hurdles
regulatory authorities. and unlock the full potential of IoT in creating a sustainable
and secure future for agriculture.

VIII. FUTURE SCOPE

The future scope of pesticide detection in fruits and
vegetables is evolving rapidly due to technological
advancements, particularly in areas like IoT,
nanotechnology, spectroscopy, and data analytics. This
evolution is driven by the growing concerns about food
safety, environmental sustainability, and the need for more
efficient agricultural practices.

A. Enhanced Detection Techniques:

Explore advanced image processing and machine
Fig. 4. Sending Alert Message learning techniques to improve the accuracy and efficiency
of pesticide residue detection. This could include deep
learning approaches, such as convolutional neural networks
(CNNs), which have shown great potential in image analysis
tasks.

B. Real-Time Monitoring:
Develop the system further to enable real-time
monitoring of pesticide residues in apples. This could
involve integrating the system with IoT devices or drones for
automated and continuous monitoring of orchards.

C. Mobile Application:
Create a mobile application that allows users to capture
images of apples and instantly receive alerts about the
presence of pesticide residues. This would enable consumers
Fig. 5. Alert Message Acknowledgement
to make informed decisions when purchasing apples.
By employing the GSM module and SMS alerts, the
pesticide content monitoring system ensures rapid and D. Expansion to Other Fruits:
reliable communication, enabling stakeholders to take timely Extend the system to detect pesticide residues in other
actions in response to elevated pesticide levels, ultimately fruits and vegetables. This would broaden its applicability
contributing to improved food safety and agricultural and impact on food safety.
practices.
E. Data Sharing and Collaboration:
VII. CONCLUSION Establish a platform for sharing data and collaborating
The implementation of IoT-based detection systems for with researchers, farmers, and regulatory bodies to improve
pesticides in apple and potato cultivation presents a pesticide management practices and food safety standards.
significant leap forward in ensuring food safety,
environmental sustainability, and efficient agricultural F. sensor Integration:
practices. This technology offers real-time monitoring and Explore the integration of sensors capable of detecting
data-driven insights, enabling farmers and stakeholders to pesticide residues directly on apples. This could provide a
make informed decisions regarding pest management. By more direct and efficient method of detection compared to
leveraging sensor networks and data analytics, the system not image-based approaches.
only detects pesticide levels but also provides a
comprehensive understanding of environmental conditions, G. Multi-Spectral Imaging:
crop health, and potential risks. [1]The integration of IoT in
pesticide detection aligns with the global push for smart Investigate the use of multi-spectral imaging techniques
agriculture, where technology plays a pivotal role in to capture more detailed information about the chemical
addressing the challenges faced by the agricultural sector. As composition of apples, including the presence of pesticide
we move towards a future of interconnected devices and residues. This could lead to more precise and reliable
smart farming practices, the IoT-based detection of detection.
pesticides in apple and potato cultivation stands as a
testament to the potential of technology to revolutionize and H. Automation and Robotics:
optimize agricultural processes. However, it is crucial to Develop automated systems or robots for inspecting
address challenges such as data security, standardization, and apples in orchards or processing facilities. These systems
accessibility to ensure the widespread adoption of these could use AI and machine learning to detect pesticide
technologies. Collaborative efforts from researchers, farmers, residues and sort apples accordingly.
 IX. REFERENCES
[1] D. Nagajyothi, "IOT Based prediction of
pesticides and fruits diseases detection using SVM
and Neural Networks," International Journel of
Mechanical Engineering, Jan 22.
[2] S. K. V. Selvi, "Organic Farming Technology for
Envirnment-friendly agriculture," International
Journal Conference onn advances in Engineering,
Science and Management, 2012.
[3] H. A. B. Mnif w, "Effect of endocrine Disruptor
pesticides: a review," Int J Enviro Res Public
Health.
[4] Sharma, "Worldwide pesticide usage and its
impacts on ecosystem," SN Appl Sci1:1446, 2019.
[5] N. ST, "Review of pesticide residueanalysis in
fruits and vegetables.Pre treatment analysis and
detection techniques," Food ResInt jul
133:109141, 2020.
[6] Archana, "Different methods to determine
pesticide residue in Fruits and Vegetables," IRJET,
April 2021.
[7] Aradhana, "Quality and pesticide detection in
fruits and vegetables," JETIR, 2021.
[8] V. K. Sujitha, "An implementation of soft
computing approach of smart control for induction
motor using ANFIS," 4th International
Conference on Smart Systems and Inventative
Technology, 2022.
[9] S. Gnanavel, "Quality detection of fresh fruits and
vegetables to improve Horticulture and Agro-
Industries," IEEE Xplore, 2017.
[10] V. Sujitha, "Analysis of regulated PV fed switched
reluctance motor drives using repression resistor
converter," International Journal of Engineering
and Technology, 2014.