Abstract: A system (10) for recognizing numerical characters which appear as both numerical characters and words on a document includes an image capture device (12) for capturing the image of a document. A recognition circuit (14) performs recognition of the numerical characters. A recognition circuit (16) performs recognition of the words corresponding to the numerical characters. A comparator circuit (20) compares the recognition signals generated by the recognition circuit (16) to the recognition signals generated by the recognition circuit (14) to determine if the numerical characters recognized by the recognition circuit (14) is correct.

Abstract: A video image processor for generating an enhanced image of data based upon input gray scale picture element values of image data of a document background and printed information contained thereon generates a histogram of the gray scale picture element values for each tile of image data for an area of the document representing the frequency of occurrence of picture elements of a particular gray scale value. A correlator processes each histogram to determine a back reference level and a white reference level for each tile. Normalization factors are calculated based upon the black reference level and the white reference level. The input gray scale picture element values for each tile of the image data associated with the normalization factors of each tile are normalized based upon the normalization factors for each tile and correlated to generate a normalized and correlated representation of the image data.

Abstract: A neural-simulating system for processing input stimuli includes a plurality of layers, each layer receives layer input signals and generates layer output signals, the layer input signals include signals from the input stimuli and ones of the layer output signals from only previous layers within the plurality of layers. Each of the plurality of layers includes a plurality of neurons operating in parallel on the layer input signals applied to the plurality of layers. Each of the neurons derives neuron output signals from a continuously differentiable transfer function for each of the neurons based upon a combination of sets of weights associated with the neurons and the layer input signals. An adaptive network is associated with each neuron for generating weight correction signals based upon gradient estimate signals and convergence factors signals of each neuron and for processing the weight correction signals to thereby modify the weights associated with each neuron.

Abstract: A neural-simulating system for an image processing system includes a plurality of networks arranged in a plurality of layers, the output signals of ones of the layers provide input signals to the others of the layers. Each of the plurality of layers include a plurality of neurons operating in parallel on the input signals to the layers. The plurality of neurons within a layer are arranged in groups. Each of the neurons within a group operates in parallel on the input signals. Each neuron within a group of neurons operates to extract a specific feature of an area of the image being processed. Each of the neurons derives output signals from the input signals representing the relative weight of the input signal applied thereto based upon a continuously differential transfer function for each function.

Abstract: A signal processing system (10) is provided for digitally filtering an input digital signal. The signal processing system (10) includes a digital filter (12) without normal finite precision arithmetic whose output is applied to an adaptive quantizer (14). The output of the adoptive quantizer (14) includes a digital signal having a number of bits per sample which is less than or equal to the number of bits per sample contained in the output of the digital filter (12).

Abstract: An image processing system (10) for displaying a video image of a document includes structure for scanning (12) the document and for generating (24) pixel signals representing gray level data values of the image of the document. Circuitry (80) generates a set of color data values and assigns one of the color data values to pixels having a predetermined gray level data value. Structure (42, 84) displays the pixels having an assigned color data value and generates a color image of the document.

Abstract: A dot matrix print detector (24) for use in an optical character recognition system for detecting the presence of characters on a document printed in a dot matrix print format is provided. A scanning device (26) generates pixel signals representing pixel values of portions of characters in a predetermined area appearing on the document. A filter (28) functions to generate filtered signals representing energy levels of pixels associated with the edges of the characters. Summation circuitry (30, 32) is provided for determining the energy levels of the filtered signals for the predetermined area and for generating a summation signal. Summation circuitry (38, 40) generates a summation signal representing the energy levels of unfiltered pixel elements or low pass filtered pixel elements. The output of the summation circuitry (40, 32) is applied to a decision circuit (34) for determining the presence of dot matrix print in the predetermined area.