ABES ENGINEERING COLLEGE GHAZIABAD

DEPARTMENT OF COMPUTER SCIENCE

REPTILE PRESENCE DETECTION IN FARMS

Presented By: Supervisor Name:

ISHITA VERMA MR. MADAN SINGH

CSME A
 OUTLINES OF PRESENTATION
PROBLEM STATEMENT:
Detection of reptiles in the farms and diverting them without hurting them.

Explanation:
Through the help of this presentation , it would be quite evident that as farmers spend most of their time
irrigating and yielding the fields, it becomes too hazardous for them to work especially at the night time
due to the presence of various reptiles . It is found in the various researches that almost 65% of the
farmers unnatural deaths occur due to the silent attacks by the reptiles like snakes, dragon lizards, etc.
So, in order to aid the farmers with a solution to this problem , this presentation would lead to the
formation of a hardware device having various resources such as Raspberry Pi which runs the
machine algorithm, which is interfaced with the ESP8266 Wireless Fidelity module, Pi
Camera, Buzzer, and LED. Machine learning algorithms like Region-
based Convolutional Neural Network and Single Shot Detection technology plays an
important role to detect the object in the images and classify the animals. The
experimentation reveals that the Single Shot Detection algorithm outperforms than
Region-based Convolutional Neural Network algorithm. Finally, the Twilio API
interfaced software decimates the information to the farmers to take decisive action
in their farm field.
 CONTENTS
SR NO. CONTENTS SLIDE NO.
1. Outlines of presentation 2
2. Introduction 4-5
3. Objectives 6
4. Related Work 7-8
5. Gap Analysis 9
6. Applications 10
7. Proposed Methodology 11
8. Tools/Technology Used 12-14
9. Conclusion 15
10. References 16
 INTRODUCTION
Agriculture is the backbone of the Indian economy, where more
Pie Chart Depicting The Ratios of
Various Reasons for Unnatural than 60% of the country's population directly or indirectly
Deaths of Farmers depends on this sector.
As per the research done by the National Farmers Association , it was found that
more than 65% of the unnatural deaths are caused by the attacks by the reptiles
either during the nurturing , yielding or irrigation of the crops. Hence, it becomes
a major area of concern to be brought into concern.
Also, few attempts were tried to solve this conflict by using
technology such as IoT and Machine learning, which is called A
IoT (Artificial Intelligence for the Infrastructure of Internet of
Things). Our proposed model uses IoT and Machine learning
concepts based solution to it.
IoT (Internet of Things) controls the Things that are connected to
it and transfers the data over the network. The IoT technology
enables the collection of real-time data from the farm field using
Rrptiles Loans and Mental pressue Sensors and various electronic components . In this work, we
Attacks by intruding animals Natural calamity
present the coordination of Pi Camera, LED, and Buzzer
interacting with the cloud a new service in the domain. The
peripheral part adopted wireless technologies such as Wi Fi for
Pi Camera is used to capture real-time images in a farm field 24. The low cost and ease of
programming controller Raspberry Pi for coordination of hardware part and data transferred
through ESP8266 WiFi module, which uses TCP/IP protocol .
Machine learning is a branch of artificial intelligence used for data analysis to automate the
analytical model building, which identifies the pattern and objects to make a decision . In this
work, a deep machine learning algorithm for object detection and classification model is
trained and tested. Twilio communication is API interface software which used to
communicate throughout globally by creating its own/private network. It is used to forward
the decisive information to the framers, which has control through the Raspberry Pi controller.
Here from all this motivated to propose an effective model to protect the crop from animal
intrusion through modern technology.
 OBJECTIVES
• To detect the motion of any reptile in order to detect the presence of the reptile in the farm.
• After the detection of the presence of reptile in the farm, alarm the farmer about the presence of
the reptile.
• To design a system which will notify the farmer about the presence of the reptile and also it’s
category and then a way to divert it’s way.
 RELATED WORK
Year Writer Proposed outcome
2020 S. Giordano worked and developed an IoT application for the protection of crops from animal intrusion in the crop field. To
collect or monitor the crop field, authors used the wireless technologies such as 6LoWPAN, Wi Fi, and ZigBee
with the advanced IoT gateway. An ultrasound repeller device was developed to work even in partial and total
darkness using the solar panel along with LiPo batteries. To improve the efficiency of this device used PIR
(Passive Infrared) sensor, which takes care of the frequency transmission and networking operation by
transmitting a small size frame at a distance of 50m. This communication happens using the RIOT-OS software,
when the animal is detected it produces a sound of 120dB. The performance of this device reduces below 90%
if the distance from the gateway is above 60m and it will never work if the distance crosses 100m.

2021 Mukesh Mahajan worked on protecting the crop in the farm field from animals such as buffaloes, cows, goats, and birds using a
PIC microcontroller. The developed model uses the motion sensor to detect the animals that are near to farm
field and the sensor signals the microcontroller to take appropriate action by farmers. The PIR-based motion
sensor is used to detect the animals and the buzzer is used to notify the farmers based on the microcontroller
instruction. Here authors claim that this model avoids the farmers staying for 24 hours in the farm field and
barricade their crops.

