README

Kongfu Panda Security System

This project is a security system.

It contains several intereting technique components.
1. face_recognition
   The world's simplest facial recognition api for Python and
   the command line.
   It works based on dlib which is a modern C++ toolkit containing
   machine learning algorithms and tools for creating complex
   software in C++ to solve real world problems.
2. UVC Camera
   UVC stands for USB Video Class.
   The USB Device Class Definition for Video Devices, or USB Video
   Class, defines video streaming functionality on the Universal
   Serial Bus. Much like nearly all mass storage devices (USB flash
   disks, external SATA disk enclosures, ...) can be managed by a
   single driver because they conform to the USB Mass Storage
   specification, UVC compliant peripherals only need a generic driver.
3. Human Detection Sensor
   This is a sensor which can detect human and give a trigger signal
   if there is a person.
4. Helix Device Cloud
   Wind River® Helix™ Device Cloud delivers secure, enterprise-grade
   device lifecycle management so you can easily design your Internet
   of Things (IoT) solutions with performance that scales to meet your
   deployment needs.
5. Web Pages
   This include lighttpd, websocket and HTML/JS/CSS.
   
This system uses camera to capture images, sends them to
face_recognition to compare to the authentication database, if successes,
the log with contents of person and time will be send to HDC server, if fails, the
picture of unknown person will be send to the HDC server and also an alert email
will be sent to administrator to remind him whether to call the police.

How to build and run the system:

NOTICE: This is only a rough guide, please fetch more informations this guide mentioned.

1.  Build wrlinux and install it to Intel X64 board
    See Wind River Linux part of WindRiver Knowledge Library.
2.  Install machine learning libary (dlib)
    Download the dlib source , build and install it.
3.  Install Python
    You may use the python in the wrlinux or rebuild yourself.
4.  Install face_recognition component and corresponded components.
5.  Make the uvc binary, and copy it to device_cloud directory.
    Buy a camera which is supportted by the UVC and plug it into the computer.
6.  Reconfigure the helix_cloud
    See HDC part of WindRiver Knowledge Library.
7.  Get the device_id and modify the corresponding parts of webapp.
8.  Add ti572x codes to vxWorks7 related BSP, build and load to board.
9.  Install serial component and corresponded components
a.  Run "python Kongfu.py" to run the application.