SpoilerAlert

The SpoilerAlert UI
The Computer Vision model
The team!

Inspiration

Food recalls are becoming more frequent, yet many shoppers remain unaware of them until it’s too late. Manually checking barcodes or browsing through FDA recall lists is time-consuming and impractical. We wanted to create a simple, efficient, and accessible way for consumers to verify food safety instantly—giving them peace of mind while shopping or checking their fridge.

What it does

SpoilerAlert uses computer vision and the FDA’s recall database to let users quickly scan food items and check if they are part of a recall. It provides instant feedback on whether an item is safe to buy or consume. Additionally, SpoilerAlert tracks how many users have found a recalled product still for sale in their area, raising awareness of food safety issues and preventing unnecessary food waste.

How we built it

We developed a computer vision model to recognize food items and match them against FDA recall records. We integrated this with the FDA recall database for real-time updates. The user interface was designed for seamless scanning, and we implemented a tracking system to show how many users flagged a recalled product in their city.

Challenges we ran into

Ensuring high accuracy in the computer vision model, especially for varied packaging and lighting conditions
Keeping the recall database up to date in real-time
Designing a user-friendly interface that makes scanning fast and effortless
Balancing between false positives and false negatives in recall detection

Accomplishments that we're proud of

Successfully integrating computer vision with a real-time recall database
Creating a functional prototype that can accurately identify and flag recalled products
Building a system that not only protects consumers but also raises awareness about food safety and waste
Developing a community-driven aspect where users can report and track recall sightings in their city

What we learned

The importance of real-time data in food safety applications
How to fine-tune a computer vision model for varied real-world conditions
The significance of an intuitive and seamless user experience
The impact of community-driven reporting in solving widespread problems

What's next for SpoilerAlert

Improving model accuracy by enhancing the recognition system to reduce errors
Developing a mobile app for more accessibility and ease of use
Expanding to medicine recalls by adding FDA-recalled medications to protect consumers beyond food products
Notifying retailers if recalled products are still on shelves
Partnering with grocery stores and food safety organizations for wider adoption