| 摘要 |
A method and apparatus for compressing YUV color image data representing an image. The method includes the steps of generating base image data containing low frequency image information where the base image data is a function of the Y component, the U component, and the V component, and generating detail image data containing high frequency image information where the detail image data is a function of the Y component alone. The base image data can be decompressed to from a decompressed base image data that is an approximate facsimile of the original image and the detail image data can be decompressed and selectively added to the decompressed base image data to form a facsimile which more accurately duplicates the original image than the decompressed base image data alone. The step of generating the detail image data includes the steps of determining an average of Y component signals of a sub-array of pixels, determining an approximate variance of Y component signals of the sub-array, and forming a bit mask wherein each bit in the mask indicates whether the corresponding pixel is greater than or less than the mean. The compression algorithm is optimized for execution on general purpose commodity microprocessors so that it can be executed with reasonable performance on conventional desk-top personal computers without specialized hardware.
|
