A Practical Method of Feature Extraction for Water Surface Temperature Based on Thermal Infrared Remote Sensing Image

In this paper, the retrieval model of water surface temperature is studied on the base on ASTER Thermal Infrared Multispectral data. The model of the radiation brightness switching, water surface brightness temperature and water surface temperature are constituted for the ASTER thermal infrared Multispectral TIR. The practical model base on the mono-window algorithm is detruded to be fit for the temperature retrieval on the water surface temperature, it need not estimate the temperature parameter. Not only simplifies the regression estimate but also increases the retrieval precision of the mono-window algorithm and reliability. The Haizhou Bay area of sea surface temperature are extracted, and the arithmetic precision are estimated to make use of the observation data of the weather.