Mybot (Putt and Max)

Inspiration

We were very interested in the technology of image classification and object detection as it has lots of benefits for many industries including for our project seeds. The reason we chose the topic related to healthcare and safety, specifically mask detection is because there is a global pandemic occurring right now thus we wanted to do something that is related to current events. In addition, a mask no mask image classification is very easy and beneficial for us to try out and learn new information.

What it does

Mask or no mask detection using image classification. Can detect still images and live videos.

How we built it

Python, keras, tensorflow, opencv. We had help from our project's developers. The AI concepts used include Image classification, CNN, maxpool etc...

Challenges we ran into

I think that although the algorithm that we produce was simple, we faced multiple challenges along the way, most of which are related to methodology and coding. Luckily we were able to use resources like the internet, Github and our developer friends to assist with our project. Another challenge we faced was making the algorithm accurate, which required lots of training data to work. Overall I think that these challenges have helped us stay outside our comfort zone and learn new and interesting things.

Accomplishments that we're proud of

What we learned

Our learning experience includes understanding the coding methodology and the different AI techniques that are used for image classification, making the algorithm more complex and accurate through various improvements in the AI database. We understood the uses and benefits of AI and machine learning, and how it can impact various industries through proper implementation. Finally, the most important learning experience for our group is how image classification is applied into our other projects.

What's next for Mybot

Our project’s expansion can be categorized into 4 parts. First, we will be able to identify the symptoms by detecting their temperature or whether they are coughing. Second, is classify, which the algorithm will be able to identify the demographic of the person. Third, we can track where the person went within the monitored zone. And finally, we would like to increase the accuracy of our program so that it can detect masks in large groups of people, while also using machine learning at the same time.

Built With

• keras
• opencv
• python
• tensorflow