Beautify

A fun and interactive way to clean and better your community. Join others in your local area in helping to reduce landfill and litter while earning fulfilling rewards along the way.

Comment

Inspiration

Beautify was largely inspired by three big passions we all shared. The first was our belief in promoting cleanup for the environment and improving the health of our community. The second was in the power of collaborative efforts between people and friends to accomplish significant things. And lastly, in exploring new areas around us that are often overlooked. In particular, the idea of Pokemon Go really provided inspiration for its ability to allow people to visit new places and meet new people while having fun.

What it does

At it's core, Beautify is an app that aims to encourage people to clean up and "beautify" the community by picking up litter or removing debris. Using an interactive map, players can travel to predetermined locations where they can help in cleaning up environmental waste and help reduce pollution. Users are encouraged to help clean up through a rewards system where they can earn points to redeem prizes such as free trip to the museum or a discount on a local favorite coffee shop. The app includes an image verification system to all users to submit pictures of their progress, as well as to identify areas to clean up.

How we built it

To generate the interactive map containing dynamic site locations, a MongoDB Atlas cluster was created to store location information such as name and coordinates. Interacting with the database is a custom made RESTful API which performs CRUD operations (Create, Read, Update, Destroy) on the collections. This API interacts with the Beautify App to provide nearby cleanup site locations was written with C# and utilizes the PyMongo library for processing. The Beautify app implements this API, along with several Google APIs for user location and map data, and computes all user information locally. It was written in Kotlin on Android Studio, and was contributed to by all teammates through Git. Tensorflow lite was also used to train a model to detect waste from images, and would be used to verify photos taken in the Beautify App.

Challenges we ran into

Our team of three freshmen, with two aerospace majors and one computer science, did not have any prior experience in the realm of Android app development. It was definitely a steep learning curve learning how to use MongoDB, coding the RESTful web API in C#, applying beginner Tensorflow knowledge to train a trash-detecting model, and building an Android app in Android Studio coded in Kotlin, all from scratch. After 36 hours, complete with an all-nighter together at the library, we overcame error after error to fulfill much of what we sought out to do.

What's next for Beautify

In the near future, we hope to improve the functionality of Beautify even more to encourage collaboration among friends. In addition to expanding the use of the web API and database system, we hope to apply machine learning to a greater number of applications (due to Android Studio file size constraints, our trained model could not be implemented in the current version of Beautify). We hope to partner with local governments and environmental groups to further help clean up communities across the world, while providing greater rewards for users too.

Share this project:

Updates