Consequences of pixel precision for compression rate and measurement
| Autor: | Simen Gaure, Ingvil Hovig |
|---|---|
| Rok vydání: | 1997 |
| Předmět: |
Pixel
Uniform quantization business.industry ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Data compression ratio Data_CODINGANDINFORMATIONTHEORY Homogeneous Computer Science::Computer Vision and Pattern Recognition Signal Processing Compressibility Entropy (information theory) Computer vision Computer Vision and Pattern Recognition Artificial intelligence Electrical and Electronic Engineering business Algorithm Software Mathematics |
| Zdroj: | Signal Processing: Image Communication. 9:137-142 |
| ISSN: | 0923-5965 |
| DOI: | 10.1016/s0923-5965(96)00015-x |
| Popis: | An increase in image pixel precision makes performance drop in many compression methods. Many entropy reducing transforms are homogeneous and this homogeneity is not influenced if this transform is followed by a uniform quantization step. The entropy in the resulting data after these steps will therefore be additive as a function of the pixel precision. The differences in compressibility dependent on increasing pixel precision or entropy values of original image data would be expressed more explicitly by using proposed new methods of measuring compression gain. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect