Methods of color document compression, processing, and transmission

Guotong Feng, Purdue University

Abstract

A mixed content document typically contains a mixture of text, graphics, halftone regions, and pictures. Since the characteristics and spatial behaviors in these regions are dramatically distinct, efficient and accurate representation of a mixed content document becomes very challenging in many document imaging applications. This thesis includes three chapters focusing on three different application areas. ^ In Chapter 1, our objective is to develop an effective and high quality color document coding method based on the mixed raster content (MRC) model. While most MRC based methods can yield very high compression ratios, the binary representation of MRC tends to distort fine document details. To address this problem, we propose a method called resolution enhanced rendering (RER), which works by adaptively dithering the encoded binary mask, and then applying a nonlinear predictor to decode a gray level mask at the same or higher resolution. This method substantially improves the decoded document quality of text and graphics at a fixed bit rate. ^ In Chapter 2, we propose an efficient and flexible solution for binary representation of mixed content documents using CCITT G3/G4 compression. The solution includes two variations which we refer to as FastFax and ReadableFax. Both methods provide accurate binary representation of image content with high compressibility. Based on FastFax, we propose a layer-based compression scheme, MixedPDF, for color document representation in the PDF format. This method includes an efficient segmentation algorithm to separate the document into a text layer and a background layer, which are then separately coded by different compressions. The MixedPDF algorithm substantially outperforms conventional JPEG compression with low computational and memory requirements. ^ In Chapter 3, we propose an efficient error diffusion algorithm optimized for PackBits compression. This method, which we refer to as POED (PackBits optimized error diffusion), is a form of threshold modulation error diffusion which takes advantage of the byte-oriented run length structure of PackBits compression by encouraging repetition of bytes in the resulting binary image. The POED method with PackBits compression yields higher compression ratios than the conventional error diffusion method, while maintaining desirable visual quality with low computational and memory requirements.^

Degree

Ph.D.

Advisors

Charles A. Bouman, Purdue University.

Subject Area

Engineering, Electronics and Electrical

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS