-
-
Vestigium: Use ML to Track Your Foodprint!
-
Inspiration - Part 1
-
Inspiration - Part 2
-
Inspiration - Part 3
-
Process
-
Target Audience
Inspiration
--> Persona 2
Carbon emissions significantly affect the planet. It is the greenhouse gas with the highest levels of emissions in the atmosphere. As a result, the Earth must quarrel with global warming, including extreme weather events. Notably tropical storms, wildfires, severe droughts and heatwaves. Furthermore, there are people like Joe everywhere, whose lives and cities are on the brink of disaster. Any strategy that reduces our emissions must be implemented. In fact, livestock agriculture is the cause of roughly half of all man-made emissions. Hence, Vestigium was born!
What it does
Vestigium uses machine learning algorithms to identify ingredients in a picture. Once this process is completed, the discovered ingredients are cross-referenced against a database containing carbon dioxide equivalent values. Finally, the app informs the user of the total carbon footprint or foodprint that their meal has on the planet. Not to mention putting it into perspective by referencing the distance a fuel powered car would travel.
How we built it
To build Verstigium, we built a website using HTML and CSS. We also incorporated python data mining to create a JSON database for the ingredient carbon equivalent values. Furthermore, we made use of a machine learning API to detect ingredients in an image. Finally, we combined all of our features, and used flask, to connect our front-end to the back-end, allowing for a smooth user-experience.
Challenges we ran into
Despite searching for publicly available carbon equivalent databases, we were not able to find an accessible solution. Thus, with no other answer in sight, we used python data mining to artificially create a database ourselves. Another challenge was the integration of the back-end and front-end using Flask as we had some issues loading functions and other processes. Nonetheless, we tried our best and are proud of our what we made!
Accomplishments that we're proud of
We are proud of the python data mining that we accomplished, along with utilizing a powerful API to further develop and enhance our project.
What we learned
While making Vestigium, we learned that each team member should be allotted a task, which helps to increase productivity and efficiency. More importantly, we realized that it is important to work with people who have different skillsets, which allows the project to be both completed in time and more enhanced. For instance, our hacking group was divided into three sets.
What's next for Vestigium
The next step for Vestigium would be to release an app for mobile devices that integrates the user's smartphone camera, allowing for quick and fast tracking.
Log in or sign up for Devpost to join the conversation.