Steps to make FaceChecker system

In order to make this system running, you must following steps:

  1. Understand Linkedface++ platform: go to Linkedface++ user guide.
  2. Prepare hardware
  3. Install system: hardware and software
  4. Register account in Linkedface++
  5. Configure modules of FaceChecker
  6. Run and monitor application

Hardware requirement

FaceChecker need following devices:

Camera Hub

We can use Raspberry Pi or equivalence mini-computer as Camera hub. In this testing, we use: Raspberry Pi 2 - https://www.raspberrypi.org/products/raspberry-pi-2-model-b/

You need a SD Card at least 8GB to store Raspbian operating system, all dependency software and device module.

Raspberry Pi Power Micro SD card Raspberry Pi

Compatible camera

Using and camera that can connect to Camera hub and capture image from hub.

We use Raspberry Pi Camera Module - RPi Camera (B), Adjustable-Focus – as described in this page: http://www.waveshare.com/RPi-Camera-B.htm

Raspberry Pi Camera

Servers

Server (module 3, 4 and 5) are for running application server and web administration application and database. You can use any server that can install Nodejs Application server and MySQL server 5.6

Installation

Hardware

  • Use Ethernet cable to connect Camera hub and all servers through LAN Port.

Raspberry Pi Camera

  • Connect Camera to Raspberry Pi through CSI Camera port

Raspberry Pi Camera

  • Entire system must be connected to internet, in order to work with Linkedface++ service

Software

Install Raspberry Pi OS

At the time of this paper, Raspberry Pi uses Raspbian as Operating system. This step takes a long time. You can do it in a service center or by yourself.

In case, you want to do it your-self, you must use an HDMI cable to connect your Raspberry Pi to a screen and follow guide in this link: https://www.raspberrypi.org/documentation/installation/

Raspberry Pi Camera

After installing Raspbian operating system, you must set IP for you Raspberry Pi so that it can access from and to servers. Type command to set IP:

# sudo vi /etc/network/interfaces

Change to pattern:

auto eth0
iface eth0 inet static
   address 10.253.0.50
   netmask 255.255.255.0
   network 10.253.0.0
   gateway 10.253.0.1
   dns-nameservers 8.8.8.8

then restart network service: sudo service networking restart

Install FaceChecker-device module

This is python module running in Raspberry. Download software from this link: https://github.com/papaiking/faceChecker_device.git

Follow the installation guide in this repository to install module.

Install FaceChecker-services

This module runs in an application server to provide services for system. Download software from this link: https://github.com/papaiking/faceChecker_services.git

This module is NodeJs application. Follow the installation guide in this repository to install module.

Install FaceChecker Admin App

This module is an AngularJS application, running in Nodejs server to provide UI for system operator. Download software from this link: https://github.com/papaiking/faceChecker_dashboard.git

Follow the installation guide in this repository to install module.

Configuration

Raspberry Pi Camera

After this step, you will have security authentication information for application: app_id and app_secret

  • Set above values for module: facechecker_server at file: facechecker_server/config/settings.js
app_identity: {
    app_id:"73705ee75cbdf46f7d68539e7523e879",
    secret:"RC3VZ53O5DJH3V687WCKOXZ81XQF80RY"
}

Running system

  1. Start Application server, web administrator application
  2. Connect to Camera Hub by SSH, go to root folder of module: facechecker_device and run command: ./bin/facechecker [-f frequency] [-i idle_time]

    System will recognize events and store in database If you specify frequency and idle_time it will overwrite default values CHECKING_INTERVAL and IDLE_INTERVAL in configuration file

  3. If you want to monitor event in real-time, go to Surveillance board web application: http://ip_address:9000.

Raspberry Pi Camera