Seminar Report

On
Pest Prediction in Rice using IoT and Feed Forward Neural
Network
Submitted in partial fulfillment for the award of the
degree of
BACHELOR OF TECHNOLOGY
In
COMPUTER SCIENCE AND ENGINEERING

Submitted by
Sam Kattiveettil James (AJC20CS121)
Under the guidance of
Ms. Tintu Alphonsa Thomas
Assistant Professor

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

AMAL JYOTHI COLLEGE OF ENGINEERING
(AUTONOMOUS)
KANJIRAPPALLY

Affiliated to APJ Abdul Kalam Technological University

December 2023
 DEPARTMENT OF COMPUTER SCIENCE &
ENGINEERING

AMAL JYOTHI COLLEGE OF ENGINEERING

(AUTONOMOUS)
KANJIRAPPALLY

CERTIFICATE

Certified that this seminar report entitled “Pest Prediction in Rice using IoT
and Feed Forward Neural Network” submitted by Sam Kattiveettil James
(AJC20CS121) in partial fulfillment for the award of the degree of Bachelor
of Technology in Computer Science and Engineering under APJ Abdul Kalam
Technological University during the year 2023-24, is the bonafide work carried
out by him under my guidance and supervision.

Tintu Alphonsa Thomas Dr. Indu John/ Krishnalal G/Manoj T Joy

Internal Guide Co-ordinators

Dr. Juby Mathew

Head of the Department
 ACKNOWLEDGEMENT

First of all, I sincerely thank the Almighty GOD who is most beneficent and merciful for
giving me the knowledge and courage to complete the seminar successfully.

I take this opportunity to express my sincere gratitude to all those who have helped me in
completing this seminar successfully.

I wish to express my sincere gratitude to our Manager Rev. Fr. Dr. Mathew Paikatt for
providing infrastructural facilities for us, our Principal Dr. Lillykutty Jacob for the kind co-
operation in all aspects of the seminar.

I owe a great depth of gratitude towards our Head of the Department Dr. Juby Mathew for
his support during the entire course of this seminar work.

I express my sincere and heartfelt gratitude to my seminar guide Ms.Tintu Alphonsa

Thomas for the constant encouragement and support throughout the seminar. I also express
my gratitude towards the supporting staff for their technical assistance.

I am indebted to my beloved teachers for their cooperation and suggestion throughout the
seminar which helped me a lot. I also thank all my friends and classmates for their interest,
dedication and encouragement shown towards the seminar. I convey my hearty thanks to my
parents for their moral support, suggestion and encouragement to make this venture a success.
 Abstract
This research paper presents a study on the use of IoT and Feed Forward Neural Network
for pest prediction in rice crops. The authors discuss the hardware and software components
of the system, including sensors, microcontrollers, and machine learning algorithms. They
also describe the deployment of the system in a rice field and the development of an Android
application for alerting farmers about potential pest attacks. The study shows that the proposed
approach is effective in predicting pest infestations based on environmental factors such as
temperature, humidity, and rainfall. The authors conclude that the implementation of IoT and
neural networks can benefit the agricultural industry by improving crop yield and reducing
losses due to pest infestations.
Contents

1 Introduction 1

2 Literature Survey 2
2.1 Internet of Things Based Weekly Crop Pest Prediction by Using Deep Neural
Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.2 Deep Learning model for early prediction of plant disease . . . . . . . . . . . . 2
2.3 Towards Precision Agriculture: IoT-enabled Intelligent Irrigation Systems Us-
ing Deep Learning Neural Network . . . . . . . . . . . . . . . . . . . . . . . 3
2.4 Crop Yield Prediction using Machine Learning Algorithm . . . . . . . . . . . 3
2.5 Crop Recommender System Using Machine Learning Approach . . . . . . . . 3

3 Proposed Technique 5
3.1 The Basic idea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3.2 Architecture of suggested Prediction Method . . . . . . . . . . . . . . . . . . 5
3.3 External Environment Monitoring and Prediction
Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3.4 Working of the Proposed System . . . . . . . . . . . . . . . . . . . . . . . . . 7

4 Performance Analysis 8
4.1 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

5 Future Works 10

6 Conclusion 11

Bibliography 12
List of Figures

3.1 Environment factors effecting insect growth . . . . . . . . . . . . . . . . . . . 5

3.2 Architecture of Pest Prediction . . . . . . . . . . . . . . . . . . . . . . . . . . 6
List of Tables
Chapter 1

Introduction

This research paper titled “Pest Prediction in Rice using IoT and Feed Forward Neural Net-
work ”,presents a solution to the challenges posed by pests such as stem borers in rice crops.
The paper highlights the importance of rice as a staple food commodity globally and in Pak-
istan, where the research was conducted. The authors discuss the limitations of traditional
methods of pest control and propose a solution using IoT and neural networks to predict and
prevent pest infestations in rice crops.

