0% found this document useful (0 votes)
71 views

Project Progress Presentation

This document presents a project progress presentation on intrusion detection using face recognition with a CCTV camera and GSM notification. The project is being developed by three fourth-year computer science students for their department. It aims to enhance security systems by using face recognition and alerting registered users via text message if an intruder is detected, at which point an alarm will sound. The project is being developed in two sections - face detection and matching using LBPH modeling, and an SOS notification/alarm system. The document outlines the working description, platform, face recognition model used, and face detection/matching process developed so far.

Uploaded by

Priyanshu Mangal
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
71 views

Project Progress Presentation

This document presents a project progress presentation on intrusion detection using face recognition with a CCTV camera and GSM notification. The project is being developed by three fourth-year computer science students for their department. It aims to enhance security systems by using face recognition and alerting registered users via text message if an intruder is detected, at which point an alarm will sound. The project is being developed in two sections - face detection and matching using LBPH modeling, and an SOS notification/alarm system. The document outlines the working description, platform, face recognition model used, and face detection/matching process developed so far.

Uploaded by

Priyanshu Mangal
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 14

RAJA BALWANT SINGH ENGINEERING TECHNICAL CAMPUS

BICHPURI, AGRA - 283105

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING


A PROJECT PROGRESS PRESENTATION ON
INTRUSION DETECTION USING FACE RECOGNITION CCTV & GSM NOTIFICATION

PROJECT GUIDE: PRESENTED BY: AKSHAY KUMAR


ER. LAVKUSH SHARMA ANSHUMAN JAISWAL
SURYANSH CHAUHAN
C.S.E. (IVth YEAR)
CONTENTS

1. Introduction
2. Working Description
3. Sections
4. Platform & Environment Used
5. Model For Face Recognition
6. Face Detection & Matching
7. Conclusion Of This Presentation
INTRODUCTION
• Our main purpose of this project is to enhance the security system,
• The motto of our project revolves around intruder & intrusion.
• We are using face recognition along with GSM module to built an advance secure CCTV
Camera.
WORKING DESCRIPTION

• Our proposed work is a software embedded module consists of various units that make up the
system: the power supply unit, Raspberry Pi 4, Raspberry Camera, GSM Modem, & buzzer
unit.
• We are going to use Raspberry Pi 4 for image processing i.e Face Recognition Module.
• We have built up two sections along with four phases in our project for the ease of development
and working.
SECTIONS

• We have divided the development of project in two main sections:


1. Face Detection & Matching :
We can say that it is the main section of this project as it holds the major working of
the project. It is used to gasp the faces that come across in the CCTV Surveillance and
matches with the dataset of the stores faces.
2. SOS & Buzzer Unit:
In this section we had send a message to the registered user informing him/her about
the intrusion. Following this the buzzer unit comes up in action and starts the burglar
alarm.
PLATFORM & ENVIRONMENT USED
• We have used Python 3 (version 3.7) as our main programming language to build up the
module for “Face Detection & Matching”.
• We have used Anaconda for various tools and packages of Python for face recognition
module.
• The python IDE we used is VSCode along with virtual environment name “face_env” in
Anaconda.
MODEL FOR FACE RECOGNITION

• We have come across many problems when we have started to build our project.
• The main concern for us is to choose a model for the implementation of face
recognition system.
• The model we finally choose to work on is LBPH (Local Binary Pattern Histogram)
Model.
• Local Binary Patterns Histogram was proposed in 2006.
• It is based on local binary operator. It is widely used in facial recognition due to its
computational simplicity and discriminative power.
• It involves three phases:
1. Face Detection & Data Gathering
2. Train the Recognizer
3. Face Recognition
FACE DETECTION & MATCHING
• This is used for detection and recognition of human faces.
• In this section we built the four sub modules along with a main module that connects all of
them.
• In this module we start with a sub module “capture_video.py” which is use to the test the
CCTV Camera.
• Then second sub module “capture_images.py” is used to capture & stored the faces of
registered users along with their name and user id which creates a CSV file along with their
name & id which is stored in “files” folder.
• The capture faces are stored in “training_images” folder.
• Then we switch to another sub module “train_images.py” in which the capture images are train using
LBPH (Local Binary Pattern Histogram) Model along with creating a “trainer.yml” file which is a dataset
stored in “files” folder.
• The last sub module “recognize.py” is used for ‘Initialising the CCTV Surveillance’ which will recognise
the stored faces in database and detect the unknown faces i.e. ‘Intruders’.
• This sub module will also store the images of intruder in “unknown_images” folder as for future instances.
• In this module we had also used “Haar Cascade Classifier” for the detection and training the dataset of
human faces.
CONCLUSION OF THIS PRESENTATION

• We had completed the development of section one which we have shown in


previous slides.
• Now we are headed towards the development of second section.
• When the development phase is completed then we will compile the source code in
Raspberry Pi which we will connect along with Raspberry Pi Camera.
• Then after that we will connect it with a GSM Modem and a buzzer along with a
power supply.
THANK YOU

You might also like