tushgaurav / CV-SecuritySystem

Computer-Vision-Security-System

A Security System using the face recognition, which can be monitored from anywhere using a HTTP server, coded using Python and Jinja2.

Installation

System Requirements

1. Python3

Install Python3 for Windows/Linux/Mac OS from https://www.python.org/downloads/release/python-3

2. SqliteBrowser(for sqlite database)

Download and Install SQLite Browser , to view/modify the sqlite database from https://sqlitebrowser.org/

3. pip

1. Download it get-pip.py file from https://bootstrap.pypa.io
   
2. Open terminal and install pip python get-pip.py

Python Packages Requirements

pip install numpy pillow flask opencv-python opencv-contrib-python twilio dropbox

Hardware Requirements

1. 2 cameras, 1 for surveillance and another to take face data for training face recognizer
   
2. A server to host the Web server
   
3. Router (optional)

Cloning the code from git

1. git clone https://github.com/shangeth/Computer-Vision-Security-System.git

Running the program

1. cd server
2. python server.py this will open the webserver on 127.0.0.1:5000

How does it work

1. Home page will show live feed of camera and log of people who visited the facility.
2. Add Visitor allows you to enter the details of new visitor who need to be added/permitted into the facility.
3. After adding details of the new visitor, python detect_faces.py. Before running it make sure the new visitor’s face is in front of the second USB camera.

Enabling SMS notification(optional)

1. Go to twilio.com and create an developer account and get the auth token and SID for the application.
2. Go to dropbox.com and create an developer account and generate auth token for the application in create app section.
3. nano camera_recognition.py and edit line 40 and 61 to enter your credentials.