The paper presents a detailed methodology for the proposed solution, including the hard-
ware and software components of the IoT system and the neural network model. The authors
explain how sensors are used to collect data on environmental factors such as temperature,
humidity, and rainfall, which are known to affect pest infestations. The data is then analyzed
using a feed-forward neural network, which is trained to predict the likelihood of pest infesta-
tions based on the environmental data.

The authors demonstrate that their solution is effective in predicting and preventing pest
infestations in rice crops, with a prediction accuracy of up to 90%. The paper also discusses
the potential future directions of the research, including the use of additional sensors to col-
lect more environmental data and the optimization of the neural network model to improve
prediction accuracy.

1
Chapter 2

Literature Survey

2.1 Internet of Things Based Weekly Crop Pest Prediction

by Using Deep Neural Network
RM Saleem, RN Bashir, M Faheem, MA Haq, A Alhussen, ZS Alzamil, S Khan, IEEE
Access, 2023
This research paper proposes an Internet of Things (IoT) and deep learning-based model for
predicting crop pests using environmental parameters such as temperature, humidity, sunshine,
rainfall, and wind speed. The model is implemented in cotton crops for whitefly pest predic-
tion and is designed to be flexible and applicable to any pest prediction on any crop. The study
identifies the important environmental parameters that are considered for the development rate
of a pest and proposes pest prediction on the environmental conditions. The proposed model
stands out due to its distinctiveness in considering multiple parameters during predictions and
its utilization of directly sensed environmental conditions. The weekly predictions are gen-
erated from the last seven days of environmental conditions, which are useful for short-term
measures against pest attacks like the use of pesticides. The study concludes that the proposed
model is precise and expendable for other crops as well and can ease the execution of real-time
crop field monitoring.

2.2 Deep Learning model for early prediction of plant dis-

ease
PE Rubini, P Kavitha 2021 Third International Conference on Intelligent Communica-
tion, 2021
This paper proposes a deep learning model for early prediction of plant disease, which aims
to help farmers in the early diagnosis of diseased plants, significantly impacting agricultural
production. The proposed model accurately classifies leaf images and provides information on

2
Pest Prediction in Rice using IoT and Feed Forward Neural Network 2023

the type of disease. The dataset used in the proposed work contains around 17000 images of
tomato leaves, which are split into three groups: test, train, and validation. Transfer learning is
used, and the model architecture used here is VGG-16 and Dense Net. The proposed system,
in blending with its measured interpretability and accuracy, may effectively aid farmers in the
early diagnosis of the diseased leaf.

2.3 Towards Precision Agriculture: IoT-enabled Intelligent

Irrigation Systems Using Deep Learning Neural Network
PK Kashyap, S Kumar, A Jaiswal, M Prasad, AH Gandomi,IEEE Sensors Journal, 2021
This paper presents an IoT-enabled intelligent irrigation system for precision agriculture.
The system includes a smart irrigation model, an associated sensing IoT network model, and an
LSTM RNN model to predict soil moisture. The irrigation scheduler decides whether irrigation
is required, estimates water volume, and evaluates the irrigation period using sprinkled deploy-
ment and its flow rate. The proposed LSTM RNN-based intelligent irrigation algorithm DLiSA
is compared with state-of-the-art algorithms. The proposed model is calibrated and tested using
real-time series data of three different regions. The paper concludes with a discussion of related
works, the results of the simulation, and the importance of sustainable agriculture practices.

2.4 Crop Yield Prediction using Machine Learning Algorithm

Reddy, D. Jayanarayana, and M. Rudra Kumar, 2021 5th International Conference on
Intelligent Computing and Control Systems (ICICCS). IEEE, 2021.
This research paper discusses the use of machine learning algorithms in predicting crop
yields, which is important for the success of agriculture in India. The paper provides a step-
wise process for crop yield analysis and compares several ML methods used for crop yield
prediction. The paper also includes a comparative analysis of several types of research. The au-
thors discuss the advantages and limitations of various ML techniques, including SVM, CNN,
and H-ANN. Overall, the paper provides valuable insights into the use of machine learning in
agriculture and its potential to improve crop yield prediction.

2.5 Crop Recommender System Using Machine Learning Ap-

proach
Pande, Shilpa Mangesh, 2021 5th international conference on computing methodologies
and communication (ICCMC). IEEE, 2021.

