论文部分内容阅读
遥感影像计算机自动分类技术的基本原理,是根据地物类别在遥感影像上反映出来的光谱、纹理和时相等特征而对地物进行分类的,并在影像上确定每一地物类别的平面分布。以光谱特征为基础的模式识别方法,是当前遥感应用中最常见的计算机自动分类方法,它是根据不同地物类别的不同反射光谱特性、以地物影像光谱特征的相似性作为判别
The basic principle of computerized automatic classification of remote sensing images is to classify the objects according to the characteristics of spectrum, texture and time reflected by them in the remote sensing images and to determine the plane distribution of each feature category on the images. The pattern recognition method based on spectral features is the most common computer automatic classification method in current remote sensing applications. It is based on the different reflection spectral characteristics of different feature types and the similarity of the spectral features of the image features