Abstract:The three-dimensional temperature and salinity field information of the ocean is an important parameter for describing the physical properties of the ocean and mastering the process of ocean physical movement. Obtaining accurate three-dimensional temperature-salinity information of the ocean is of great significance for understanding the ocean, developing and utilizing the ocean, and marine scientific research. With the development of artificial intelligence and deep learning methods, the intelligent detection of subsurface three-dimensional temperature and salinity fields using deep learning methods has become one of the hotspots. This article reviews the research progress related to the intelligent detection of ocean three-dimensional temperature and salinity fields from the perspectives of ocean temperature and salinity observation datasets, traditional machine learning methods for three-dimensional temperature and salinity intelligent detection, general neural network three-dimensional temperature and salinity intelligent detection, and deep learning three-dimensional temperature and salinity intelligent detection. Finally, the paper summarizes and looks forward to the problems and future development trends of intelligent detection of three-dimensional temperature and salinity fields.