Inspiration

Nutritional labels tell you about the % of the daily recommended nutrients and vitamins the food you consume provides you with. Now, what if there was also a way to track the environmental impact of the food you consume? The Footprint app allows its users to scan grocery store items and get their estimated carbon and water footprints so that they can make informed decisions regarding the food they buy. The app also allows its users to track their carbon and water footprints over time, to see the impact of their changes in lifestyle and diet. We aim to make a more sustainable life! :joy:

What it does

There are multiple functions inside the app, but our goal is to make it as simple as possible for the users. Here are the steps:

  1. Open the app to the home page
  2. Press "SCAN PRODUCT" button, which will navigate to the product scanner page
  3. On the scanner page, users can either use the camera to take picture of grocery products or select an image from their phone album.
  4. Not satisfied with the photo? No worries! We have a retake option!
  5. The "Use Photo" button on the button right navigates the user to the carbon footprint and water footprint data and the comparison with similar grocery items. This is accomplished by prediction the product using image classification. The image is uploaded to the cloud and processed. The prediction data is stored in output file that shows the grocery item detected.
  6. On the home page, the other button "VIEW PROFILE" links to a user's personal account to keep track of their carbon footprint and output a pie chart for different products that are documented. It also shows the percentage difference of the user's carbon footprint compared to that of the average American citizen.

How we built it

First, we brainstormed about the topic: "Sustainability" for innovative ideas and solutions. Later, we benchmarked through current apps in the genre for more inspiration and understand what is insufficient in the market that we could contribute to.

A Convenient and Informative Tool to Raise the Awareness of Sustainability

To save the user's time, our goal is to make it as handy as possible. We implement image recognition, since vision is the most straightforward information, and prediction, to save customer's time to type in brands and search for the right product. Charts and plots are shown for effortless understanding of carbon footprint and water footprint data. After having a direction to proceed, we started gathering data for data analysis. The datasets contain carbon and water footprint information. JavaScript is used to construct front-end coding and python in addition to ImageAI are used to compute the prediction. The app is made via expo and react.

Challenges we ran into

Some challenges we ran into involved being able to integrate our image classification model with JavaScript code and being able to run image classification real-time on a mobile device.

Accomplishments that we're proud of

For this project, all team members stepped out of their comfort zones and pushed themselves to work in unfamiliar areas all while having fun!! We are proud to have created a product that aims to bring sustainability awareness to the general population and are proud at how far we progressed in the development of our app.

What we learned

From a general perspective, we worked with data related to carbon and water footprints, and in the process, discovered interesting and shocking facts related to carbon consumption and emission and water usage. From a technical perspective, some members of the team had never worked with web or mobile development before, so we learned much about the coding details and all the new and exciting web development technologies that are out there. Additionally, we also delved into the existing research, models, and approaches in the field of grocery and retail image classification.

What's next for Footprint

Although we are proud of our work, there are a few functions that we would like to improve in the future:

  • Expand database and train more data for prediction
  • Upload real-time image to local server for faster image detection
  • Real-time video for camera and real-time image detection
  • Augmented reality icons to show if grocery items are sustainable

Built With

Share this project:

Updates