SELECTED DATA COMPRESSION FOR DIGITAL PICTORIAL INFORMATION

摘要

This invention provides a method and apparatus to selectively compress digital image data for the supply of that information to digital display screens or storage. The invention utilizes a digital input containing a non-linear representation between the value and the intended intensity and linearizes this into a higher order number. If the higher order number is below at least a first threshold, it is transmitted or stored on a lower order data system noting that the higher order bits or channels should all be zero. If the higher order value is above the threshold, the highest order bits of that value are transmitted or stored on the lower order data system so that only the lowest order bits are lost. The received input from the lower order data system can then be decoded and the data sent to drive a digital display or otherwise utilized. As the linearization process introduces substantial errors in the lower order numbers and minimal errors in the higher order numbers, this invention seeks to increase the accuracy of those lower order numbers by offsetting this against introducing some additional error into the higher order numbers. Further embodiments utilize multiple stages of compression.