Ai Mask detector - SCOPES Digital Fabrication

Lesson Details

Age Ranges
Fab Tools
Additional Contributors

You need to login or register to bookmark/favorite this content.


Studio 5/6
Studio 5/6
K-12 teacher
In line with the Ministry of Communications and Information Technology’s Digital Youth Strategy, STUDIO 5/6 is mainly aimed at nurturing youth as digital learners by sharpening their 21st century learning skills, as they develop in an all-pervasive digital environment. In… Read More


this is 1-hour online lesson to teach participants about machine learning through practical activity by using the online tool – MIT – Personal image classifier.

What You'll Need


Internet connection

The Instructions

AI Mask Detector

Machine learning is a branch of artificial intelligence based on the idea that systems can learn from data. Participants in this session will get introduced to machine learning in which they will train a model to recognize and people's faces who are wearing a safety mask or who are not, by using a web-based tool.


  • Introduce participants to the engineering design challenge 
  • Explain what is image classification
  • Discuss the ability and limitation of a machine to classify images.
  • Introduce machine learning and give an example of the mask detector.
  • Introduce a video about ML
  • Discuss the use of AI in our daily lives.
  • Introduce personal image classifier tool.
  • Explain step by step how to train a model.


Share the tool link and ask them to create their own prototypes.

  • Create labels
  • Add pictures to train the model
  • Testing their models.

Wrap up

  • Ask participants to share their results and discuss how human bias plays a role in machine learning.




ask participants

Can a computer recognize the photos and emotions like us?


Is this hard or easy for a machine? 



Have you heard about machine learning?

What do you know about machine learning?

What do you understand from the picture?





Explain traditional programming and machine learning from the point of paper rock and scissor game.

Say >> Instead of thinking all about of these rules and we have to write and express all of these rules in a code, what if we could provide a lot of answers and we could label those answers and then have a machine infer those rules

Say >> For example, instead of writing a complicated code to describe the shape of a hand for the computer, we can just get a lot of images of people doing a rock so we can diverse hands, diverse skins, and tones. and then we just say to the machine hey this is what rock looks like, this is what a paper looks like.