Abstract: A digital telephone (60) comprises send equipment (40) and receive equipment (50), coupled via a signal splitter/combiner (61) onto a digital telephone line (62) for communication with a remote party (90). The send equipment (40) comprises an A-law map (12) and a mu-law map (43). Receive equipment (50) comprises an inverse A-law map (14) and an inverse mu-law map (52). Maps (12,43) are selectable by the user (80). Similarly, inverse maps (14,52) are also selectable by the user (80). Each of send equipment (40) and receive equipment (50) may further comprise an un-quick-fix module (45) for removing unwanted insertions of quick fixes (25) inserted into the telephone line (62) by international telephone companies. Receive equipment (50) may further comprise a digital signal processor (57), computer (56), a set of prestored templates (58) corresponding to possible incoming telephone characteristics, and a caller parameter decipher module (51).

Abstract: Pattern recognition, for instance optical character recognition, is achieved by training a neural network, scanning an image, segmenting the image to detect a pattern, preprocessing the detected pattern, and applying the preprocessed detected pattern to the trained neural network. The preprocessing includes determining a centroid of the pattern and centrally positioning the centroid in a frame containing the pattern. The training of the neural network includes randomly displacing template patterns within frames before applying the template patterns to the neural network.

Abstract: Method and system for selectively retrieving information contained in a stored document set using a metric-based or "fuzzy" finite-state non-deterministic automaton. An automaton is constructed (501) corresponding to a text string query, text strings are read (502) from storage and corresponding dissimilarity values are generated (505). Those strings resulting in values less than a given threshold are recorded (508) and listed for the user. Dissimilarity values are determined based on penalties associated with missing characters, extra characters, incorrect characters, and other differences between the text string query and a text string read from storage.

Abstract: Method and system for selectively retrieving information contained in a stored document set using a non-literal, or "fuzzy", search strategy. A text string query is transmitted (200) to a computer processor, and a dissimilarity value D.sub.i is assigned (208) to selected ones of stored text strings representative of information contained in a stored document set, based upon a first set of rules (106). A set of retrieved text strings representative of stored information and related to the text string query is generated (212), based upon a second set of rules (107). Each of the retrieved text strings has an associated dissimilarity value D.sub.i, which is a function of at least one rule R.sub.n from the first set of rules (106) used to retrieve the text string and a weight value w.sub.n associated with that rule R.sub.n. The retrieved text strings are displayed (216) preferably in an order based on their associated dissimilarity value D.sub.i.

Abstract: Display standards in common use for the display of computer or television images on high resolution displays, commonly assume that the output image will be displayed on a output device having a high refresh rate. A high refresh rate is normally required to avoid the viewer observing flicker, stilted motion or other visual artifacts if a lower refresh rate were used. It is difficult to drive a high resolution Ferroelectric Liquid Crystal Display at a high refresh rate. The subject apparatus is provided for using the memory function characteristics of such a display and driving such a display at a slower refresh rate while still maintaining the appearance of a device having a higher refresh rate by refreshing those portions of the screen where motion has been detected at a high rate and only occasionally refreshing the whole screen.

Abstract: Display standards in common use for the display of computer or television images on high resolution displays, commonly assume that the output image will be displayed on an output device having a high refresh rate. A high refresh rate is normally required so the viewer does not observe flicker, stilted motion or other visual artifacts if a lower refresh rate were used. It is difficult to drive some displays at a high refresh rate. An apparatus is disclosed for detecting areas of an input image that are changing over time, even where the input image may be subject to some degree of noise. A priority value is determined for providing a measure of this change for use with a system to determine which areas of the image may need changing as a matter of high priority.

Abstract: A general method of accelerating the classification of an unclassified input symbol S into a library of pre-classified templates employs a comparison inequality function. The correlation calculation is initiated between the input symbol S and templates retrieved from the library of templates. Templates with a low degree of correlation are excluded prior to completion of the correlation calculation to accelerate the correlation. Sorting by temple intensity permits the temple foreground pixels to be correlated first, which also accelerates the correlation calculation. A second sort by discrimination power permits critical conflict pixels to be correlated first for further accelerating the correlation calculation.

Abstract: In order to display continuous tone colour image on a discrete colour level display, methods of halftoning must be used. The high display rate of colour output devices means that serial methods of real time halftoning are difficult to use. A method and apparatus are disclosed for reducing the speed with which a halftoning method must be performed by performing the halftoning of an output image in parallel by simultaneously error diffusing more than one line of input at a time.

Abstract: A scanner (12), computer (2), and printer (13) are preferably combined within a single housing (1). The scanner (12) optically scans an input paper document (11) into a storage buffer (22) within the computer (2). The scanned information (22) is edited by automatic means (21), human-activated means (23, 3), or some combination thereof. The edited information is printed by the printer (13) onto the original input paper document (11). A single preferably linear paper transport mechanism (4) is used to move the document (11) through the scanner (12) and the printer (13). An authentication code (14, 16) may be superimposed onto the document (11) to prevent the unauthorized editing thereof.

