Abstract: A classification neural network for piecewise linearly separating an input space to classify input patterns is described. The multilayered neural network comprises an input node, a plurality of difference nodes in a first layer, a minimum node, a plurality of perceptron nodes in a second layer and an output node. In operation, the input node broadcasts the input pattern to all of the difference nodes. The difference nodes, along with the minimum node, identify in which vornoi cell of the piecewise linear separation the input pattern lies. The difference node defining the vornoi cell localizes input pattern to a local coordinate space and sends it to a corresponding perceptron, which produces a class designator for the input pattern.

Abstract: A biometric recognition system involves two phases: creation of a master pattern set of authorized users biometric indicia and authentication using a classification neural network. To create the master pattern set, an image of an authorized biometric user's indicia is divided into a plurality of regions of interest or "features". The system determines which features are the most useful for identification purposes. Master patterns are then created from these master features, thus creating a master pattern set. During authentication, a sample pattern set of a user to be authenticated is similarly created. A neural network is used to compare the sample pattern set with the master pattern set to determine whether the user should be authenticated.

Abstract: A method and apparatus for classification and tracking objects in three-dimensional space is described. A machine vision system acquires images from roadway scenes and processes the images by analyzing the intensities of edge elements within the image. The system then applies fuzzy set theory to the location and angles of each pixel after the pixel intensities have been characterized by vectors. A neural network interprets the data created by the fuzzy set operators and classifies objects within the roadway scene. The system can also track objects within the roadway scene, such as vehicles, by forecasting potential track regions and then calculating match scores for each potential track region based on how well the edge elements from the target track regions match those from the source region as weighted by the extent the edge elements have moved.

Abstract: A method and apparatus for producing a background image from a plurality of images of a scene and for subtracting a background image from an input image are described. A background image is produced by dividing an image into subimages, acquiring reference subimages for each subimage location and comparing subsequent subimages with the reference subimage to determine if any objects have passed between the reference subimage and the video camera that acquired images. When objects have passed between the reference subimage and the video camera, the reference subimage is designated as background and stored in a background image. Background portions of an input image can be removed or their intensity diminished with a background image. Foreground weights can be determined by comparing the difference between a background image and an input image. To the extent that corresponding pixels are the same, the pixel is given a low foreground weight, indicating that the pixel is a background weight.

Abstract: A method and apparatus defines boundaries of the roadway and the lanes therein from images provided by video. The images of the roadway are analyzed by measuring motion between images and detecting edges within motion images to locate edges moving parallel to the motion of the objects, such as vehicles, thereby defining the approximate boundaries of a lane or roadway. A curve is then generated based on the approximate boundaries to define the boundaries of the lane or roadway.

Abstract: A system for classifying vehicles based on the sound waved produced by the vehicles receives analog sound pressure levels and converts them to a power spectrum. Fuzzification functions, such as asymmetric wedge shaped functions, are convoluted with the power spectrum to create a vector that characterizes the power spectrum while reducing the dimensionality of the characterizing vector. A neural network analyzes the characterizing vector and produces a classification designator indicative of the class of the object associated with the analog sound pressure levels received by the system.

Abstract: A calibration apparatus for calibrating three-dimensional space for a machine vision system and a method for using the calibration apparatus for calibrating the machine vision system are disclosed. The calibration apparatus comprises pairs of indicators, such as lights, mounted on a frame, each pair of lights separated by a known horizontal distance. The calibration apparatus may be mounted on the back of a vehicle that moves the apparatus through a roadway scene at a known speed. A machine vision system tracks the calibration apparatus, specifically the pairs of lights mounted on the frame, and maps the pixel space of the machine vision system to three-dimensional space.

Abstract: A calibration apparatus for calibrating three-dimensional space for a machine vision system and a method for using the calibration apparatus for calibrating the machine vision system are disclosed. The calibration apparatus comprises pairs of indicators, such as lights, mounted on a frame, each pair of lights separated by a known horizontal distance. The calibration apparatus may be mounted on the back of a vehicle that moves the apparatus through a roadway scene at a known speed. A machine vision system tracks the calibration apparatus, specifically the pairs of lights mounted on the frame, and maps the pixel space of the machine vision system to three-dimensional space.

Abstract: A method and apparatus for classification and tracking objects in three-dimensional space is described. A machine vision system acquires images from roadway scenes and processes the images by analyzing the intensities of edge elements within the image. The system then applies fuzzy set theory to the location and angles of each pixel after the pixel intensities have been characterized by vectors. A neural network interprets the data created by the fuzzy set operators and classifies objects within the roadway scene. The system can also track objects within the roadway scene, such as vehicle, by forecasting potential track regions and then calculating match scores for each potential track region based on how well the edge elements from the target track regions match those from the source region as weighted by the extent the edge elements have moved.