Coral Health Classification Computer Vision Project

The purpose of my model is to classify and distinguish between three different types of coral species: Staghorn Coral Acropora cervicornis, Brain Coral Diploria labyrinthiformis, and Elkhorn Coral Acropora palmata. All three coral species are deemed “healthy” corals (compared to bleached or dead corals).This model is the first iteration of being able to eventually be used to identify the health of different corals and reefs. A future model intends to classify coral health by distinguishing healthy living corals of these three classes in this current model, to bleached corals of these classes. While including these classifications, my aim is to train my model to identify the major health patterns among coral.The intended audience would be for scientists, conservationists, or researchers to help in field monitoring. Monitoring coral health can be an indicator of how the ecosystem is doing. Having a model that identifies coral health can indicate what issues may be affecting the environment. Scientists and conservationists can use this information to understand how environmental stressors are impacting reef health. Using this identification, scientists can create solutions to help support the reef based on its specific needs.Using this model as a monitoring system, scientists can have more insight on coral health, and can address coral bleaching as soon as there are signs. Having a model to classify these stages as they are changing gives scientists more time to support and conserve the corals and their ecosystem, in order to keep more corals healthy.