Abstract: Apparatus and accompanying method(s) for digitally scaling, particularly reducing, an original image to yield a scaled image particularly suited for display by a document workstation in an image management system as one image in a composite multi-image display. Reduction scaling is achieved by logically combining pixels values in each block of pixels in an original bi-tonal image, preferably through an "OR" operation, in order to generate a corresponding single pixel value in a scaled bi-tonal image. To reduce the original image by a non-integer scale factor, the pixel blocks that form the original image are assigned differing groupsizes that periodically vary on a two-dimensional pattern from block to block throughout the original image. The average value of the periodic variation occurring in each direction throughout a sufficiently large two-dimensional area in the original image substantially equals the fractional portion of the non-integer scale factor.

Abstract: Apparatus, and accompanying methods, for determining appropriate sampling points located throughout a scanned microfilmed bit-mapped "data strip" in order to properly extract the values of stored data bits therefrom. Each data bit is formed of a contiguous group of one or more pixels in the data strip. Specifically, my technique involves parsing the data strip into successive vertical strips that each has successive rows of pixels, determining inter-row and inter-pixel differential intensity values for each vertical strip, and defining vertical line addresses and horizontal pixel addresses in each of the strips as a function of the pixel positions of the inter-row and inter-pixel differential intensity values associated therewith. The individual pixels in each vertical strip, which are situated at the horizontal pixel addresses and which lie on those rows in that vertical strip specified by the vertical line addresses, are sampled in order to yield sampled pixel data.

Abstract: During segmentation of individual characters in a column, the invention predicts for the next character an optimal minimum gray level sensitivity threshold while scanning pixel rows between characters by lowering the threshold unless or until noise in the image causes the scanner to detect too many "ON" pixels between adjacent numerals. During subsequent attempts to recognize the segmented character, the invention computes a confidence score and increases the sensitivity threshold above the predicted level if the confidence score is insufficient.

Abstract: A feature-based character recognition identification and confidence level are determined for an unknown symbol. If the confidence level is within an intermediate range, the feature-based identification is confirmed by matching the unknown character with a reference template corresponding to the feature-based identification. If the confidence level is below the intermediate range, template matching character recognition is substituted in place of the feature-based identification. If the template matching recognition identifies more than one symbol, corresponding templates from a second set of templates having thicker character strokes are employed to resolve the ambiguity.

Abstract: A halftoning method for creating a color binary image from a continuous tone color image or computer generated graphics signal is improved by diffusing color vector error and by including the visual color blur functions in the recursive error propagation algorithm so that perceived color vector rather than binary color error is propagated.

Abstract: A recursive spatial filter interchanges the value of a control pixel with that of the value of a predetermined one of the surrounding pixels in a pixel mask when a specific image pattern occurs in the digital image. The use of such a filter eliminates certain highly visible grain patterns from the halftoned digital image while preserving the average gray level.

Abstract: A film image retrieval and scanning apparatus having split first and second optical paths with each optical path being precisely aligned with the other, and each path containing coupled zoom optics. The first path projects a film image onto a first scanner array in the first path to provide a real time display of the image on a TV monitor. The second optical path projects an image onto a second scanner array, which may have a higher resolution than the first array to provide an analog signal that is converted to a digital signal of the image that may be stored, or transmitted on a network, or both. The first scanner provides a real time image and serves as a viewfinder that can be used to locate the proper position of an image; thus ensuring that the image will be aligned with the second array so that it can be scanned by the second scanner array when desired.

Abstract: A method and apparatus for producing complex magnetization patterns in a sheet of magnetic material. A fixture having top and bottom plates with a matching conductor pattern embedded in one surface of each plate. The conductor pattern covers substantially the entire surface of each plate. To magnetize the sheet of magnetic material, it is placed between the plates and a magnetizing current is applied to the conductors with a capacitor discharge magnetizer. The resulting sheet of magnetized material is useful as a magnetic deactivator for electromagnetic security articles.

Abstract: In a laser scanner of the type having beam shaping means and a stationary diffraction grating for directing the shaped beam to a hologon scanning disc, the beam shaping means is combined with the stationary diffraction grating by providing a shaped stationary diffraction grating, thereby eliminating the need for other beam shaping means prior to the shaped diffraction grating, thereby simplifying construction of the apparatus.

