Marine Animal Monitoring, Anomaly Detection, Deep Learning, Artificial Intelligence, Object Detection Algorithms, Convolutional Neural Networks, Flask Web-Framework. Human activities at water bodies, such as concerted fishing and operating offshore windfarms, can have a negative impact on marine megafauna. Operations such as Surveying and tracking of marine species are done to reduce the impacts, which are frequently done at the locations where operations occur. Currently, tracking wild animals may be done remotely and the analysis of the recorded photos which are regularly undertaken at the sites of action. As of now, wild animal monitoring may be conducted through remote means, and the analysis of the obtained images taken by high precision cameras is possible to be automated using object detection algorithms and the development of machine learning techniques. Thermal cameras are used to track warm blooded species. But most of species possesses cold blood feature. Also, it causes people who swim near the coastal areas, so it is important to develop pattern recognition techniques to identify marine species and alert swimmers when they approach the coast. This project evaluates several object recognition methods of the machine learning algorithms and proposes the optimal model with respect to the performance evaluation.