Overfishing is currently one of the most devastating issues that faces our oceans today. Fish that are endangered and threatened are rapidly decreasing among the world's oceans as recreational fisherman and commercial fishers continue to catch them using bad practices. We wanted to create an app that makes it easier for them to be educated and use sustainable practices as they continue fishing.
What it does
Fish Detect uses machine learning and artificial intelligence to allow the user to take a picture of the fish they catch and the app will tell them what fish they caught and information about it, including whether it is okay to consume or not.
How we built it
We used Xcode to create this single view iOS application and used the brand new framework from apple, Create ML, as well as their Core ML modules. Fish data and information was collected from the Monterey Bay Aquarium Seafood Watch Guide and other sources.
Challenges we ran into
We were unsure how to exactly go about creating this project as neither of us had experience with Core ML and Create ML modules and APIs. Create ML is a new technology, only about a few months old, so it was difficult to find resources. We did not have any iOS devices running iOS 12 so we could not run the app until morning. In the end, we were only able to get four fish on the database but hope to add more in the future.
Accomplishments that we're proud of
Throughout this process we were able to better understand Swift and learn how Create ML and Core ML functions. Only one of us had some experience with iOS development so we were able to learn more about Swift, its UI capabilities, and new data science components.
What we learned
This was our first hackathon experience and we were able to learn more about what it takes to develop an app in a very short time period. We learned how to create an entire iOS app an implement its back end and front end code. Create ML allowed us to learn more about data science and machine learning as well.
What's next for Fish Detect
We definitely plan to expand our database to include more fish information specific to the state of California. In addition, we will implement better UI components and work more on the front end programming. A possible incentive may be added and make it more game-like to attract a greater audience. Hopefully we can also pitch this to the Monterey Bay Aquarium in the future!