With its superior capability to deliver abundant and instantaneous ground truth information, remote sensing image is highly regarded as the most popular data to support geospatial decision making. Recent internet technology innovation has dramatically reduced the difficulty of searching and accessing geographic data in distributed environment. For RS image users, the challenge is no longer "where to find images", but "how to select the best image from the huge internet-based image archive environment." This issue requires all of the candidate RS images to be "ranked" in a way to meet users' application preference. Spatial constraint, i.e., the relationship between the AOI (Area of Interest) and RS images, is no doubt the most dominant factor during this selection process. From the spatial perspective, we proposed 3 indicators in this paper to serve as the basis for determining the spatial ranking of individual RS image with given AOI. Every indicator represents a mathematical formalization of users' preferred scenario while selecting best images. These indicators are tested and analyzed extensively based on different dimensionality of AOI (point, curve, surface) to reflect its behaviors in different spatial layouts of AOI and images.
展开▼
