| 摘要 |
A digital video processing system for correcting video color and other attributes, such as pixel location and sharpness. The method described may be employed by a general purpose computer or a secondary digital color processor to process images from standard or high-definition telecines, or from non-real time film scanners. Only the pixels and/or regions of a video picture to be modified are processed, so that most pixels, which are unmodified, remain free from any potential corruption. Pixels to be modified are identified by means of lookup tables which can be loaded for each frame with data indicating attributes such as hue, saturation, and luminance values, or ranges which are to be present in a pixel if that pixel is to be selected for modification. The pixel's attributes are applied to each lookup table as an address. The system employs multiple channels, with precedence logic, to control the priority order of the channels, in order to apply only one particular set of modification increments associated with only one channel, in the event that more than one set of selection attributes are applicable to a given pixel. The system is capable of standard- or reduced-definition modification of portions within a high-definition picture. Minute modifications may be automatically disregarded, thereby preventing any small mathematical errors from inadvertently affecting the image. The system also has wash mode capability for permitting blanket modification of one or more attributes of the entire video picture.
|
