Publication Date
Fall 2025
Degree Type
Master's Project
Degree Name
Master of Science in Bioinformatics (MSBI)
Department
Computer Science
First Advisor
Dr. Philip Heller
Second Advisor
Dr. Nada Attar
Third Advisor
Dr. Wendy Lee
Keywords
Coral transplantation, Costa Rica, computer vision, image segmentation, deep learning, and machine learning
Abstract
Coral reefs are dying at a rapid rate and efforts are being made to preserve them in the ocean. The University of Costa Rica is raising coral nurseries in the ocean in Northwestern Costa Rica to replenish coral coverage lost due to unfavorable water conditions. Ideally, the coral fragments that are placed by the university are able to survive and grow on the structures they are attached to and, throughout the growth cycle of the coral, images are constantly being taken of these structures to monitor growth progress. Each of these images needs to be annotated to identify where corals are located in the image and then analyzed by a specialized team of bioinformaticians and marine biologists to track coral growth and experiment success. It is unrealistic to annotate each coral in these thousands of images by hand, so machine learning is used to aid in the process. This paper examines the application and performance of a computer program developed by Dr. Philip Heller at San José State University called Coral Vision that implements machine learning techniques to automatically identify coral in images to help automate the image analysis process.
Recommended Citation
Valderrama, Diana, "Coral Vision: A Machine Learning Approach to Aid Coral Restoration in Northwestern Costa Rica" (2025). Master's Projects. 1667.
https://scholarworks.sjsu.edu/etd_projects/1667
