We wanted a quick and easy way to determine if food items are compatible with people's dietary restrictions.
What it does
The API takes an image of an ingredient list and perform an OCR scan, which extracts the ingredients and checks for commonly restricted food items such as allergens and certain meats.
How I built it
The API is powered by node.js and makes backend calls to Microsoft's Azure Computer Vision OCR library.
Challenges I ran into
Our main challenge was filtering out ingredients from other non-relevant text found in most pictures.
Accomplishments that I'm proud of
We are especially proud of our ingredient-filtering algorithm.
What I learned
We learned how to create scalable yet simple APIs.
What's next for Ingredient Reader
Next, we want to improve the OCR capabilities of the API while enabling it to recognize a wider variety of allergens and other restricted ingredients.