Abstract: A compact disc (CD) read sensor apparatus for reading data from a CD track uses multiple laser beams for performing both tracking and reading data. In one embodiment, two laser beams are used to track by controlling the radial position of the two beams that are nominally radially disposed towards opposite edges of the recorded track. A differential signal is used to obtain tracking information and a sum signal is used to produce a read signal. In another embodiment, three laser beams are used corresponding to prior art three beam systems using two beams nominally disposed towards opposite edges for tracking and a center beam for reading. The three beams are combined to form an enhanced read signal with an improved signal-to-noise ratio over that obtained by any one beam.

Abstract: A multi-function machine for combining and routing image data is disclosed. The multi-function machine includes a number of functional units, for instance, a digital copier which produces scanned image data, a printer which prints image data from a computer, and a facsimile device which receives and sends facsimile image data. Image data from one or a number of sources may be combined to form merged image data, corresponding to a merger of the individual images. The merged image data may then be conveyed to any of a number of user selectable destinations, including a printer, a host computer, or a facsimile device.

Abstract: A method and apparatus for designing a multilayer feed forward neural network that produces a design having a minimum number of connecting weights is based on a novel iterative procedure for inverting the full Hessian matrix of the neural network. The inversion of the full Hessian matrix results in a practical strategy for pruning weights of a trained neural network. The error caused by pruning is minimized by a correction that is applied to remaining (un-pruned) weights thus reducing the need for retraining. However, retraining may be applied to the network possibly leading to further the simplification of the network design.

Abstract: Method for two-way digital communication between Data Terminal Equipment (DTE) and Data Communication Equipment (DCE), includes a novel command set. The inventive method involves initiating communication by transmitting one command from a set of DTE commands to the command interpreter block. That block interprets the command, and acts on the command, directing the command to the appropriate control block. The control block receiving the decoded command then executes the command by sending a signal representative of the command to a hardware interface device which transmits the signal to the receiving DCE.

Abstract: A speaker recognition method uses visual image representations of mouth movements associated with the generation of an acoustic utterance by a speaker that is the person to be recognized. No acoustic data is used and normal ambient lighting conditions are used. The method generates a spatiotemporal gray-level function representative of the spatiotemporal inner month area confined between the lips during the utterance from which a cue-block is generated that isolates the essential information from which a feature vector for recognition is generated. The feature vector includes utterance duration, maximum lip-to-lip separation, and location in time, or speed of lip movement opening, speed of lip movement closure, and a spatiotemporal area measure representative of the area enclosed between the lips during the utterance and representative of the frontal area of the oral cavity during the utterance. Experimental data shows distinct clustering in feature space for different speakers.

Abstract: An apparatus and method for increasing the throughput of a data processing system such as an acoustic or image compression system. An addition or subtraction of two pluralities of numbers is accomplished in parallel by combining numbers into a pair of "doublevectors," adding or subtracting the pair of doublevectors, and separating the resultant doublevector to provide values representative of the results of the addition or subtraction on the original plurality of numbers. Similarly, a left-shift of a plurality of numbers is accomplished by combining the numbers into a single doublevector, left-shifting the doublevector, and extracting a plurality of output values representative of the left-shifted values of the original numbers. Using additions, subtractions and shifts a linear transform such as the Generalized Chen Transform may be performed using doublevectors to provide a substantial increase in computation speed.

Abstract: A system for restoring images with undefined pixel values at known locations is described. The threshold value and a neighborhood configuration are defined and are used to restore the image. The neighborhood configuration defines a geometric region, typically a fixed number of pixels, surrounding the target pixel. The threshold value specifies a number of pixels in the neighborhood configuration for which pixel values are known. In our system, for each pixel in one of the unknown regions an analysis is performed over the entire area defined by the neighborhood configuration. If the threshold number of pixels within that region is known, then the value of the unknown pixel is calculated. If the threshold value is not achieved, then analysis proceeds to the next pixel location. By continuing the process and reducing the threshold value when necessary or desirable, the complete image can be restored.

Abstract: The apparatus for the recognition of speech includes an acoustic preprocessor, a visual preprocessor, and a speech classifier that operates on the acoustic and visual preprocessed data. The acoustic preprocessor comprises a log mel spectrum analyzer that produces an equal mel bandwidth log power spectrum. The visual processor detects the motion of a set of fiducial markers on the speaker's face and extracts a set of normalized distance vectors describing lip and mouth movement. The speech classifier uses a multilevel time-delay neural network operating on the preprocessed acoustic and visual data to form an output probability distribution that indicates the probability of each candidate utterance having been spoken, based on the acoustic and visual data. The training system includes the speech recognition apparatus and a control processor with an associated memory.

Abstract: Transforms such as the DCT are useful for image compression. One close relative of the DCT is preferred for its arithmetic simplicity. A method and apparatus is described whereby the quantization operates with speeds comparable to 16 bit transforms yet the mean squared error is comparable to that of a 32 bit transform. The factorability of the transform into a relatively rapid set of additions and a set of multiplications permits an efficient pipelined data flow wherein the addition sections of the vertical and horizontal transforms are performed by the same hardware prior to a final multiplication.

