| 皇甫润,李傲.基于遥感影像的矿区地物信息提取方法对比分析[J].唐山学院学报,2020,33(3):42-46 |
| 基于遥感影像的矿区地物信息提取方法对比分析 |
| Contrastive Analysis of Extraction Methods for Ground Object Information in Mining Area Based on Remote Sensing Image |
| |
| DOI：10.16160/j.cnki.tsxyxb.2020.03.009 |
| 中文关键词: Landsat-8遥感影像 监督分类方法 非监督分类方法 面向对象分类方法 地物信息 |
| 英文关键词: Landsat-8 remote sensing image supervised classification method unsupervised classification method object-oriented classification method ground object information |
| 基金项目: |
|
| 摘要点击次数: 9316 |
| 全文下载次数: 6334 |
| 中文摘要: |
| 以包头市白云鄂博矿区为试验区，以Landsat-8卫星影像为数据源，分别采用非监督分类、监督分类两种传统的分类方法与面向对象分类方法对遥感影像进行分类提取，并对三种方法的分类结果进行定性和定量比较。结果表明：传统的分类方法得到的分类图像连续性较差，很多点状分布的地物与实际地物不符合；而面向对象分类方法得到的图像各类地物之间边界明显，与实际的地物分布相似度高，更适合于矿区的空间信息提取工作。 |
| 英文摘要: |
| With Baiyun Ebo mining area in Baotou as the experimental area and Landsat-8 satellite image as the data source, two traditional classification methods of unsupervised and supervised ones and the object-oriented classification method are used to classify and extract the remote sensing images. Then, the classification results by the three methods are compared qualitatively and quantitatively. The results show that the classification image obtained by the two traditional methods has poor continuity, and many point-distributed ground objects do not match the actual ones. However, in the images obtained by the object-oriented classification method the boundary between various types of objects is obvious, and the degree of similarity to the actual ones is high. Therefore, this method is more suitable for the abstraction of the space information in the mining area. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |