Method of and apparatus for compressing image representing signals
US5542008A · kind A · utility
Assignee
Inventors
Key dates
| Filing date | Jan 28, 1994 |
| Grant date | Jul 30, 1996 |
| Priority date | — |
| Expiry date | Jan 28, 2014 |
Classification
- Technology area (CPC H)Electricity
- CPC primaryH04N19/152
- WIPO fieldAudio-visual technology
- WIPO sectorElectrical engineering
Abstract
An image representing input signal is processed so the data therein are filtered in a predetermined manner to derive a filtered signal. The values of the filtered signal are combined at successive time intervals, and the values are combined so there is an addition of a function of values of the filtered signal at the successive time intervals to derive a combined signal. The combined signal is converted into a signal representing a predicted number of bits in a frame of the image. In response to the signal representing a predicted number of bits in a frame of the image, a signal representing the magnitude of a control for quantization step size is derived which is as a function of an error between the predicted number of bits and a predetermined target value therefor. A function of the input signal data is orthogonally transformed so a transformed signal is derived. Output data represented by the transformed signal are quantized so they have quantized step size determined by the control signal representing the magnitude of a quantization step size and there is a reduction in the number of data bits for the image relative to the number of data bits in the input signal.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.