School Smart

To revloutionze the school system


Looking at the current schooling system we are in, we realized a few issues and places that can be improved, such as attendance, lunch pin automation, text scanning, and school security. We found that teachers normally took valuble class time to take attedance using a long time, and that in lunch lines, the fact that we are entering our pins to a pin pad makes it unsanitary , slow, and easy to spill food. Moreover, school security is curcial to a safe school enviornment! On the other hand, we realized that some teachers need to scan their text and upload it to a Google doc, or some visually impaired students may need to understand what is on a piece of paper.

What it does

Our application has four main features built upon the fact that our code would run on a security camera, an attendance feature that uses facial recogntion, lunch pin automation that also uses facial recognition, a school security camera logging system where it will log who entered the building at what time, the gender of the person, and a photo of them, and lastly a built-in text scanner that uses ml for teachers to scan documents easily.

How we built it

For the lunch pin automation and the attendance features, we incoorperated tensorflow, opencv, facial-recognition, numpy, firestore, and datetime to recognize faces and communicate with firestore.

On the other hand, we used Google mediapipe, facial-recognition, clib and all the packages used in our facial recognition scripts to detect, compare, and log data locally and on the cloud

Lastly, the text recognition system is using the latest local flutter ml.

Challenges we ran into

There were a lot of challenges we ran into, the most significant ones are package version control and the logic behind the communication between front end and backend using firestore, and the misnaming of a few functions.

Accomplishments that we're proud of

We are really proud of our flawless communication between frontend and backend using firesotre, we are also proud of the fact that we made this under 2 days.

What we learned

The most crucial thing we learned through this hackathon is the power of cooperation and the importance of communication between one another. One of the team member learned alot of flutter ml and the other learned a lot about the latest and most developed ml and computer vision packages on Github and on pip.

What's next for School Smart

We are planning to integrate image display on our logging system using 64byte string encoded images uploaded to firebase.

Some more ideas are

  • more advanced and less laggy image logging system
  • teach attendance system
  • total automation with food detection.

How to use the demo link

When prompt with a login screen, please enter the following credentials email: password: coolnice

Built With

Share this project: