Inspiration
Researchers estimate that 32 million Americans have food allergies, including 5.6 million children under age 18. That’s one in 13 children, or roughly two in every classroom. And, It is very hard to manually identify and avoid food allergens from food every single time. In 21st century with advanced technology, we have to do better for helping people in needs.
What it does
LifeGuard is a mobile application automatically verifying if food contains an user's specific food allergens by simply taking a picture of his or her dish or scanning barcodes of a branded food product.
How it works
- Write out a list of ingredients users want to avoid
- Open the camera from the app whenever they want to check possible allegerns
- Select if they want to take a picture of a dish or scan barcodes of a branded food product
- Take a picture 5-1. (Food Capture) LifeGuard recognizes the name of the dish and figure out a possible recipe of the dish. Then, return a list of ingredients in the recipe. (Using ClarifaiV2 API, Recipe-Food-Nutrition API) 5-2. (Barcode Scan) LifeGuard recognizes the scanned food product and returns a list of ingredients in the product. (Using Nuritionix API)
- Finally, LifeGuard warns wheter the dish or the product contains possible allergic ingredients!
How we built it
We built an Android application by utilizing 3 exteranal apis on Rakuten Rapid Api marketplace and Firebase sdk. We used ClarifaiV2 API to recognize which food is in a picture, Recipe-Food-Nutrition API to figure out a possible recipe of a recognized dish, Nuritionix API to retrieve information of a food product with barcode, and Firebase sdk to scan barcodes.
Challenges we ran into
After extensively testing the food image recognition api, we realized a list of foods it recognized from a picture is widely ranged from an ingredient to a main dish without any hierarchical inforamtion. So, it failed to retrieve a proper list of ingredients time to time.
Accomplishments that we're proud of
We are proud of our innovative and practical approach to improve daily lives of people with food allergies with the utilization of cutting edge technologies including image recognition.
What we learned
We learned we have so many things we can do and help with our technical knowlege.
What's next for LifeGuard
In order to solve the problems mentions above and improve its usability, we will make food dictionary. It will contain the hierarchical relationship between ingredient and main dish. With the dictionary, we will be able to warn people with better accuracy and context.
Log in or sign up for Devpost to join the conversation.