2021 Iniyaa K K worked on protecting crops by animals using deep machine learning and Convolutional neural
network algorithm. The author aims to protect the crops from animals and not harm both animals as well as
humans in the conflict. Due to this, the authors developed a model to divert the animals nearer to the crop
fields. The machine-learning algorithm-based model is developed to detect the animals coming nearer to the
farm field using the neural network concept through the computer vision technique. In this model, the farm
field is monitored using a camera placed at the farm field at a regular interval of time. The algorithm detects
the animal's movement through the camera frames using various libraries function and concepts of neural
networks and plays appropriate sound to divert the animals away from the farm field.
 RELATED WORK
Year Writer Proposed outcome

2020 D. Kalra worked and developed a model to protect the crop from insects and small animals through the
sensor and also for control irrigation using the IoT technologies. The Arduino UNO microcontroller
works like a heart for the proposed model in managing proper irrigation and crop protection. The
irrigation is managed automatically on/off water siphons depending on the dampness parameters
of the farm field. The crops are protected by insects, animals, etc through the use of deliberate
sensors connected in the farm field; sensors estimate the motion of insects and animals nearer to
the crop and sent the signal to the Arduino Uno microcontroller for calculation of distance and all.
Based on the calculated distance values, the microcontroller enables high-frequency sound.

2021 Raksha R and worked and developed a prototype to monitor the crops and warning the wild animals based on
Surekha P two emerging technologies such as IoT and Machine learning. The IoT components used are like
PTZ (Pan-Tilt-Zoom) camera, GSM module, Sensors, and Arduino UNO microcontroller and
Machine learning algorithm for classification of the animals are done using KNN (K-Nearest
Neighbor) Algorithm, Logistic Regression, and SVM (Support Vector Machine) Algorithm. Datasets
of elephants, horses, Zebra, etc are taken in total in some 605 images. SVM provides an accuracy
of 89.6% compared to the KNN and Logistic Regression model for the iterated regularization
parameter of C=100.
 GAP ANALYSIS
From going through the various research papers and literature reviews , the following gaps have been
identified in the prior works:
• More emphasis have been put on the Iot parts of the project
• More emphasis is put on the large pace or scale of work such as animal intrusion and herd handelling
• There is no system for ground level detection
• More work is done in the field of cattle and crop protection process
• Only limited for animal intrusion

How these gaps can be fulfilled by this project?

• More emphasis would be on the Machine learning part of the project
• More emphasis would be on the reptile detection on the ground
• There would not only be ground level detection but also notification of the presence of the reptile on
the ground
• More emphasis would be on the farmers’ life security
• Not only serves the purpose of reptile detection but also ensure life security of the farmers
 APPLICATIONS
This project is majorly applicable in the agricultural
field. Till now in majority of the projects more
emphasis have been put forward on the crop
protection measures and methods as they were
only limited to the animal intrusion detection.
But this project aims more towards the protection
or security of the lives of farmers through the
detection of the presence of reptiles in the farms
and then alarming them about the presence of the
reptiles in the farms.
This project involves both the application of
Internet of Things as well as machine learning,
hence in totality known as A IoT (Artificial
Intelligence for the Infrastructure of
Internet of Things)
 PROPOSED METHODOLOGY
DATA COLLECTION

DATA PRE PROCESSING

FEATURE EXTRACTION

NOTIFICATION

OUTCOME
 TOOLS/TECHNOLOGY REQUIRED
The alternating current of 12V is converted into a direct current of 5V through
a bridge rectifier circuit and 7805 voltage regulator to operate Raspberry pi.
ESP8266 is interfaced with a Raspberry pi board to provide firebase cloud
connectivity to the system . Raspberry pi4 is used to run a machine learning
algorithm. Pi cam is used to capture the reptile images entering the agricultural
field. The images are then analyzed by the machine learning algorithm running on
the Raspberry pi board and conclusions are drawn as an output of machine learning
algorithm. If any danger is sensed the Raspberry pi generates sounds of different
frequencies with the help of a buzzer and an input signal is sent to ESP8266
nodemcu which is in communication with the firebase IoT cloud which sends
messages to the farmer through the android application . The proposed model
consists of two parts as Hardware and Software. Raspberry pi acts like a heart for
the hardware part and it is interfaced with components like a voltage regulator, Pi
Camera, LED lights, Wi Fi, and Buzzer. The software part for the hardware is done
through embedded C and for computer vision prediction used machine learning
models like R-CNN and SSD for object detection and predicts the animals.
 TOOLS/TECHNOLOGY REQUIRED
Hardware components