Department of Computer Science and Engineering 3

Pest Prediction in Rice using IoT and Feed Forward Neural Network 2023

This paper proposes a crop recommender system for farmers in India that predicts crop yield
using machine learning algorithms. The system takes into account factors such as rainfall,
temperature, area, season, and soil type to recommend the most profitable crop and suggest
the best time to use fertilizers. The proposed system is a mobile-based application that is
user-friendly and accessible. The paper also discusses the background work of researchers in
the field of agriculture and yield prediction, presents the proposed model for yield prediction
and crop recommendation, and discusses the results. Overall, the paper aims to provide a
solution to help farmers make informed decisions about their farming practices and improve
their livelihoods.

Department of Computer Science and Engineering 4

Chapter 3

Proposed Technique

3.1 The Basic idea

Adults of yellow stem borer are attracted to light conditions, which triggers the initiation
and movement of offspring to a specific location. Along with light, some other influential fac-
tors for the exponential growth of yellow stem borer are relative humidity and rainfall [39].
However, the development and growth of the yellow stem borer and its life cycle is mainly in-
fluenced and derived by temperature [40]. Day length changes and cooler daytime temperatures
induce temporary arrest or diapause in the growth and development of mature larvae.

Figure 3.1: Environment factors effecting insect growth

The Fig. shows about external environment factors making favorable conditions for insect
pest growth which ultimately attacks on rice plant.

3.2 Architecture of suggested Prediction Method

The proposed system monitors the different environmental factor like temperature, hu-
midity, and rainfall. Proposed system is divided into two modules i.e., sensor base module

5
Pest Prediction in Rice using IoT and Feed Forward Neural Network 2023

for environmental data gathering and Neural Network module for data examination. Detailed
working of proposed system is consisting of following steps:

1. Data Collection from environmental observing system consists of Temperature and Hu-
midity sensor (DHT22) and Rainfall Sensor (FC-37) and Arduino device.

2. Recording of respective values in proposed system for analysis.

3. Data examination by NN model.

4. Demonstrating the generated results of planted sensors on mobile.

5. Informing farmers about the current environmental situation of their crop to take neces-
sary steps.

3.3 External Environment Monitoring and Prediction

Module
In proposed approach, data is collected using sensors and a hardware device Arduino as a
microcontroller.

Figure 3.2: Architecture of Pest Prediction

Fig. elaborates details of working system containing Temperature, Humidity and Rainfall
sensors, collected data is stored in cloud storage and processed using custom designed FFNN

Department of Computer Science and Engineering 6

Pest Prediction in Rice using IoT and Feed Forward Neural Network 2023

and after analysis output is generated as alert on mobile device. In this study FFNN is proposed
for classifying existing environmental situation to build effective pest prediction modeling sys-
tem. Although there are several machine learning algorithms that can perform this type of
classification, the results obtained from machine learning algorithms are not so much satisfy-
ing. That’s why most researchers prefer neural network. Among many reasons of using NN
some of them are:

1. Neural Network has ability to manage large dataset without getting overfit [73]. Training
data-size may vary in our projected scenario and in future if any factor like temperature,
humidity and rainfall are deliberated for better results of proposed modeling system.

2. NN can be customized by changing the training and testing data size. We preferred over
other popular machine learning methods, which would eventually affect our output and
no additional parameter would be required if any change occur in future.

3. Another main reason of choosing ANN is that its output performance continues to in-
crease by increasing its training data set

3.4 Working of the Proposed System

1. Data Collection: The system collects data from environmental observing systems con-
sisting of temperature and humidity sensors (DHT22) and a rainfall sensor (FC-37) con-
nected to an Arduino device. The sensors collect data on environmental factors such as
temperature, humidity, and rainfall, which are known to affect pest infestations.

2. Data Recording: The system records the data collected from the sensors in a database for
analysis.

3. Data Examination: The system uses a feed-forward neural network to analyze the data
collected from the sensors. The neural network is trained on a dataset of environmental
data and pest infestation data to predict the likelihood of pest infestations based on the
environmental data.

4. Results Demonstration: The system generates results of the planted sensors on mobile
devices, which can be accessed by farmers. The results show the current environmental
situation of their crop, including the likelihood of pest infestations.

5. Farmer Notification: The system informs farmers about the current environmental situa-
tion of their crop and the likelihood of pest infestations. Farmers can take necessary steps
to prevent pest infestations based on the information provided by the system.

Department of Computer Science and Engineering 7

Chapter 4

Performance Analysis

4.1 Results and Discussion

We used NN (Neural Network) and binary classification model to predict environmental
type. Neural Network has already implemented in many fields and research areas. We trained
neural networks on the training dataset and then tested this trained neural network model by
providing some unseen data taken from the fields. We evaluated attained output of our trained
Neural Network with the formula given below:

Initially we collected 340 records from the sensors. 285 samples are used for training the
model while 55 samples are used for validation purpose. 85.6gradually increased after repeat-
ing several training rounds (epochs). The ANN algorithm binary classification designed in
python language with keras and sklearn libraries.

