Image processing method and apparatus with reduction of granular noise,

Abstract

According to the present invention, when image data are subjected to quantization processing using an error diffusion method, the generation of a striped pattern which is a cause for deterioration of picture quality in the error diffusion method is prevented. In the present invention, an original image is read and image data are generated. The image data obtained by reading the original image are quantized using an error diffusion method, and the quantized image data can be output and recorded by a thermal-head printer. Furthermore, according to the present invention, an error between the density of an input image and the density of an output image after being subjected to a quantization processing using an error diffusion method is perfectly preserved. In an image processing apparatus for performing a quantization by dispersing an error between input image data and output image data which arises when the input image data are quantized, to image data of surrounding picture elements, the error between the input image data and the output image data is computed or otherwise determined, the error is subjected to a predetermined weighting processing, the error subjected to the weighting processing is dispersed to surrounding picture elements, and a surplus of the error generated in the weighting processing is corrected.