Lossless image compression using a simple prediction method

✍ Scribed by Kenneth M. Dawson-Howe

• Publisher: John Wiley and Sons
• Year: 1996
• Tongue: English
• Weight: 402 KB
• Volume: 7
• Category: Article
• ISSN: 0899-9457
• DOI: 10.1002/(sici)1098-1098(199623)7:3<227::aid-ima10>3.0.co;2-w

✦ Synopsis

This article describes a new straightforward technique for lossless image compression, entitled simple prediction method, which results in compression ratios similar to those achieved by the most powerful techniques described in the literature. The predictive model used by the method is one in which the current point is predicted as a weighted average of the preceding neighbouring points. The weights for this mask are encoded within the compressed image. The difference between this prediction and the actual value is generally small with a symmetric exponential distribution, and this difference is represented using arithmetic encoding.