US 6608928

Patent Overview — US6608928B1

Title: Generic pre-processing of mixed raster content planes

Patent number / application: US6608928B1, application US09/432,600

Assignee: Xerox Corporation (original assignee; later security-interest and release assignments to Bank One, NA and JPMorgan Chase Bank appear in the record)

Dates: Filed 1999-11-03; priority date 1999-11-03; granted/published 2003-08-19; anticipated expiration 2019-11-03. Status listed as "Expired – Lifetime."

Inventor: Ricardo de Queiroz

Purpose: The patent addresses compression of scanned images, which are often large and therefore good candidates for compression. Mixed Raster Content (MRC) separates a composite document image into multiple planes—described as an upper plane, a lower plane, and a binary selector plane that chooses between them—and applies an appropriate compression technique to each plane. The problem: segmentation leaves data in the upper and lower planes in areas the selector plane will never choose, so those pixels are unused during reconstruction, adding useless bits and lowering the compression ratio. The invention uses the selector plane as a reference to reduce the data needed to encode the associated planes.

Plain-language scope: The systems and methods pre-process the upper and lower planes using selector-plane information, determining which data is useless, smoothing the boundary between actual image data and useless data, and replacing the useless data with values that improve compression. The compression/decompression approach is described as applicable to any set of compression techniques for separately processing different data types, and to systems such as digital printers and copiers. Embodiments include an image segmenting portion with upper and lower plane pre-processors, a compressor, decompressor, and output controller, illustrated in FIGS. 1–9.

Source: https://patents.google.com/patent/[US6608928](/patent/US6608928)/en

What the excerpt does not establish: The supplied text is explicitly an excerpt with omitted portions. It does not include the full claims, detailed description, or references, so the precise claim scope, dependencies, and any prior-art citations are not established here. It also does not establish litigation, ownership outcomes, validity, or the accuracy of the listed legal status or assignee data.