Marine Ecosystem Monitoring: Applying Remote Sensing and AI to Track and Predict Coral Reef Health

Coral reefs are essential ecosystems, supporting a diverse range of marine life and offering considerable ecological, economic, and cultural benefits. However, they face increasing threats from climate change, pollution, and other human activities. Thus, effective monitoring and management of coral reefs are crucial for their conservation and sustainability. In this study, we employed a hybrid model, HCNN-SVM, which combines convolutional neural networks (CNNs) for feature extraction and support vector machines (SVMs) for classification. We utilized a coral-reef dataset from Kaggle, containing images from CoralNet and the Moorea Coral Reef Long Term Ecological Research (MCR LTER). To enhance image quality and ensure accurate reflectance values, we applied radiometric, geometric, and water column corrections during preprocessing. Feature extraction involved the use of spectral indices such as the Coral Bleaching Index (CBI) and Normalized Difference Vegetation Index (NDVI) to detect healthy and stressed corals, along with texture analysis to differentiate substrates. The model was trained on this diverse dataset, which captures various environmental conditions and reef types, achieving an accuracy of 98.55%. Cross-validation methods were employed to evaluate model performance, ensuring robustness and generalizability. The HCNN-SVM model demonstrated high accuracy in identifying and mapping coral reef components, making it a powerful tool for monitoring coral reef health and supporting conservation efforts.