Abstract: An optical scanning system of the type in which an oscillating mirror located between the focusing lens and the image plane is provided to scan the reflected optical axis of a focusing lens across a flat field, wherein the pivot axis of the mirror is translated during oscillation of the mirror. The pivot axis of the mirror is translated at an angle of 45.degree. with respect to the optical axis of the focusing lens thereby enabling the mirror to be translated at a uniform velocity during scanning to provide a uniform velocity scan.

Abstract: Apparatus for converting multilevel video images into a bilevel image having an accurate rendition of the original grayscale of the image. The apparatus can provide an improved rendition of low contrast images.

Abstract: A feature-based optical character recognition system, employing a feature-based recognition device such as a neural network or an absolute distance measure device, extracts a set of features from segmented character images in a document, at least some of the extracted features being at least nearly impervious to rotation or skew of the document image, so as to enhance the reliability of the system. One rotationally invariant feature extracted by the system is the number of intercepts between boundary transitions in the image with at least a selected one of a plurality of radii centered at the centroid of the character in the image.

Abstract: Document contour vectorization and the use of a modified Hough transform are used in combination to detect the skew angle of a digitized form image so that image skew may then be corrected in the preprocessing of form images prior to document analysis and classification.

Abstract: In an optical character recognition system, character strokes are automatically enhanced prior to processing by matching the character bit map with a set of templates. Each template has a stripe of thickness W composed of binary ones surrounded by zeroes along a horizontal, vertical, major diagonal or minor diagonal direction. If the template matches a character stroke to a predetermined degree, the stroke is enhanced by increasing its thickness to a predetermined number N of binary ones.

Abstract: A halftoning method for creating a binary image from a continuous-tone image or computer generated graphics signal is improved by including the human visual blur function in a recursive error propagation algorithm so that perceived error rather than binary error is propagated.

Abstract: Apparatus, and an accompanying method, for a neural network, particularly one suited for use in optical character recognition (OCR) systems, which through controlling back propagation and adjustment of neural weight and bias values through an output confidence measure, smoothly, rapidly and accurately adapts its response to actual changing input data (characters). Specifically, the results of appropriate actual unknown input characters, which have been recognized with an output confidence measure that lies within a pre-defined range, are used to adaptively re-train the network during pattern recognition. By limiting the maximum value of the output confidence measure at which this re-training will occur, the network re-trains itself only when the input characters have changed by a sufficient margin from initial training data such that this re-training is likely to produce a subsequent noticeable increase in the recognition accuracy provided by the network.

Abstract: Vertically segmented characters are pre-processed for optical character recognition so as to remove skew and noise by convolution with a mask consisting of parallel vertical lines with a one-pixel spatial frequency. The horizontal alignment of the mask with respect to each segmented character is adjusted until a maximum cross-correlation is found. Then, the product of the mask and the character image is formed. A histogram of the number of pixels in the product image along each vertical line of the mask is compared with a library of such histograms of known symbols and the best match identifies the symbol represented by the character image. Alternatively, the histogram is converted to a binary codeword which is compared with a library of such codewords corresponding to a set of known symbols.

Abstract: The invention relates to a system that uses human readable microfilm to store information and retrieve that information for later use in a computer. The microfilm is put into a Computer Input Microfilm (CIM) Reader that scans the film and converts the characters or text in the image into computer codes. An Optical Character Recognition algorithm is used to identify the text character and the CIM Reader converts that character into a computer code.

Abstract: A polygon-based graphics/text separation method is comprised of two sequential processes. First a raster to contour vector conversion step is used to convert a digitized bitmap into a collection of simple polygons. Next a component classification process is used to extract six particularly defined features of each of the individual polygon-based components to enable the separation of graphics and text polygons. Graphical polygons are further classified into four subclasses. Textual polygons are grouped into polygon strings (text strings). Each string contains a sequence of segmented character contour polygons which is ready for an optical character recognition algorithm to convert them into computer understandable ASCII characters.