| 摘要 |
A raster image processing system and method accepts a digital page description as a series of page elements (text, graphics, images, etc.) and creates output video-ready (“flattened”) raster without allocating large amounts of raster memory (arrays of pixels) and without creating a display list of all page elements. This technique improves performance (greater speed with fewer resources consumed) in the typical output process: rendering, image processing, compression, and transmission. The method records each page element in memory in a manner that largely retains the inherent compression of the element description, and fully retains the positional relationships with its neighboring elements. Where an incoming page element overlaps an existing one, the intersections are calculated on the fly, and elements underneath are immediately split, merged, shrunken, or deleted. Each incoming page element is immediately processed and recorded as described above, and then discarded. The result is a continuously precise, simple, non-overlapping (“flat”) representation of the final output raster. In other words, in one embodiment, the representation starts empty, then stays compressed and flat throughout while accumulating representations of page elements.
|
