Inspiration
We found that the one common problem we all faced, aside from Mackey's homework, was keeping our bodies fit and healthy. The fact that so much information is available, and yet we are unable to process and use it, seemed too much of a shame. Our project aims to assist people to promote a healthy, lifestyle with minimal possible effort.
What it does
Our project is an Android App that keeps track of one's nutritional intake. Instead of manually the excessive amount of data, we used text-identification software and edge detection partitioning to automatically input and keep track of what we eat.
How we built it
We used Android Studio to build a application. None of us had previous experience with the software (some in java). So the whole 28 hour period was both hard and exciting.
Challenges we ran into
We made some assumptions of how accurate image-recognition software was. This meant that our goal was much harder to achieve than we had planned. We tried to use Google Cloud AI for this. However, we ended up using several different algorithms. We pre-processed the images before we were able to compute with Google Firebase.
What we learned
We've learnt that simply knowing concepts from lectures and lessons, such as OOP or Machine Learning, is not sufficient in helping us tackle challenging real-world tasks. The implementation of knowledge is just as, if not more, important than the knowledge itself.
What's next for NutriScan
We aim to improve our application, such as a social aspect where people can encourage each other to keep up with their healthy lifestyle.
Log in or sign up for Devpost to join the conversation.