山区林地多季相TM影像综合分类研究

Study on Integrated Vegetation Classification by Using Multi-temporal TM Data in Mountainous Area of Northeast China

为了提高陆地卫星TM数据在山区林地的分类精度,利用4个季节的帽儿山TM数据进行植被分类,并辅助以物候特征和地面GIS专题信息 采用由简单到复杂的信息提取过程,先利用基于光谱知识的林地提取模型提取林地边界,再用有监分类方法分别进行林地和非林地内部类型信息的提取,生成多季相综合分类图 分类精度比单时相提高了19 6%

In order to improve the vegetation classification accuracy of Landsat TM data in mountainous area,vegetation classification supported by GIS data and phonological information is made by using four-seasonal TM data of maoer mountain in this paper. The procedure of information extraction is from simple to complex: firstly, forest area boundary was extracted based on the spectral knowledge;secondly, supervised classifications were performed within the forest area and non-forest area respectively. In this course,four vegetation classification maps were produced from four TM datasets;finally,integrated vegetation map was derived through GIS analysis model. The overall accuracy is increased by 19.6% compared to the classification result of single seasonal data.