chapter 8, Lossy Plus Lossless Residual Encoding

In III Lossless Compression Techniques from: Digital Image Compression Techniques

Author(s): Majid Rabbani, Paul W. Jones

Published: 01 February 1991

Chapter DOI: 10.1117/3.34917.ch8

Chapter Page Count: 2 pages

Previous Chapter | Next Chapter

Front Matter

I Background
1. Digital Images and Image Compression

References to Part 1

II Information Theory Concepts
2. Source Models and Entropy

3. Variable-Length Codes

4. Entropy Estimation and Lossless Compression

5. Rate-Distortion Theory and Lossy Compression

References to Part 2

III Lossless Compression Techniques
6. Bit Plane Encoding

7. Lossless Predictive Coding

8. Lossy Plus Lossless Residual Encoding

References to Part 3

IV Lossy Compression Techniques
9. Lossy Predictive Coding

10. Transform Coding

11. Block Truncation Coding

12. Vector Quantization

13. Subband Coding

14. Hierarchical Coding

15. Choosing a Lossy Compression Technique

References to Part 4

A. Compression of Color Images

Back Matter

Preview

Excerpt

In most image transmission applications, the requirement of a short transmission time (and correspondingly low bit rate) often precludes the use of a numerically lossless compression algorithm at the outset. However, in certain applications, it may be necessary to send a lossless version of the image at some later stage of the transmission. Such an application might exist in the medical field, where two physicians are discussing a possible patient referral from remote locations. One of the physicians may wish to transmit a digital radiograph over the phone line, and in the interest of a short transmission time, a lossy (but high quality) version of the image is sent. If the referral is accepted, the remaining difference (residual) image required to perfectly reconstruct the original image could be sent, perhaps at the conclusion of the telephone conversation. In general, a lossy plus lossless residual encoding scheme consists of the following steps:

• Generate a low bit rate image through the use of an efficient lossy compression scheme.

• Form a residual by computing the difference between the lossy reconstruction and the original image.

• Encode the residual using an appropriate lossless technique.

A lossy plus residual scheme can be considered a type of predictive coding where the lossy image constitutes a nonlinear, noncausal prediction. The prediction in this case is not based on previously transmitted information as it is in DPCM and must be explicitly transmitted as “overhead.”

DOI: 10.1117/3.34917.ch8

Your library does not subscribe to the eBooks portion of the SPIE Digital Library.