Abstract: The apparatus for the recognition of speech comprises an acoustic preprocessor, a visual preprocessor, and a speech classifier that operates on the acoustic and visual preprocessed data. The acoustic preprocessor comprises a log mel spectrum analyzer that produces an equal mel bandwidth log power spectrum. The visual processor detects the motion of a set of fiducial markers on the speaker's face and extracts a set of normalized distance vectors describing lip and mouth movement. The speech classifier uses a multilevel time-delay neural network operating on the preprocessed acoustic and visual data to form an output probability distribution that indicates the probability of each candidate utterance having been spoken, based on the acoustic and visual data.

Abstract: The present invention provides a method and apparatus for encoding and decoding data in parallel. The present invention provides a system for decompressing a data stream having multiple codewords. The system includes an input channel that receives the data stream. The system also includes a decoder which decodes each bit of the data stream, wherein at least two of the codewords in the data stream are decoded at the same time, such that the data stream is decoded in parallel.

Abstract: A method and apparatus for controlling and communicating with business office devices, such as copiers, facsimiles and/or printers. The present invention communicates and controls various modules of business devices which allow an external device such as an operation panel to access the state of a target device, such as a copier, printer or facsimile. The operation panel can communicate with the target device and control the same target device. Also, a remote diagnostics station can provide remote diagnostics of the target device.

Abstract: A method and apparatus for formatting a document and creating a best document layout from an input list of picture and text objects is disclosed. The method includes calculating multiple document layouts while maintaining the correct reading order of the picture and text objects at all times. The method positions each picture and text object at multiple anchor points to create multiple document layouts, and then selects a best document layout which is the layout using the least number of pages to display the entire list of objects. If more than one layout uses the least number of pages, the layout positioning the least number of objects on the last page is the best layout.

Abstract: A system for analyzing pages has a circuit for generating page information, a circuit for determining if the page information represents a blank page, a blank page counter that is incremented if a blank page is detected, a comparator for comparing the blank page counter to a blank page threshold memory, and an error indicating circuit that indicates an error signal if the blank page counter exceeds the blank page threshold and a detect signal is active. The page analysis system detects blank pages within a job and distinguishes between initial blank pages caused by incorrectly oriented original pages within a machine, and intentionally blank pages within the middle of a job.

Abstract: A data communication apparatus and associated method for switching from the Compatible mode to the ECM mode. The data communication apparatus receives a protocol message transmitted by a second data communication apparatus in accordance with a Compatible mode of operation. The first data communication apparatus then inverts the polarity of the received protocol message so that the inverted polarity protocol message is formatted in accordance with a G3 mode of operation. Then, the first data communication apparatus determines from the inverted polarity protocol message that the second data communication apparatus will receive the image data if it is transmitted in accordance with an ECM mode of operation. The image data is then transmitted in accordance with the ECM mode of operation to the second data communication apparatus.

Abstract: A system for compressing numbers represented in clear text format of a Standard Page Description Language (SPDL) in which numbers are compressed to a binary format represented by N/2.sup.r. Only numbers which are integers or have a decimal portion following a certain format can be compressed using this format. If the number is an integer, it could be converted setting N equal to the number and r equal to zero. If the number contains a decimal portion, the number is doubled and then checked to see if it has become an integer. If it has not become an integer, it is repeatedly doubled until the integer portion N or r go out of range. A second embodiment of the compression system operates in a similar manner if the number to be compressed is an integer. If it is not an integer, the decimal portion of the number is broken into groups having four consecutive digits. The right most group of digits is examined to determine if it is characteristics which allow compression using the N/2.sup.r format.

Abstract: A system is described for automatically invoking computational resources without intervention or request from a user of the system. In the system a query-free information retrieval system is described in which the exact technical documentation contained in existing user or other technical manuals is provided to a user investigating apparatus having a fault. The user enters symptoms based upon the user's analysis of the apparatus, and in response the system provides information concerning likely faults with the apparatus. As the symptoms are entered, the relative value of individual faults is determined and related to the symptoms they cause. The user can then select technical information relating to probable faults in the system.

Abstract: A method and apparatus for controlling and communicating with business office devices, such as copiers, facsimiles and/or printers. The present invention communicates and controls various modules of business devices which allow an external device such as an operation panel to access the state of a target device, such as a copier, printer or facsimile. The operation panel can communicate with the target device and control the same target device. Also, a remote diagnostics station can provide remote diagnostics of the target device.

Abstract: A method and apparatus for compressing images having a binary extractor, a binary compressor and a JPEG compressor. The binary extractor produces a binary image representing an estimate of the binary information in the original image. The binary image is compressed by the binary compressor. The binary image is also subtracted from the original image to produce a difference image representing the continuous-tone information in the original image. The difference image is then compressed using the JPEG compressor. In this manner, the present invention compresses a document image without having to perform segmentation on the image.

Abstract: A method and apparatus for controlling and communicating with business office devices, such as copiers, facsimiles and/or printers. The present invention communicates and controls various modules of business devices which allow an external device such as an operation panel to access the state of a target device, such as a copier, printer or facsimile. The operation panel can communicate with the target device and control the same target device. Also, a remote diagnostics station can provide remote diagnostics of the target device.