基于监督分类矿山复垦土地利用分析

Analysis of land use of Mine Reclamation Based on supervision and classification

本文通过使用监督分类来解决大部分矿山复垦土地利用动态监测问题。首先,阐述监督分类理论基础;其次,利用遥感图像数据,根据矿区复垦土地利用现状,将矿区复垦土地进行分类,然后使用监督分类的神经网络方法,针对不同的影像特征进行遥感图像归类;最后,应用可视化技术对样本成果进行精度评价,制作相应的地图;结果表明,2018年矿区复垦土地的遥感图像分类变化明显,复垦土地变化不大。

In this paper,the dynamic monitoring of land use in most mines is solved by using supervised classification.Firstly,the theoretical basis of supervised classification is described;secondly,the reclaimed land in mining area is classified by using remote sensing image data according to the current situation of land use in mining area,and then the neural network method of supervised classification is used to classify remote sensing image according to different image features;finally,the accuracy of sample results is evaluated by using visualization technology,and the corresponding map is made;finally,the conclusion is drawn The results show that in 2018,the remote sensing image classification of the reclaimed land in the mining area changed significantly,and the reclaimed land changed ttle.