imentos / dubai-teched

Smart City & Traffic Control powered by Deep Learning

Setup:

1. Go to root folder:

npm install

2. Go to littlebits folder:

npm install

3. Go to web folder

bower install

Startup:

1. Go to root folder:

node index.js

2. Go to littlebits folder:

node index.js

3. Go to http://localhost:3000/index.html in Chrome

NOTE: Make sure that NO cars are in the parking lots before you start the website or restart the application. The free parking should always start with 6.

Output Image Location:

We capture images from webcam every second and save into this location for Karthik's algorithm. It is important to have the right location. Otherwise, car counting won't work.

Go to root\config.json

{
	"imageLocation": "<YOUR IMAGE OUTPUT LOCATION>"
}

Sensors:

We use one Ardunio board (UNO) to read 6 light sensors. Based on the USB ports, this board should be assigned to ttyACM0. Please see the following circuit diagram for the light sensor setting. https://github.com/rwaldron/johnny-five/blob/master/docs/photoresistor.md

Sensors Adjustment:

Go to littlebits\config.json

{
	// This value is the Ubuntu USB port for Ardunio board.
	"port": "/dev/ttyACM0",

	// This threhold controls when the status changes is triggered. For example, 'free' or 'occupied'. 
	// This value is scaled down to 0-10.
	"statusThreshold": 5,

	// This threhold controls when the value changes is triggered. For example, the change ONLY happens when the change amount is over 500.
	// This value is between 0-1023.
	"changeThreshold": 500,

	// This value controls how often the value is read. 
	// This value is based on milliseconds. 
	"checkFreq": 500
}