MASK ALERT

by Ajith_V_Anil in Circuits > Assistive Tech

208 Views, 0 Favorites, 0 Comments

MASK ALERT

mask alert.JPG

Many of us forget to wear or take a mask when going outside our house. This project is to alert you when you are going out without a mask. This project is based on machine learning. The hardware part is fixed on the front door of our house.

Supplies

  • camera
  • Machine learning compactable board like raspberry pi
  • Buzzer
  • Power source

Understanding the Problems

Many of us forget to wear or take a mask when going outside our house or office. This project is to alert you when you are going out without a mask.

Project Objective

  • Our model needs to identify the face with a mask and without a mask.
  • The model needs to give an alert like sound or light when you are passing without a mask.
  • Need a kill switch or reset the model if the model gives the wrong data.

Basic Working

Basic working.JPG

Material Used

  • camera
  • Machine learning compactable board like raspberry pi

  • Buzzer

  • Power source

Working

The hardware part is fixed on the front door of our house (facing inside the house). The hardware part contains a camera and an electronic board with our Machine learning model. The camera is live all the time.

When a person comes in front of that camera the model will make a decision whether the person is wearing a mask or not. If the person is not wearing a mask it makes some sound using the buzzer or flask LEDs. If the person is wearing a mask then the model detects it and bypass it.

Machine learning model

The Machine learning model is made by giving different possible face sample with mask and without a mask. The demo is done using Google Teachable Machines

Making This Project Using Google Teachable Machines

  1. Open Google Teachable Machines by clicking this
  2. Click get started
  3. Sign in with your Google account
  4. Click Image Project
  5. Change the name of 'Class 1' to "With mask" (optional)
  6. Click Webcam
  7. Hold and record your photo with mask upload at least 100 photos for better result
  8. Change the name of 'Class 2' to "Without mask" (optional)
  9. Click Webcam
  10. Hold and record your photo without mask upload at least 100 photos for better result
  11. now click 'Train Model' (it will take time)
  12. now your model is ready.

Method

i6.JPG
i1.JPG
i2.JPG
i3.JPG
i4.JPG
i5.JPG

Teachable Machine Output

Teachablemachine output.JPG

Diagram

Diagram.JPG

Scope

  • We can connect with wifi for alerting with text message
  • Can be using in offices
  • Can be used in critical areas.