IMAGE PROCESSING METHOD AND APPARATUS, THRESHOLD VALUE MATRIX CREATING METHOD, IMAGE FORMING APPARATUS, SUB-MATRIX CREATING METHOD AND PROGRAM

The image processing method comprises: a recording failure position determination step of determining a recording failure position on an image corresponding to a malfunctioning recording element; a basic threshold value matrix storage step of storing a basic threshold value matrix set with threshold values used for halftoning of converting multiple-value input image data into dot data of a number of tonal graduations smaller than that of the multiple-value input image data by quantizing the multiple-value input image data; a sub-matrix storage step of storing a plurality of sub-matrices in association with recording failure positions in the basic threshold value matrix, each of the sub-matrices being set with threshold values which are substituted for the threshold values in a partial region of the basic threshold value matrix, the partial region including a pixel position corresponding to the recording failure position and having a width of a prescribed number of pixels; a replacement region determination step of determining a corresponding region with the width of the prescribed number of pixels including the recording failure position in the basic threshold value matrix, according to the pixel position in the input image data and the recording failure position determined in the recording failure position determination step; a sub-matrix selection step of selecting one of the sub-matrices stored in the sub-matrix storage step to use for substituting for the corresponding region determined in the replacement region determination step; a threshold value replacement step of creating a reformed threshold value matrix by replacing the threshold values of the corresponding region including the recording failure position in the basic threshold value matrix, with the one of the sub-matrices selected in the sub-matrix selection step; and a quantization processing step of quantizing the input image data by selectively using the basic threshold value matrix and the reformed threshold value matrix.