研究论文

ETM+影像提取伏沙地信息的最佳波段组合——以内蒙古锡林郭勒盟西乌旗为例

  • 张韬;赵宇飞;安慧君;陈秀兰
展开
  • 内蒙古农业大学林学院,呼和浩特 010019

收稿日期: 2011-05-06

  修回日期: 2011-05-19

  网络出版日期: 2011-06-18

Selection of ETM+ Remote Sensing Image Optimum Waveband Combination in Information Extraction of Sinking Sandy Land——The Case in Xiwu Flag, Xilin Gol League, Inner Mongolia

  • ZHANG Tao;ZHAO Yufei;AN Huijun;CHEN Xiulan
Expand
  • College of Forestry, Inner Mongolia Agricultural University, Huhhot 010019, China

Received date: 2011-05-06

  Revised date: 2011-05-19

  Online published: 2011-06-18

摘要

以伏沙地信息提取为研究目的,依次应用单波段信息量、相关系数矩阵、最佳指数法(OIF)、典型地物影像光谱特征曲线分析的方法,对内蒙古锡林郭勒盟西乌旗ETM+遥感影像数据的最佳波段选择进行研究。结果表明,第4波段信息量大,与其他波段间相关性小,作为必选波段;5、3波段次之;3、4、5波段组合的OIF较高。最后根据目视辨别认定波段(红)4、(绿)5、(蓝)3组合是针对伏沙地信息的最佳波段组合。

本文引用格式

张韬;赵宇飞;安慧君;陈秀兰 . ETM+影像提取伏沙地信息的最佳波段组合——以内蒙古锡林郭勒盟西乌旗为例[J]. 科技导报, 2011 , 29(17) : 29 -33 . DOI: 10.3981/j.issn.1000-7857.2011.17.002

Abstract

This paper discusses the optimum waveband combination in information extraction of sinking sandy land. The single band dataset, the correlation coefficient, the Optimum Index Factor (OIF) and the typical object spectral image feature curve are used to analyze the ETM+ multi-spectral band combination for sinking sandy information extraction in Xiwu Flag, Xilin Gol League, Inner Mongolia. The results show that: (1) Band 4 contains the greatest amount of information with the smallest correlation with other bands, band 5 and band 3 follow in their importance. The higher OIF can be obtained with the band 3, band 4, and band 5 combination, which is the best combination for sinking sandy land information extraction. (2) To select the best band combination according to remote sensing research target, different research objects have different best band combinations. To calculate the optimum band combination for relevant index for sinking sandy, one can use ERDAS and EXCEL tools to establish the function model, so as to improve the calculation efficiency. (3) In a typical ground, the feature image characteristic curve, sinking sandy land and typical steppe spectrum may be similar, as their difference is in the underground soil horizon part, while the ground vegetation in the remote sensing image through the spectra is difficult to be distinguished. The texture characteristics through images, the position relations etc are indirectly conditional to discriminate accurately. (4) Through the visual translation process for sinking sandy land with sand and grassland, wetlands, forest multiple properties, to accurately draw the space position of the sinking sandy land, one should considersinking sandy landcauses factors. In addition, because the sinking sandy land soil composition is not complete, humus layer is relatively thin, the structure is loose, the quality of the material is coarse, susceptible to damage, the foundation for sand, once damages, will lead to desertification, and the sand recovery is very difficult. So the relevant Government departments or scholars should try to attach great importance to protection and systematic research.
文章导航

/