Plate Guardian

Scan for allergens
Create meal plan
Create recipe

Inspiration

In today’s world, allergies have become a growing concern, especially amongst foods where it’s not so apparent what certain ingredients might contain, and it’s not always easy to detect allergens in such foods. There’s also a growing focus on nutrition and hitting certain nutrient/weight goals, and our app helps users with both of these concerns!

What it does

PlateGuardian allows users to upload an image file of a food label and their allergens, and generates an allergy report detecting whether the allergens are found within the ingredients of the food product. It also allows users to help meet their dietary goals, allowing them to enter certain macronutrients that they would like to incorporate in their diet and providing recipes containing those nutrients. Further, it also allows users to input a certain weight goal they would like to meet while inputting certain macronutrients they’d like to hit in their diet, and provides them with certain meal ideas that they can use to meal-prep for the week. This makes planning much easier while bringing users closer to their weight and nutrition goals!

How we built it

We utilized an image processing python library (easyocr) to convert an image file to text. Our frontend was designed using streamlit, which is a python library that transforms data scripts into web applications. We styled our application using the Streamlit theming feature, which enables us to design without using CSS. The backend was developed using python.

Challenges we ran into

Our team initially faced some difficulties converting our image to text. However, after finding the easyocr library and putting some time into learning how to work with it, we were able to utilize it to help us optimize our image to text conversions. We also faced some difficulties with installing certain libraries and technologies. We solved this issue by creating an individualized environment for our project and by then adding each library’s directory in our systems’ PATH variable.

Accomplishments that we're proud of

Creating a working web application in 36 hours.

What we learned

We learned so many new things about image processing as well as Langchain, and learned so much more about collaborating well as a team in a challenging, fast-paced environment to build a project from scratch and bring it to completion!