Abstract: Pattern recognition, for instance optical character recognition, is achieved by defining a minimal bounding rectangle around a pattern, dividing the pattern into a grid of boxes, comparing a vector derived from this partitioned pattern to vectors similarly derived from known patterns, choosing a set of Pareto non-inferior candidate patterns, and selecting a recognized pattern from the set of candidates. The vectors include pixel density matrices, matrices of horizontal connectivity of boxes, and matrices of vertical connectivity of boxes.

Abstract: A library of L unenhanced images (pixel bit maps) is enhanced for optical character recognition (OCR) with respect to a pre-existing group of G input symbol classes for creating a library of G recognition enhanced templates (pixel bit maps) of the G input symbol classes. The enhancement is accomplished by comparing each image of the library with each pixel version of each symbol class (forming V sets of L comparisons), and weighting the images with the highest potential for confusion. Divergent pixel version are further divided into sub-classes, each with a sub-image which evolves into a sub-template. A primary comparison C* and a secondary comparison C** are identified from the L comparisons within each of the V sets of comparisons. A recognition margin is determined between each pair of identified comparisons C* and C**. The single pair of identified comparisons C* and C** is selected forming the smallest recognition margin M*. The single pair of images I.sub.j * and I.sub.

Abstract: A process of producing large dither matrices suitable in use in halftoning images is disclosed. This process includes forming an objective function based on the placement of the dither values within the matrix and optimizing the objective function so as to produce an improved matrix. The preferred optimizations process includes the use of simulated annealing with additional optimizations of the simulated annealing process being disclosed.

Abstract: In order to display a continuous tone colour image on a discrete colour level display, methods of halftoning must be used. The high display rate of colour output devices means that serial methods of real time halftoning are difficult to use. A method and apparatus is disclosed for reducing the speed with which a halftoning method must be performed by performing the halftoning of an output image by simultaneously dividing the input image into a number of areas and simultaneously halftoning the areas individually making special provisions for pixels located in the boarder regions of a given area.

Abstract: Apparatus for sending and receiving both analog and digital information from a single information transmission medium. The apparatus includes a core processor (200), including a digital signal processor (DSP) (212) or a CPU/DSP, in communication with an integrated services digital network (ISDN) interface (204). The core processor (200) also may be in communication with a personal computer (PC) (202). The invention also is directed to a related method, including a series of modules.

Abstract: Methods of correcting color images often involve a complex interrelation of many color components and require extensive training to be able to be fully utilized. The present disclosure involves a simplified form of color correction. From the selection of a series of actual (16,17) and desired (18,19) colors, a series of contrast and brightness values are derived, which can then be applied to the color image to produce the color corrected image.

Abstract: Stroke-based symbols such as conventional alpha-numerics are formed by individual strokes. Each stroke has intersections or node regions common to two or more strokes and main inter-node body regions unique to a single stroke. Strokes are a primary recognition cue to the human eye and for OCR techniques. The length and orientation of each stroke is determined and entered into a library of extracted strokes for analysis and matching. After selection each stroke is physically cut from the stroke based symbol to assist in selecting the next stroke. The body regions are cut first. The common node regions are temporarily protected to preserve the end nodes of the remaining strokes. Each node is cut during the extraction sequence of the last stroke common to that node. This sequential removal of strokes results in a progressively diminishing symbol residue and a progressively increasing library of extracted strokes.

Abstract: Display standards in common use for the display of computer or television images on high resolution displays commonly assume that the output image will be displayed on an output device having a high refresh rate. A high refresh rate is normally required so the viewer does not observe flicker, stilted motion or other visual artifacts that one might observe if a lower refresh rate were used. It is difficult to drive some displays at a high refresh rate. An apparatus is disclosed for detecting areas of an input image that are changing over time, even where the input image may be subject to some degree of noise. A priority value is determined for providing a measure of this change for use with a system to determine which areas of the image may need changing as a matter of high priority.

Abstract: A digital data base 10 having record data codes representing an initial digital record is printed at predetermined code sites 12 within two dimensional code storage area 14 on the surface of on base medium 10B. The digital data base includes redundant check codes and one or more redundant check-check codes for permitting data error and data loss correction. The two dimensional code storage area has a coordinate system for defining code positions therein. The code sites are physically positioned within the code storage area at predetermined coordinates. A plurality of error checking sets are formed by the code sites, each containing one or more data codes and one or more check codes. The bits in the check codes in each error checking set are determined by the bits in the data codes of the same error checking set, and are in error correcting relationship therewith. Each data code and each check code are included in "L" error checking sets to provide "L" levels of inter-locking error checking and correcting.

Abstract: Pattern recognition, for instance optical character recognition, is achieved by training a neural network, scanning an image, segmenting the image to detect a pattern, preprocessing the detected pattern, and applying the preprocessed detected pattern to the trained neural network. The preprocessing includes determining a centroid of the pattern and centrally positioning the centroid in a frame containing the pattern. The training of the neural network includes randomly displacing template patterns within frames before applying the template patterns to the neural network.

Abstract: A window assembly for transmitting relatively high power microwave energy from a waveguide, held at substantially atmospheric pressure levels, into a microwave reaction chamber at sub-atmospheric pressure levels. The window assembly provides for the transmission of microwave energy to generate a glow discharge plasma without suffering from catastrophic failure as a result of excessive temperature and pressure conditions.