地球信息科学学报

地球信息科学学报 ›› 2017, Vol. 19 ›› Issue (4): 559-569.doi: 10.3724/SP.J.1047.2017.00559

• 遥感科学与应用技术 • 上一篇    下一篇

江西省都昌县山地与平原植被界线提取

齐文娟1,2(), 杨晓梅1,*()   

  1. 1. 中国科学院地理资源与科学研究所 资源与环境信息系统国家重点实验室,北京 100101
    2. 中国科学院大学,北京 100049
  • 收稿日期:2016-08-12 修回日期:2016-12-08 出版日期:2017-04-20 发布日期:2017-04-20
  • 通讯作者: 杨晓梅 E-mail:qiwj@lreis.ac.cn;yangxm@lreis.ac.cn
  • 作者简介:

    作者简介：齐文娟（1983-）,女,河北承德人,博士生,研究方向为高分辨率遥感影像解译与分析。E-mail:qiwj@lreis.ac.cn

  • 基金资助:
    国家重点研发计划项目“基于国产遥感卫星的典型要素提取技术”（2016YFB0501404）;江西省科技计划项目（2015ACF60025）

Mountain and Plain Vegetation Boundaries Extraction in Duchang County Province Jiangxi

QI Wenjuan1,2(), YANG Xiaomei1,*()   

  1. 1. State Key Laboratory of Resources and Environmental Information system, Institute of Geographic Science and Natural Resource Researches, Chinese Academy of Sciences, Beijing 100101, China
    2. University of Chinese Academy of Sciences, Beijing 100049, China
  • Received:2016-08-12 Revised:2016-12-08 Online:2017-04-20 Published:2017-04-20
  • Contact: YANG Xiaomei E-mail:qiwj@lreis.ac.cn;yangxm@lreis.ac.cn

RichHTML

PDF (PC)

赞

被引次数

1 | 1

摘要：

由于生长环境、地带性分布规律和垂直差异,山地与平原地区的不同植被类型在遥感影像上存在“同物异谱”及“异物同谱”现象,易导致土地类型的误分。为避免此类错分,在进行土地覆被类型解译之前,应首先确定平原、山地植被的边界。本文在遥感影像聚类分析、GIS空间分析及数理统计分析技术的支持下,以江西省都昌县北部地区为研究区域,基于高分一号（GF-1）遥感卫星影像及地形要素,完成了山地植被与平原植被的界线提取。实验结果显示,本研究得到的植被界线精度高达99.47%和96.28%。与单纯基于遥感影像分类得到的植被边界成果相比,精度提高了近25%和23%。研究证明,高分辨率遥感影像与地形因子结合后,计算得到的山地平原划分界线精度有明显提高,可以满足高分辨率遥感影像土地覆被类型解译研究的需要。

关键词: GF_1高分遥感影像, 绝对高程, 地形起伏度, 地形部位指数, 地貌位置类型

Abstract:

Due to differences in zonal and vertical distribution of growth environment, there could be "foreign body with same spectrum" phenomenon of different vegetation types in mountain and plain areas, resulting in false land cover classification. To avoid such misclassification, we defined vegetation boundaries between mountain zone and plain zone before interpretation of land cover types. This research is carried on based on clustering analysis of remote sensing images, spatial analysis of GIS and technology support of mathematical statistics analysis. The study area is located in the northern region of Duchang County, Jiangxi province. Based on the remote sensing images of high resolution GF_1, 3 kinds of terrain factors, which are area absolute elevation, relief and terrain landform position, division of mountain vegetation and plain vegetation are completed. Experimental results show that terrain factors combining with GF_1 image will achieve division accuracy of 99.47% and 96.28%, respectively. Compared to the calculated results of pure terrain factors, the accuracy of boundary extraction increased by more than 40% and 5%, respectively. Compared to experiments based on pure remote sensing image classification, the accuracy increased by nearly 25% and 23%, respectively. Besides, if compared with the 1:250,000 geomorphologic map, the accuracy is increased by nearly 15% and 23%, respectively. This research proves that using GF_1 images and terrain factor together to calculate vegetation boundary of plain and mountain will obtain results with an accuracy which can meet the requirements of interpretation of high resolution of remote sensing images of land cover.

Key words: Gf_1 high resolution of remote sensing images, absolute elevation, relief, topographic position index, types of landform position