通过统计分析多景MODIS图像,提出了一种基于MODIS图像识别I类水体和II类水体的经验方法。该经验方法是利用I类 水和II类水在可见光区的反射率差别大、而在近红外波段两者的反射率基本相同的特性来完成的。把该经验方法用在了中 国渤海区域,对不同年份的两景MODIS图像进行了I类水与II类水识别,并对结果进行了分析,结果证明该经验方法是正确和有效的。

By statisically analyzing many MODIS images, an empirical algorithm is described, which is used to determine the case II water from the MODIS image. The algorithm takes advantage of the characteristic that the apparent reflectances of the case I and case II waters are the same in SWIR band, but considerably different in the visible band. The empirical algorithm was applied to MODIS images over Bohai, China, the waters in Bohai were classified into two types: case I and case II. The results show that the proposed classification empirical algorithm is precise and effective.