摘要
通过对遥感影像中河流光谱特征的研究,提出了一种基于区域生长的遥感影像河流提取方法。该方法首先根据遥感影像生成灰度图像,选取河流生长初始区域,设置生长范围和生长阈值,通过区域生长得到河流边界,剔除河流内部"孤岛"并对边界细化,最后生成矢量文件。本文使用具有不规则河流边界的Landsat8遥感影像作为实验对象,验证方法的可行性和有效性。
After study on the spectral characteristics of remote sensing image, this paper presents a river extraction method.The first step of the method is change remote sensing image into gray image, select the initial growth area of the river, and set growing range and threshold, we can get the river border through gegion growing.The next step is to remove the noise in the river and thinner image, convert images to vector data at last.This paper takes a Landsat8 remote sensing image with the rivers has irregular borders as test ob-ject to verify the feasibility and effectiveness of this method.
出处
《测绘与空间地理信息》
2015年第3期198-200,共3页
Geomatics & Spatial Information Technology
关键词
遥感影像
区域生长
河流
图像细化
矢量化
remote sensing image
region growing
river
image thinning
vectorization
