Multi-scale approach for coral condition assessment and Drupella sp. identification using a versatile low-cost camera...

Area of Science:

• Marine biology
• Ecosystem monitoring
• Conservation technology

Background:

• Coral reefs face increasing threats from climate change and human activities.
• Traditional monitoring methods are often inefficient, lack coverage, and provide limited spatial data.
• Conservation organizations need improved tools for habitat assessment and corallivore monitoring.

Purpose of the Study:

• To develop a low-cost, versatile underwater camera system for coral reef monitoring.
• To integrate Structure-from-Motion (SfM) photogrammetry and semantic segmentation for 3D habitat reconstruction and analysis.
• To enable efficient, spatially explicit monitoring of corallivore dynamics and coral health.

Main Methods:

• Developed a multi-mode underwater camera system (surface, horizontal, curved) with three synchronized cameras.
• Utilized SfM photogrammetry to create 3D point cloud reconstructions from video data.
• Applied a KPConv semantic segmentation model to classify point clouds, identifying Drupella snails and compromised coral.

Main Results:

• The camera system provided high point densities at a lower cost than conventional methods.
• The KPConv model achieved 0.916 overall accuracy and 0.601 mean IoU for segmentation.
• The system successfully identified compromised corals and cryptic Drupella snails in field surveys.
• Quantitative outputs enabled detailed coral condition assessments and early warnings for Drupella outbreaks.

Conclusions:

• The integrated hardware-software system offers an efficient and cost-effective solution for coral reef monitoring.
• Spatially explicit insights are generated, particularly valuable for understanding small-scale corallivore dynamics.
• This technology supports community-based conservation efforts by improving survey efficiency and data quality.