Abstract: A method and apparatus for processing a video image so as to feature the boundaries between light and dark regions of the image. The method and apparatus are applicable to a video image which is represented by an array of pixels, in which each pixel is associated with a gray scale intensity value. According to the invention a center pixel is selected along with a plurality of pixels forming a two-dimensional neighborhood of the center pixel. From the intensity values associated with the pixels of the neighborhood, a determination is made whether or not the center pixel lies in a transition region, in which the image undergoes a rapid variation between light and dark. A bit value is then assigned to the center pixel, indicating whether the center pixel is to be black or white in the processed image. The bit value is assigned according to one of two predetermined algorithms depending on whether the center pixel was determined to lie within or not to lie within a transition region.