AUTOMATIC DETERMINATION OF BLANK PAGES AND BINARY IMAGES' BOUNDING BOXES

A method, and apparatus for executing the method, for processing digitized image data representative of a page. The method includes the steps of (a) processing the image data on a scanline by scanline basis to partition the image data into a plurality of rows and columns of N.times.M arrays (14a) of image pixels. The step of partitioning includes the steps of, for each row of an N.times.M array of image pixels, (b) detecting occurrences, if any, along any scanline of an image pixel having a foreground value; for a non-zero number of detected occurrences within one N.times.M array of image pixels, (c) incrementing a corresponding array row count and a corresponding array column count; and (d) for at least one detected occurrence, adjusting the boundary of a bounding box so as to enclose the at least one detected occurrence and substantially all previously detected occurrences. The method includes the terminal steps of examining the boundary of the bounding box and, if the boundary is set at an initial unrealizable condition, declaring the page to be blank, else if the boundary is not set at the initial condition, declaring that the binary image of the page includes information within the boundary of the bounding box. Methods for deriving an iconic image of the digitized image data are also described.