AI-Powered Marine Life Identification: A Machine Learning-Based Approach for Detecting Endangered Sea Animals and Underwater Biodiversity

Identifying marine species is crucial for conserving biodiversity and protecting endangered sea life. This study introduces a deep learning approach for classifying six key sea animals-lobsters, crabs, eels, jellyfish, dolphins, and corals. The study employs ResNet and EfficientNet models, with the latter utilizing transfer learning to enhance performance. Using a simplified version of the SeaAnimals dataset, specifically curated to focus on these six classes, the models are evaluated on Core indicators, including recall, precision, and F1-score, support, accuracy, and loss. The EfficientNet model demonstrates superior performance, showcasing its capability to generalize effectively across diverse underwater imagery. This approach provides a scalable solution for marine biodiversity monitoring, contributing to the global effort to save endangered species and foster the conservation of marine ecosystems.