In recent years hyperspectral remote sensing has been widely used in the applications of geology agriculture forest ocean etc. This work assessed the feasibility of hyperspectral technique in coastal zone remote sensing. Data was acquired by Operational Modular Imaging Spectrometer (OMIS). Field spectrum of each class was measured and analyzed to extract certain spectral feature. We proffer a hybrid decision tree classification algorithm combined with optimum spectral features of class pairs to every tree node and step by step classified out coastal vegetation water body rocky shore sand beach mudflat and artificial objects etc. The results show that hyperspectral data can be used to classify coastal landscape more accurate than multispectral image.