BINARY QUANTIFICATION OF AN IMAGE HAVING MULTIPLE LEVELS OF GREYS

摘要

Method for transforming a video type image having multiple levels of greys into a binary image, method wherein each pixel is either white or black. With each pixel of the initial image there is associated a partial image centered on said pixel and comprising a plurality of pixels in at least one dimension. If the histogram of the grey levels of said partial image is considered as bimodal (D > E), a threshold (S) median with respect to said histogram is selected. If the histogram is considered as monomodal (D < e), a threshold (S) outside said histogram is selected, the position of this threshold with respect to the histogram being determined by comparison of the latter with a preset mean background value (FM). Application to telecopy, electronic storage of documents and drawings, preprocessing in order to facilitate the reading of handwritten messages or recognition of patterns in general.