Although coral reefs play an important role in maintaining biodiversity on our planet, they are under severe threat from climate change, pollution, and overfishing. Evaluating and monitoring coral reefs is essential to their survival. Traditional methods of tracking and monitoring reefs is tedious and expensive. This research proposes a real-time monitoring system for coral reefs using YOLOv5, a deep learning based object detection algorithm, which allows for fast and accurate species identification and health assessment of corals. The addition of temperature, salinity, and pH to the system will help to classify the health of the coral more accurately. The effectiveness of the system was tested using a labelled dataset of coral images collected via underwater drones and remote sensing. Research shows that, among the many techniques, YOLOv5 stands out as the most accurate in recognizing different species of corals and detecting the initial stages of bleaching, proving the model’s capability for real-time monitoring and management of coral reefs.