In our experiment, we have set a parameter called ‘patience’ on the loss of the model to
monitor. If the loss starts to increase gradually then after two epoch patience the model stops
training automatically.

Besides good accuracy, there are few limitations in the proposed system as the designed
system is highly dependent on the network. If the network is unstable or our hardware gets

8
Pest Prediction in Rice using IoT and Feed Forward Neural Network 2023

disconnected then data collection may be affected. Moreover, sensors can be affected by ex-
treme weather conditions like storms or floods making them give a false reading. Therefore, a
good network is necessary and a good mounting position of hardware is also required to work
properly.

Department of Computer Science and Engineering 9

Chapter 5

Future Works

the research paper proposes several future works that can be done to improve the proposed
system for pest prediction in rice using IoT and feed-forward neural network. Some of these
future works are:

• Feature Selection: The authors suggest that future research can focus on using feature
selection techniques to optimize the input and output results of the proposed model. This
can help to improve the accuracy of prediction by 5-10

• Additional Sensors: The authors propose that more sensors could be added to the existing
system, such as light sensors, soil moisture sensors, and air monitoring sensors. This can
help to collect more environmental data and make predictions that are more accurate.

• Pest Identification: The authors suggest that future research can focus on identifying the
specific type of pest infestation based on the environmental data collected. This can help
farmers to take more targeted measures to prevent pest infestations.

• Integration with Crop Management Systems: The authors propose that the system can be
integrated with crop management systems to provide farmers with a more comprehensive
solution for pest control. This can help farmers to take a more proactive approach to pest
control and improve crop yield.

• Real-time Monitoring: The authors suggest that future research can focus on developing
a real-time monitoring system that can provide farmers with continuous updates on the
environmental situation of their crop. This can help farmers to take immediate action to
prevent pest infestations and improve crop yield.

10
Chapter 6

Conclusion

In conclusion, this research paper proposes a solution for pest prediction in rice using IoT
and feed-forward neural network. The paper highlights the importance of rice as a staple food
commodity globally and the challenges posed by pests such as stem borers in rice crops. The
authors discuss the limitations of traditional methods of pest control and propose a solution
using IoT and neural networks to predict and prevent pest infestations in rice crops.

The paper presents a detailed methodology for the proposed solution, including the hard-
ware and software components of the IoT system and the neural network model. The authors
explain how sensors are used to collect data on environmental factors such as temperature,
humidity, and rainfall, which are known to affect pest infestations. The data is then analyzed
using a feed-forward neural network, which is trained to predict the likelihood of pest infesta-
tions based on the environmental data.

The paper presents the results of experiments conducted to test the effectiveness of the
proposed solution. The authors demonstrate that their solution is effective in predicting and
preventing pest infestations in rice crops, with a prediction accuracy of up to 90%. The paper
also discusses the potential future directions of the research, including the use of additional
sensors to collect more environmental data and the optimization of the neural network model
to improve prediction accuracy.

Overall, the proposed solution has the potential to significantly reduce yield losses and im-
prove food security, making it a promising area for further research and development. The
authors suggest that future research can focus on feature selection, additional sensors, pest
identification, integration with crop management systems, and real-time monitoring to enhance
the accuracy and effectiveness of the system. The proposed solution has the potential to revo-
lutionize pest control in agriculture and improve the livelihoods of farmers globally.

11
Bibliography

[1] Saleem, R.M., Bashir, R.N., Faheem, M., Haq, M.A., Alhussen, A., Alzamil, Z.S. and Khan,
S., 2023. Internet of things based weekly crop pest prediction by using deep neural network.
IEEE Access.

[2] Rubini, P.E. and Kavitha, P., 2021, February. Deep Learning model for early prediction of
plant disease. In 2021 Third International Conference on Intelligent Communication Technolo-
gies and Virtual Mobile Networks (ICICV) (pp. 1104-1107). IEEE.

[3] Kashyap, P.K., Kumar, S., Jaiswal, A., Prasad, M. and Gandomi, A.H., 2021. Towards
precision agriculture: IoT-enabled intelligent irrigation systems using deep learning neural net-
work. IEEE Sensors Journal, 21(16), pp.17479-17491.

[4] Reddy, D.J. and Kumar, M.R., 2021, May. Crop yield prediction using machine learning al-
gorithm. In 2021 5th International Conference on Intelligent Computing and Control Systems
(ICICCS) (pp. 1466-1470). IEEE.

[5]Pande, S.M., Ramesh, P.K., ANMOL, A., Aishwarya, B.R., ROHILLA, K. and SHAU-
RYA, K., 2021, April. Crop recommender system using machine learning approach. In 2021
5th international conference on computing methodologies and communication (ICCMC) (pp.
1066-1071). IEEE.

12