Automated image processing: Algorithms can be trained on extensive datasets, such as the open-source FathomNet, to automatically detect and classify marine species from underwater images and videos. For instance, AI is being used to monitor coral reefs for signs of bleaching or disease and to track the populations of fish and endangered species. This technology can reduce the human effort required for image analysis by a significant margin.
Acoustic monitoring: AI analyzes soundscapes collected by hydrophones to identify, localize, and track vocalizing marine animals like whales and dolphins. Machine learning algorithms can distinguish the calls of different species or even individual animals, providing crucial insights into their behavior, migration patterns, and communication. Projects like SurfPerch, developed by Google Research and DeepMind, have used AI to create bioacoustic libraries that assess coral reef health based on ambient noise levels.
Habitat mapping and exploration
AI is enabling unprecedented views of the ocean floor and crucial habitats, which helps researchers understand the distribution of marine life and the impacts of human activity.
Autonomous underwater vehicles (AUVs): AI-powered AUVs can autonomously navigate and map the deep sea, collecting high-resolution sonar and imaging data. These algorithms can process data in real time, identify points of interest, and adapt their missions on the fly to investigate new discoveries.
Coral reef and habitat mapping: The Allen Coral Atlas uses machine learning to map and monitor the state of coral reefs by comparing satellite imagery with field data. By analyzing satellite images, AI can also detect changes in marine habitats like mangroves and track the health of coastal ecosystems.
Predictive modeling and conservation
AI's predictive capabilities are providing marine biologists and conservationists with tools to forecast changes and manage resources more effectively.
Climate change impacts: Machine learning models predict how phenomena like ocean warming, acidification, and changing currents will affect marine life, including shifting fish populations and the spread of invasive species.
Illegal fishing detection: AI systems analyze satellite imagery and vessel data to identify anomalous behavior indicative of illegal, unreported, and unregulated (IUU) fishing. Organizations like OceanMind use this technology to alert authorities to potential violations, protecting marine resources and biodiversity.
Bycatch reduction: AI-powered cameras on fishing gear can identify non-target species in real time, alerting fishers and enabling them to adjust their methods to reduce bycatch.
Challenges and the future of AI in marine biology
While AI is a powerful tool, its application in marine biology is not without challenges. These include the high initial cost of technology, the need for high-quality, labeled datasets, and ethical considerations surrounding its deployment. Data collected from the marine environment can be particularly difficult to acquire and interpret due to the vast scale and variable conditions.
Looking ahead, advancements in edge AI, multi-modal systems, and increased collaboration among scientists, technologists, and policymakers will further integrate AI into marine research and conservation. AI will continue to be a crucial ally in our efforts to understand, predict, and protect the world's oceans in the face of accelerating environmental change.
Despite significant advances, artificial intelligence (AI) faces several fundamental limitations when analyzing underwater images for marine research, many of which stem directly from the unique and challenging nature of the marine environment. Unlike land-based imagery, underwater footage is often characterized by poor visibility, distorted colors, and a dynamic, unpredictable setting that can mislead even the most sophisticated AI algorithms.
Environmental distortions
Poor image quality: The ocean's properties—specifically the absorption and scattering of light—are the single greatest challenge for underwater AI. This leads to images with low contrast, reduced resolution, and a color cast (often blue or green) that varies with depth. These factors can create blurry images and obscure details, making accurate object detection and classification extremely difficult.
Irregular lighting and shadows: The availability and quality of light underwater are highly variable. Images can have uneven illumination, harsh shadows, and intense backscatter from artificial light sources. This inconsistency makes it hard for AI models to generalize, as they might struggle to perform accurately in a different location or time of day.
Data-related challenges
Insufficient labeled data: For AI models to learn to identify marine species, they need large, diverse datasets of e