Abstract: A method for removing noise in the form of small black spots in a black and white raster picture, which includes the following steps: Generating a moving window of 5.times.5 picture elements having an inner window (R1) of nine and a surrounding outer window (R2) of sixteen picture elements; detecting for each such window whether there are more black picture elements in the inner window than a given threshold, or whether there is any connection between black picture elements in the inner window and the outer window; if this is true, leaving the inner window unchanged; otherwise, i.e. if there are less black picture elements in the inner window than the threshold, and if they are not connected to black picture elements in the outer window, making the inner window all white. Thus, isolated small black spots are eliminated without eroding edges or erasing very thin lines.

Abstract: For recognizing a three-dimensional object from its two-dimensional image which was produced e.g. by a TV camera, a Hough transform representation is generated of the image and specific configurations or structures of the cluster points which constitute the Hough transform representation are determined. The information about these specific configurations is compared to similar information stored for the Hough representation of known object models. By thus relating portions of the image to portions of one or several object models, vertices of the image which are present at line or edge intersections, are related to vertices of the known object model(s). This knowledge about the correspondence of model and object vertex points allows the exact fitting of vertices and thus recognition of the unknown object and its relative orientation.

Abstract: A method and apparatus for automatically recognizing the presence of an image area and a text or graphics area on a printed master which is being examined for determination of the textual content thereof employ an optoelectronic video scanner, the scanned values obtained therefrom being digitized and stored in a memory. The input master is schematically divided into small overlapping windows and a number of length of run histograms are generated by a computer for each such window. The particular characteristics of each of the histograms are examined for determination as to whether the histograms correspond to a text or graphics area or an image area of printed material on the master.