Fishie

• 

  A roundscale spear fish used for the ML training

• 

  A White marlin used for the ML training

Inspiration

 We were originally inspired in part to help the environment and one of the best ways to do so is by gathering accurate real-time data about various different species and how they react to the current global climate.

What it does

 Currently, It will allow users to take photos and cross reference them with a database of different species of fish and and, in the future, diversify to different marine species. The app also record prevalence of a species in the location of the fisher. 

How we built it

 We utilized Java and the google cloud platform to initiate a machine learning algorithm which predicts the species presented by a user image. Android Studio was used to built the bulk of the application. 

Challenges we ran into

 Installing the sdk for both android studio and GCP was a main cause of concern as well as the issue with the auto-ml api in trying to train it to detect between predetermined species of fish.

Accomplishments that we're proud of

 We were able to put together the app and connect android studio with the GCP.

What we learned

 Although many fishing apps exist, ours is much needed because the internet barely has data on rare species of fish that need dire attention for being studied on. 

What's next for Fishie

 The next attempt is to connect Fishie with researchers and scientists to further the goal of real-time status of various fish in different biomes in the world. 

Special Mentions

 Rozanna Abramtsis contributed her efforts greatly along with the mentors at HackDavis, so would like to thank them for their hard work and dedication!

Built With

• android
• auto-ml
• gcp
• java