X img





作者投稿 专家审稿 编辑办公 编委办公 主编办公

冰川冻土 ›› 2002, Vol. 24 ›› Issue (4): 452-456.

• 研究论文 • 上一篇    下一篇


马明国1, 角媛梅1,3, 王雪梅2, 程国栋1   

  1. 1. 中国科学院寒区旱区环境与工程研究所冻土工程国家重点实验室, 甘肃兰州, 730000;
    2. 中国科学院寒区旱区环境与工程研究所冻土工程国家重点实验室, 甘肃兰州, 730000;
    3. 云南师范大学旅游与地理科学学院, 云南昆明, 650092
  • 收稿日期:2001-12-17 修回日期:2002-04-10 出版日期:2002-08-25 发布日期:2012-04-26
  • 基金资助:

The TM/ETM+Mosaic Image Processing and Application in the Heihe River Basin

MA Ming-guo1, JIAO Yuan-mei1,3, WANG Xue-mei2, CHENG Guo-dong1   

  1. 1. State Key Laboratory of Frozen Soil Engineering, CRAEERI, CAS, Lanzhou Gansu 730000, China;
    2. Cold and Arid Regions Environmental and Engineering Research Institute, CAS, Lanzhou Gansu 730000, China;
    3. Yunnan Normal University, Kunming Yunnan 650092, China
  • Received:2001-12-17 Revised:2002-04-10 Online:2002-08-25 Published:2012-04-26

摘要: 对研究区内两个时期(1986—1990年和1999—2000年)20景LandsatTM5或Landsat ETM7影像图进行多波段合成、色彩优化、几何配准和数字镶嵌等预处理工作,生成试验区两个时期的TM/ETM+影像数字镶嵌图,并利用均方根(Root means quare,RMS)误差法评价几何配准的精度,用采样法评价数字镶嵌的精度.结果表明,单景TM/ETM+影像图几何配准的精度可以控制在1个像元以内.多景数字镶嵌时多数点的RMS误差小于2个像元,总的RMS误差在13~15之间.利用影像图作为几何配准的数据源的方法要优于利用地形图作为几何配准的数据源的方法。

关键词: TM, 数字镶嵌, 黑河流域, 几何配准

Abstract: The study area of this paper is the Heihe River Basin, a rather big inner river basin in the arid regions of Northwest China. In the middle of this basin, there are a lot of oases, which are important commercial grain production bases. Because the oasis agriculture consumes much more water and some destroying events occur, such as disorderly reclamation, denudation, and excessive farming, the desertification processing of the basin speeds up. Especially in Ejin Banner in the lower part of the basin, under the effect of fast decreasing runoff, ecological environment seriously deteriorates, which has been widely regarded. So an efficient tool is demanded to dynamically monitor the ecological changes, in order to cumulate dependable information for ecological planning and resume. Remote sensing has the advantages of repeatable acquirement. As one of the satellite data, Landsat TM/ETM+ image data has been widely applied in a lot of ecological monitoring fields, such as land use, landscape change, water resources, city and town land change, and debris flow monitoring. Two periods of Landsat TM/ETM+image data have been used in this paper. One is from 1986 to1990 and the other is from 1999 to 2000. The fourth, third and second or fifth, fourth, third bands consist of red, green and blue bands. Then the color balance is optimized. After geometric correction and digital mosaic, TM/ETM+ mosaic image maps of the two periods are finished. The geometric correction accurateness is calculated by using root mean square method and the digital mosaic accurate is estimated by using sampling point method. The geometric correction of single TM/ETM+image could be controlled within one pixel. When multi-scope images are digitally mosaiced, RMS error of most of the sampling points is less than two pixels. The total digital mosaic RMS error is between 1.3 to 1.5 pixels. The accurateness of using images as the geometric correction data resource is higher than that of using relief maps. From the TM/ETM+ mosaic image, one can have an intuitionist understanding of the ecological characteristics and landscape pattern of the study area.

Key words: TM, digital mosaic, Heihe River Basin, geometric correction


  • P283.8