Method and apparatus for forming high resolution halftone images

Abstract

Halftone images are formed from dots selected from a dot character font memory storing information representing halftone dots of different sizes and shapes corresponding to different gray scale tone values or image densities. An original image is first scanned to sample its tone values at minute intervals exceeding the desired dot resolution of the reproduction. These sample values are digitized by an A/D converter which converts them to multibit binary numbers. Adjacent samples are compared to determine the average image density in a particular area, as well as the rapidity of rate of change of image density within that area. If the rate of change of image density in a given area is below a selected threshold value, the area is designated as low detail and the multibit binary number representing the result of the average density calculation over the area is used to select a dot character from the font memory for reproducing that area of the image. If, however, the rate of change of image density exceeds the selected threshold value, indicating an area of relatively high detail, the area samples are used to select special dot shapes and positions for use in reproducing that sampled area. This technique achieves a higher resolution halftone image for a given amount of stored data than prior techniques, as well as achieving a higher 'apparent' resolution to the observer.