If you have any questions please contact me on Discord: Chisa#6807
Inspiration
As we were meeting virtually for the first time to come up with ideas, Chisa drew inspiration from the coffee on her desk. We realized programmers and other professionals rely heavily on caffeine to stay awake, which can have negative side effects. Yet, companies are not required to list the caffeine contents of their products, making it hard to make caffeine-conscious decisions. As a result, CaffeineCulator was born to make caffeine tracking easily available.
What it does
The user has the option to either scan a product’s barcode or search for a food or beverage. The data is parsed and input into Nutrionix’s database, getting the proper caffeine amount. The user is then presented with their Caffeine Receipt, which tells them how much caffeine they consumed in that one product along with how much caffeine they can safely consume in the rest of the day. Caffeine intakes are tracked in the browser for 24 hours to keep track of daily intake.
How we built it
We split the project into front-end and back-end tasks. We created the front-end with HTML and CSS after going through a few site layouts, and added the in-site camera to get barcodes with Javascript. We configured hosting with Flask and set up barcode reading, product database searching, and nutrition database searching with Python using various libraries and the Nutritionix API. Through Flask, we were also able to assign cookies to keep track of daily caffeine intake.
Challenges we ran into
We initially wanted to use image detection algorithms to identify caffeinated products. However, we quickly ran into challenges in the specificity of identifying between similar products, along with the image detection algorithms finding a lot of excess information that was difficult to parse. We then attempted text-detection, soon finding that branding on some products makes it impossible to get a good read on. Lucas came up with the idea of reading barcodes, and through testing it worked out with great success.
## Accomplishments that we're proud of We learned how to use an array of unfamiliar programs and libraries working together in a short period of time. We are really happy with the way the front-end turned out, along with how we managed to link so many aspects of code together.
What we learned
We learned a lot about various programs and libraries and got to hone existing skills. Though we did not implement them in the final product, learning how to use Google Cloud and computer vision algorithms was a great educational experience. Through mentorship, we also all increased our knowledge of connecting various parts of code together, such as Javascript to Python and vice-versa, which is something we all mentioned we struggled on at the beginning of the project.
What's next for CaffeineCulator
A mobile application for easy, on the go caffeine tracking. Dark mode option to toggle between. Greater customization of caffeine products (mass/volume of intake, number of servings, etc). Greater personalization (ex. Keep track of the user’s name to leave a welcoming message). User accounts and databases to display data of caffeine intake over time.
Log in or sign up for Devpost to join the conversation.