摘要：A global multi-resolution image data model and a feasible solution for its seamless management and archiving remain a challenging vision. The traditional methods of the raster pixel data structure based on the idea of map projections are effective to support local or small-scale areas. However, if this structure is applied to large-scale or whole global image archiving, some significant drawbacks are unavoidable, such as data discontinuity (or overlapping), geometric distortions, etc. To overcome these deficiencies, in this paper the Quaternary Triangular Mesh (QTM) (Dutton, 1989), as a continuous, hierarchal quadtree data structure with uniform grids on a sphere, is proposed for global-scale seamless image archiving. First, the mapping relation between raster image pixels and QTM pixels is approached based on the QTM subdivision and Quaternary coding scheme (Bartholdi & Goldsman, 2001), and a corresponding algorithm of QTM pixel grey level calculation is also developed. Then, the storage structure of global-scale image archiving based on QTM pixels is presented in detail. In the end, an experiment is described using the 1km resolution NOAA data for China, comparing the differences in pixel grey levels between original image pixels and QTM pixels. The result indicates that the QTM pixel data structure can keep global-scale images seamless, and the accuracy of transformation from the imaging pixel to the QTM pixel is a loss of less than 2 grey levels for 94.5% of all pixels, the loss from 2 to 4 is 1.9%, the loss from 4 to 10 is 2%, and the rest is 1.6%. The results are good and acceptable.
