摘要 |
The present invention is a method and apparatus for the identification and tagging of antialiased pixels within regions of an image. The techniques described improve the recognition of fine gray features and enable the detection of both black and white features on high and low contrast backgrounds. The system relies on the establishment of the sense (positive/negative) of an image by first using a histogramming operation. Once the sense is determined, a plurality of logical channels are used to select a threshold, binarize a region of the image, generate a vector representative of the region, and to process the vector to identify whether an antialiased pixel is present therein. Subsequently, the output of the channels is logically combined to produce an antialias tag.
|