초록 ▼

A computerized classifier system that forms convex hulls containing all training experiences for each target class (e.g. threat/non-threat) is disclosed. The intersection of convex hulls for each pair of classes defines a region of ambiguity in feature space. Measurement of feature vector(s) outside

A computerized classifier system that forms convex hulls containing all training experiences for each target class (e.g. threat/non-threat) is disclosed. The intersection of convex hulls for each pair of classes defines a region of ambiguity in feature space. Measurement of feature vector(s) outside an ambiguous region of feature space leads to a class decision while measurement of feature vector(s) within an ambiguous region of feature space defined by convex hulls causes a transition to a new feature space with new features. In particular embodiments, measured feature data includes estimated motion states and electrical lengths of a given object, and range, velocity and acceleration image data from second phase differences for debris mitigation.

대표청구항 ▼

1. A computer-implemented method for classifying a change in the state of a target being monitored, said method comprising: measuring feature data associated with said target during a defined time duration;processing using a computer processor, real time sensor data from said target using a graph da

1. A computer-implemented method for classifying a change in the state of a target being monitored, said method comprising: measuring feature data associated with said target during a defined time duration;processing using a computer processor, real time sensor data from said target using a graph data structure that has defined therein convex hulls having regions of ambiguity corresponding to each of a set of target classes corresponding to target features, wherein the intersection of convex hulls for each pair of classes defines the region of ambiguity in feature space;transitioning, according to the graph, to another feature space having different features, in response to measurement data of feature vectors within an ambiguous region of feature space defined by convex hulls; anddetermining using a computer processor, a next sensor measurement data for obtaining additional measured feature data of the target according to the current state of the graph data structure and the region of ambiguity; andrepeating said processing, transitioning and determining steps until the change in state of the target is classified. 2. The computer-implemented method of claim 1, wherein the target being monitored is a radar target. 3. The computer-implemented method of claim 1, wherein traversal of the graph includes determining information relevant to the particular region of ambiguity associated with the target likely to reduce and/or eliminate the ambiguous region. 4. The computer-implemented method of claim 3, wherein factors including resource allocation, sensor measurement classification time constraints, feature relevance, accuracy and cost, are accounted for in generating the graph data structure. 5. The computer-implemented method of claim 3, wherein the determination and selection of next sensor measurement is accomplished according to the graph data structure and the features are sensed and analyzed using the current sensor. 6. A method for classifying unclassified vehicle tracks using real-time sensor data comprising: acquiring a computer database of vehicle track characteristics data for known vehicle tracks;forming convex hulls using the vehicle track characteristics data for each target class;defining a region of ambiguity in feature space as the intersection of convex hulls for each pair of target classes; andtransitioning to another feature space having different features in response to measurement data of feature vectors within an ambiguous region of feature space defined by convex hulls. 7. The method of claim 6, further comprising automatically load shedding by discriminating non-threatening tracks. 8. The method of claim 6, wherein the measured feature data includes estimated motion states and electrical lengths of a given object. 9. The method of claim 6, wherein the measured feature data includes range, velocity and acceleration image data from second phase differences for debris mitigation. 10. The method of claim 6, wherein the measured feature data is obtained via a neural network algorithm. 11. A hierarchical classifier computer system comprising a database of defined vehicle track signatures for known vehicle tracks;a first processor executing instructions for generating a graph from the database of defined vehicle track signatures by defining state nodes corresponding to characteristics of the vehicle track signatures and defining links between defined state nodes, the links between the defined state nodes corresponding to defined vehicle track signatures and defining regions of ambiguity within a convex hull; anda second processor executing instructions for performing run-time processing of real-time sensor data corresponding to a first unclassified vehicle track by passing the real-time sensor data through the graph and performing state transitions based on the features of the measured sensor data, the region of ambiguity, and additional select feature data, until the first unclassified vehicle track is classified.