论文部分内容阅读
运用航空遥感图像为主信息源，获取立地要素信息，在ARC／INFO系统支持下，建立空间信息库，选取550个样本数据，训练自组织人工神经网络，然后对159个“未知”样本进行立地分类预测和容错检验。结果表明，该模型的分类、容错能力强，综合了遥感图像目视判读与计算机自动分类的优点，开拓了遥感与GIS技术相结合进行智能化土地条件分类研究的新途径。
Using aerial remote sensing images as the main source of information, we get the information of site elements. With the support of ARC / INFO system, establish the spatial information base, select 550 sample data, train the self-organizing artificial neural network, and then set up 159 “unknown” samples Classification prediction and fault tolerance test. The results show that the model has strong classification and fault tolerance, integrates the advantages of visual interpretation of remote sensing images and automatic classification by computer, and opens up a new way to study intelligent land condition classification by combining remote sensing with GIS technology.