Advances in Earth Science ›› 2001, Vol. 16 ›› Issue (4): 544-548. doi: 10.11867/j.issn.1001-8166.2001.04.0544

• Articles • Previous Articles     Next Articles


SU Lihong 1,LI Xiaowen 1,3,HUANG Yuxia 2   

  1. 1. Research Center of Remote Sensing and GIS, Beijing Normal University,Beijing 100875,China;
    2. State Key Laboratory of Resource and Environment Information Systems, Institute of Geographic Sciences and Natural Resources Research,Beijing 100101,China;
    3. Dept Geography and Center for Remote Sensing, Boston University, Boston, MA02215, USA
  • Received:2001-01-02 Revised:2001-02-19 Online:2001-08-01 Published:2001-08-01

SU Lihong,LI Xiaowen,HUANG Yuxia. AN REVIEW ON SCALE IN REMOTE SENSING[J]. Advances in Earth Science, 2001, 16(4): 544-548.

The concept of scale is applied very broadly, scale is undoubtedly one of the most fundamental aspects of any research. There are four common connotations of scale in geography: cartographic scale or map scale, geographic scale or extent, resolution, and operational scale. Science of scale is considered to seek answers to the following questions. (1)The role of scale in the detection of patterns and processes, and its impact on the modeling; (2) The identification of domains of scale (invariance of scale) and scale thresholds; and (3) scaling, and the implementation of multiscale approaches for analysis and modeling. In context of remote sensing, applicability of some basic physics laws, such as, Helmholtz principle, Beer law, and Planck law, must be examined carefully, due to heterogeneity of pixels. And based on the same reasons, we should not use remote sensing physics models and applied models optionally on any resolutions, while these models are created on one special resolution. Some approaches that are used to examine scale invariance of model have been discussed in the paper. Concepts of spatial autocorrelation and spatial statistics have been introduced as a basis for understanding the effects of scale, wavelets and fractals provide a more comprehensive framework for prediction of scale effects. The quadtree data structures provide a consistent, logical way to multiscale modeling that is well grounded in theory and compatible with many ideas of physical process.

No related articles found!
Full text