介绍DavidW.Goodall的基于概率的相近指数理论,研究它被应用在遥感图像和其它空间数据综合分类中的可能性,并首次在GRASS环境下实现了基于DavidW.Goodall的相近指数的遥感图像和其它空间数据综合分类算法,并对该算法进行了测试,将分类结果与其它几种较流行的分类方法结果进行了比较。

Today many methods have been used in classifying remote sensing images. However, developing classification algorithm which is capable of processing both images and other ancillary spatial data still remains to be an active research area. In this paper, the affinity index of David W. Goodall based on probability was explained, and its application possibility in remote sensing and other spatial data integrated classification was studied. Based on Goodall's affinity index, a computer program for classifying both remote sensing and other spatial data was developed within GRASS environment. To see the result of this program, it was tested in a case study area and compared with other popular classification methods such as maximum likelihood classification and evidential classification. Through this study, we would like to know how the other spatial data can help improve the remote sensing image classification and whether the algorithm based on Goodall's affinity index is good in classifying remote sensing images and other ancillary spatial data in an integrate way.