The farming industry is shrinking while the population is rapidly increasing; as a result, farmers need all of the support they can get for supporting their careers and their missions to feed the world. We wanted to develop a tool for farmers to use and make their lives easier for both themselves and the people they feed, and Automated Agriculture was developed.
What it does
This app helps save time for farmers by using computer vision to determine whether their produce is ripe, rotten, or have any unusual anomalies so that they don't have to go around their farm every day. Through machine learning and computer vision, the app creates graphs, uses location, and recognizes patterns to help farmers increase their day-by-day efficiency.
How we built it
We split up the front-end and the back-end work between the four members. The team members in charge of the back-end development coded and performed tasks such as computer vision training, machine learning graph plotting according to colors, and data translation. On the other hand, the team members in charge of front-end development worked on the UI, implementing features such as prototypes for working with the back-end data and determining conditions to suggest to farmers and working with React Native to have an app that would be deployable with both iOS and Android.
Challenges we ran into
Working with Google Maps API and PubNub, using external modules with the alwaysAI platform, and data analysis were most of our biggest challenges.
Accomplishments that we're proud of
We're very proud of the graph and data analyzation that we developed based off of the colors that we tracked from the images and produce we trained the computer with since this data works with both machine learning and data: a topic that all of the team members were very interested in working with.
What we learned
We learned a lot about machine learning and computer vision, which are also popular topics of today. Additionally, we were also able to learn as much as we could about working between front-end and back-end development, as most of us were relatively new to hackathons.
What's next for Automated Agriculture
We want to be able to train our program so that it can be the most accurate it can be. We want the data collected to be representative of what our product is about: efficiency. As a result, we're planning on using and improving the app's data analyzation so that it can be more automated, as well as have the data work with location services to provide the most comfortable and efficient experience for farmers.