EzQ

---

[IMPORTANT] Disclaimer

The following systems must be ran in a linux based environment (in our case: ubuntu) in order to run the entirety of the project.

However, all other scripts (except server.py) can be ran in a windows environment.

Pre-requisites

The following pre-requisites are presumed to be pre-installed in your testing environment.

• libdw
• kivy
• numpy

Additional Requirements

• opencv-python: Image processing library
• cython: Cythonlibrary
• twilio: Twilio library for messaging service
• yolo34py: Image - Object detection library for detection service

1. To install the libraries in this section, go into terminal/cmd prompt and navigate to the directory: ~/EzQ/EzQ.

2. You should see our project files:

   1. cashier.py, cashier.kv, chefgui.py, chefgui.kv, credentials.txt, customer.py, customer.kv, ezq.py,firebase.py, messenger.py,server.py,twilio_credentials.txt,
   2. Folders:
      • Legacy: This folder contains all our legacy code, which are our developmental code.
      • YOLO3-4-Py: Image-Object detection library.

3. Type the following to install opencv-python, cython and twilio.

   pip3 install -r requirements.txt
   

   This line installs the above mentioned libraries (specified in requirements.txt) into the python libraries.

4. You can skip this step, if you do not want to run server.py.

   To install Yolo34py, navigate to the folder YOLO3-4-Py by typing the following. This assumes that you are currently in the directory: ~/EzQ/EzQ

   cd YOLO3-4-Py/
   

   Next, run the following:

   pip3 install .
   

   Ensure no errors occur. If there are errors: troubleshoot here.

5. Great! Now, you are all setup to run our scripts!

Scripting

This section will introduce in greater detail about our scripts as well as the running order.

1. ezq.py: Script takes an image from a web-cam and uploads it to firebase.

   Important

   Since the web-cam is not passed over, we offered a testing method that can be passed to the script as a form of testing for server.py

   To run this testing method:

   (assuming in /EzQ/EzQ/directory)

   python3 ezq.py --test
   

   This testing method sends over a test image (within the directory) over to firebase, which the script server.pycan be ran to run the detection model on the test image.

2. messenger.py: Script is ran in the background to provide messaging service

   Run it with

   python3 messenger.py
   

3. The following scripts must be ran separately but within the same step, before any inputs can be made (since they are gui-s). In short, they must be running prior to any inputs made.

4. • Customer.py: Front-end GUIfor Customers to input their hand-phone number

     python3 Customer.py
     

   • PIR new.py: PIR Sensor code to simulate power-saving feature

     python3 "PIR new.py"
     

   • cashier.py: Cashier side GUI for Cashier to submit orders and tag it with phone number inputs from customer.py

     python3 cashier.py
     

   • chefgui.py: Chef side GUI for chef to update completed orders, which triggers messenger.pyto prompt relevant customers through Whatsapp.

     python3 chefgui.py
     

   • server.py: Cloud based or Local Computer based Backendfor running of the detection module to identify number of people in queue.

     python3 server.py
     

Testing With Your WhatsApp Enabled Device

This section will provide the instructions to set up your Whatsapp enabled device to access our services.

Note for future versions These set up steps are required for the trial version of twilio. If this project recieves additional funding these steps will not be required.

1. Add +14155238886 as a contact

   You may want to save this contact as Twilio Sandbox. This is the contact where you can get the number of people in the queue You will be informed via this number when your food is completed

2. Open WhatsApp

3. Type join gasoline-egg and send

   This allows you to join the sandbox and receive messages from our free number

4. Type any message, for example hello and send it

   Whatsapp requires approval before sending messages that are unformatted.

   This approval is given the moment a message is sent to the Twilio API

5. Your number is now properly set up.

6. To obtain the number of people in the queue, type ? and send it to Twilio Sandbox

   Twilio Sandbox will reply you with the number of people in the queue