BingNutrition

Automatically scrapes Sodexo menus, uses OCR to extract the items from the images, retrieves nutritional info from the FDA database, and packages it all in a pretty website.

Menu found from scraping
Data Visualization

Key Aspects

Web Scraping
Image Processing
OCR
API interfacing
Database management
Interactive frontend
Fully integrated Python/Javascript stack

Built With

Submitted to

HackBU 2024

Created by

Worked primarily on front end using ReactJS and Tailwind, data visualization, parsing the results into an easily interactable interface.

Allen Domingo
I worked on web scraping using the bs4 library to parse pdfs links to the OCR, and used regex to extract the food price pairs provided from the text output from the OCR

Jake Juettner
I primarily worked on interfacing with the FDA nutrition API, as well as processing the returned data and handing it off to the JS frontend. Additionally, I decided on and implemented much of the data formatting and structure to ensure that all stages of the project worked seamlessly with each other.

Maxwell Parker-Blue
Worked on the backend, utilized Python libraries to manipulate webscraped data and extract its textual information using OCR, assisted with regex parsing of raw text to isolate menu item/price pairs.

Tyler Lippold

Updates

Jake Juettner started this project — Feb 18, 2024 09:57 AM EST

Leave feedback in the comments!

Log in or sign up for Devpost to join the conversation.