| 摘要 |
A method and apparatus are described for encoding a three-dimensional array of data representing a physical entity, such as an image (or sequence of frames), by means of its local symmetries. This encoding yields both compression and a resolution-independent description which allows reconstruction of the image to an arbitrary scale. Spatial zooming and interframe interpolation can be achieved without significant loss of information. The method of the invention includes the steps of storing digital data of a predetermined size in a three-dimensional format (20), dividing the array into a plurality of rectangular domain boxes to generate a plurality of uniquely addressable three-dimensional domain sets (22), dividing the array into a plurality of range boxes that are uniquely addressable (24), applying an affine transformation to each domain set and, for each domain set, selecting a range set and transformation (26), and outputting the address of a selected range set and selected transformation as a codeword (28). This set of codewords, one codeword for each of the selected domain sets, represents the original digital data. The transformations selected for each of the domain sets corresponds to one of the following criteria: (a) the minimum distance between the domain set and a transformed range set; (b) the distance between a domain set and a transformed range set that falls below a given threshold; or (c) the range set satisfies a secondary criterion and the distance between the domain set and a transformed range set falls below a given threshold.
|
