ScanHealthy

Web Interface

Inspiration

Arriving at PolyHacks and looking for a team we had breakfast and glanced at the nutrition fact labels there.. this kindled our research interests in optical character recognition, image segmentation and data processing.

What it does

The ScanHealthy app simply reads nutritional information from standard labels on food items and compares input data with recommended intake from WHO and governmental guidelines.

How we built it

We assembled open-source pre-trained data models and APIs to enable our online image processing pipeline.

Challenges we ran into

Python dependency hell makes a striking comeback! We tried to divide responsibilities but we ended up all working and debugging on the same packages, we managed to develop in parallel tracks at times though.

Accomplishments that we're proud of

We made it run and that counts for something, we're especially proud of the fact that it worked with nutritional labels from the food we've gathered here at PolyHacks.

What we learned

Overview of image processing and ML technologies that have an increasing impact nowadays. Angular UI and service development; server and DNS deployment.

What's next for ScanHealthy

After this condensed sprint, we endeavour to add a lot more features to our app to make unique and interesting including integrations with health app providers and databases such as the Open Food Facts platform that we based our work on.