Power WiFi Module Pi Camera Processor

A 12V Lead-acid battery is ESP8266 Wi Fi module allows Pi Camera V2 module is 8- Raspberry Pi 4 controller unit
used for the power supply of the Raspberry Pi controller to megapixel custom-designed is integrated with Broadcom
electronic components connect with Wi Fi network board that is equipped with a BCM2711 and Quad-core
connected in the hardware using TCP/IP (Transmission flexible ribbon cable, making Cortex-72 processor which
part, which consists of 6 cells Control Protocol/Internet it compatible with Raspberry works for 64-bit SoC (System-
connected in series each Protocol) connection. It Pi. The camera module takes on-Chip). This controller
producing 2.1V. The lead-acid operates for a frequency of pictures in two versions i.e. works for a voltage of 5V and
battery can't generate the 2.4GHz using the serial Standard version and the No communicates easily with
power itself, where it stores communication UART to IR version . The Standard other connectivity such as
power received from another transmit and receive data. version; is designed to ESP8266, Pi camera, etc. It
source for this it is called a Two general-purpose pins capture images in normal provides a set of general-
storage battery. For example such as GPIO0 and GPIO2 are light and the No IR version; is purpose input/output pins to
here the 12V battery can used to decide to designed to capture images have control over electronic
supply a current of 10mA for transmit/receive for serial in dark light using components for physical
a duration of 12.5 hours, input/output purposes, which the infrared light source. computing for IoT
anyhow if consumption of are connected to ground and technology, also the
current decreased means VCC. programming is easy and the
duration can be increased. cost is low.
 TOOLS/TECHNOLOGY REQUIRED
Software components

R-CNN approach here used for the object detection uses deep models . The R-CNN is composed of
four main parts such as selective search, pretrained CNN, category prediction, and bounding
R-CNN Model box prediction. For the input images, selective search is applied to select multiple high-quality
proposed regions, which are in multiple scales and different sizes and shapes. In between the
selective search and output, pretrained CNN is placed. Pretrained CNN works on forward
computation to extract the features for output, which gather input as required by the network
through the proposed region. For object classification, multiple Support Vector Machines were
trained using each proposed region of both features and labeled category. Then for the ground
truth bounding box prediction linear regression model is trained using each proposed region of
both features and labeled bounding box hardware part and it is interfaced with components like a
voltage regulator, Pi Camera, LED lights, Wi Fi, and Buzzer. The software part for the hardware is
Singlethrough
done Shot Multi box Detection
embedded (SSD)
C and for modelvision
computer consists series of
prediction components
used like a base
machine learning network
models like
SSD Model block and several multiscale feature blocks. In original
R-CNN and SSD for object detection and predicts the animals.images, features are extracted using the
base network block based on the deep Convolutional neural network . Here to detect the small
objects in the original images more anchor boxes are generated using the feature map. Several
multiscale feature blocks are used to reduce the size. The multiscale feature blocks detect the
objects of different sizes based on the predicted bounding boxes and anchor boxes.
 CONCLUSION
The project focuses on the reptile detection in the farms.By the end of the project, a system
would be developed which will detect the presence of the reptile in the farms through using
the combination of the internet of things and the machine learning models such as R CNN
and SSD model.
A system would be developed which will capture the real time images through the pi camera
and then train and analyse that image through the R CNN and the SSD model through which
the presence of the reptile would be detected and then the framer would be notified through
the outcome of that model.
 REFERENCES
[1] ArturFrankiewicz; RafałCupek.”Smart Passive Infrared Sensor - Hardware Platform“ Year: 2013 IECON 2013 - 39th Annual Conference of
the IEEE Industrial Electronics Society Pages: 7543 - 7547,
[2] DOI: 10.1109/IECON.2013.6700389 CITED BY: PAPERS (1)
[3] Hanshi Wang; Jingli Lu; Lizhen Liu; Wei Song; Zhaoxia Wang; “Community Alarm System Design Based On MCU And GSM” Year: 2015
[4] Volume:01 Pages:859-862, DOI:10.1109/ICCSNT.2015.7490876 , IEEE Conference Publications.
[5] Markus Borschbach; Navya Amin ,“Quality Of Obstacle Distance Measurement Using Ultrasonic Sensor And Precision Of Two Computer
Vision-Based Obstacle Detection Approaches” Year: 2015, 2015 International Conference on Smart Sensors and Systems (IC-SSS) [6] Pages:
1-6, DOI: 10.1109/SMARTSENS.2015.7873595IEEE Conference Publications
[7] Mustapha, Baharuddin, AladinZayegh, and Rezaul K. Begg. “Ultrasonic And Infrared Sensors Performance In A Wireless Obstacle
Detection System” Artificial Intelligence, Modelling and Simulation (AIMS), 2013 1st International Conference on. IEEE, 2013.
[8] Dr. Wilson, “Electric Fence” Handbook of Texas, Project report published by the Texas State Historical Association. August 4, 2011
[9] T. Mohammad, “Using Ultrasonic And Infrared Sensors For Distance Measurement” World Academy of Science, Engineering and
Technology, pp. 293-298, 2009.
[10] Discant, A. Rogozan, C. Rusu and A. Bensrhair, “Sensors For Obstacle Detection” 2007 30th International Spring Seminar on Electronics
Technology (ISSE), Cluj-Napoca, 2007, pp. 100-105. doi: 10.1109/ISSE.2007.4432828
THANK YOU