METHOD AND APPARATUS FOR PREDICTING OBJECT PROPERTIES AND EVENTS USING SIMILARITY-BASED INFORMATION RETRIEVAL AND MODELING

• US 20100332475A1
• Filed: 06/25/2010
• Published: 12/30/2010
• Est. Priority Date: 06/25/2009
• Status: Active Grant

• Current Assignee: United States Secretary of the Army, University of Tennessee Research Foundation (University of Tennessee)
• Original Assignee: University of Tennessee Research Foundation (University of Tennessee), US Army Space & Missile Defense Command/Army Forces Strategic Command
• Inventors: Stansberry, Dale V., Birdwell, J. Douglas, Horn, Sally P., Horn, Roger, Wang, Tse-Wei, Icove, David J., Rader, Mark
• Granted Patent: US 8,396,870 B2
• Time in Patent Office: Days
• US Class Current: 707/737
• CPC Class Codes: G06F 16/2246 Trees, e.g. B+trees
• CPC Class Codes: G06F 16/2264 Multidimensional index structures
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/283 Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP
• CPC Class Codes: G06F 18/2323 based on graph theory, e.g. minimum spanning trees [MST] or graph cuts
• CPC Class Codes: G06F 18/40 Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
• CPC Class Codes: G06V 10/7635 based on graphs, e.g. graph cuts or spectral clustering

Abstract

Method and apparatus for predicting properties of a target object comprise application of a search manager for analyzing parameters of a plurality of databases for a plurality of objects, the databases comprising an electrical, electromagnetic, acoustic spectral database (ESD), a micro-body assemblage database (MAD) and a database of image data whereby the databases store data objects containing identifying features, source information and information on site properties and context including time and frequency varying data. The method comprises application of multivariate statistical analysis and principal component analysis in combination with content-based image retrieval for providing two-dimensional attributes of three dimensional objects, for example, via preferential image segmentation using a tree of shapes and to predict further properties of objects by means of k-means clustering and related methods. By way of example, a fire event and residual objects may be located and qualified such that, for example, properties of the residual objects may be qualified, for example, via black body radiation and micro-body databases including charcoal assemblages.

Claims

1. A method of predicting a property of a target object via inputting a query to a processor search manager apparatus, the processor search manager apparatus being coupled to a plurality of databases and comprising a processor and memory, the plurality of databases comprising a micro-body assemblage database and a spectral database, the method comprising:

storing data objects in said micro-body assemblage database, each assemblage data object comprising micro-body assemblage data and one of a geographic location property and an environmental property associated with the micro-body assemblage data;

storing data objects in said spectral database, each spectral data object comprising spectral data and at least one attribute associated with the spectral data;

processing a query comprising information about a target object, the query information containing one of assemblage data and spectral data;

using said search manager apparatus to search the micro-body assemblage database if assemblage data are present in the query and to search the spectral database if spectral data are present in the query;

retrieving stored information about data objects having one of similar micro-body assemblage and spectra; and

using a model and the retrieved stored information to predict at least one property of the target object in response to the query.

2. The method of claim 1 wherein said spectra data comprise one of acoustic, electrical, or electromagnetic spectra.

3. The method of claim 1 wherein said spectral data comprise mass spectra.

4. The method of claim 3 wherein said mass spectra represent information about stable isotope ratios in specimens.

5. The method of claim 1 wherein the predicted property of the target object comprises information about at least one geographic location associated with the target object.

6. The method of claim 1 wherein the predicted property of the target object comprises information about at least one environmental property associated with the target object.

7. The method of claim 1 wherein the predicted property of the target object comprises one of a possible source and origin of the target object.

8. The method of claim 1 wherein at least one database is implemented using a second processor.

9. The method of claim 1 further comprising generating an index for at least one database having data objects, each data object comprising a vector of attributes, said index generated by

analyzing the attributes utilizing multivariate statistical analysis to determine a subspace of an attributes space, the subspace having at least two dimensions;

clustering the data objects into groups of data objects where a measure of similarity between data objects comprises projecting attribute vectors onto the attributes subspace; and

determining projections onto a subspace for clustering and indexing data objects in said groups of data objects to more efficiently access stored data objects having attributes similar to the attributes of a target object.

10. The method of claim 1 further comprising generating a third database of image data for said plurality of objects and the target object.

11. The method of claim 1 wherein said spectral data comprises acoustic and electromagnetic spectral data for a target object and another object displayed as a plurality of linked points from the target object to a predicted electro-acoustic spectral property.

12. The method of claim 1 wherein the attributes subspace is the object clustering and indexing subspace.

13. The method of claim 1 wherein said spectral data comprise one of acoustic and electromagnetic property data, the method further comprising displaying an evidence tree from the target object to electro-acoustic property data via a linked path.

14. The method of claim 1 wherein said spectral data comprise acoustic and electromagnetic property data, the method comprising displaying the target object comprising linked paths to a plurality of micro-body assemblages and associated data and to electro-acoustic spectral data within a circular target object space.

15. The method of claim 14 wherein micro-body assemblage data of the evidence tree is differentiated by color from electro-acoustic spectral data for the target object.

16. A method of predicting object properties of a target object via inputting a query to a plurality of databases comprising a micro-body assemblage database and a spectral database for objects coupled to a processor search manager apparatus, the processor search manager apparatus comprising a processor and memory, the method comprising:

determining a set of objects most similar to a target object depicted by a central circle via said processor search manager apparatus, the set of objects being displayed as a set of circles surrounding a circle representing the target object,

applying at least one model to predict a location property of the target object,

comparing hypothetical relationships involving the target object and a subset of the set of similar objects, and

determining a probable relationship and a location property value of the target object.

17. The method of claim 16 further comprising inferring a location property value for a newly acquired object by modeling how a property value varies with other attributes of the newly acquired object.

18. The method of claim 16 wherein the plurality of databases further comprises a database of image data for objects.

19. The method of claim 16 wherein said at least one model comprises a Bayesian model.

20. A method of predicting object properties of a target object via inputting a query to a plurality of databases for storing objects, the plurality of databases coupled to a processor search manager apparatus, the processor search manager apparatus comprising a processor and memory, the method comprising:

searching via said processor search manager apparatus each of at least two databases for stored objects having attributes similar to at least one selected attribute of said target object, said selected attribute being different for each database, the two databases comprising a micro-body assemblage database and an electro-acoustic spectral database for each object;

retrieving properties or attribute values associated with said stored objects in each database; and

determining a predicted object location property of said target object using said retrieved properties or attribute values and a model comprising a function of one of the retrieved properties and attribute values.

21. The method of claim 20 further comprising

using at least two models and predicted object properties to determine a likely relationship among objects.

22. The method of claim 21 wherein one of the objects is the target object.

23. The method of claim 20 wherein the plurality of databases comprise a micro-body assemblage database and an electro-acoustic spectra database, evidence trees for the target objects and linked objects to the target objects being displayed as circles and each circle comprising related data of the databases are represented as linked points.

24. The method of claim 20 further comprising clustering object data into groups of objects and identifying a group of objects most similar to the target object and displaying the group of similar objects as circles surrounding a central circle for the target object.

25. The method of claim 24 further comprising predicting object location properties of the target object from micro-assemblage data properties of said group of objects.

26. The method of claim 20 wherein the model comprises a principal component analysis model.

27. The method of claim 20 wherein the model comprises a multivariate data model.

28. The method of claim 20 wherein the model comprises a Bayesian model.

29. The method of claim 20 wherein the model comprises a least squares optimization model.

30. The method of claim 20 wherein the model comprises a maximum likelihood optimization model.

31. The method of claim 20 wherein the model comprises a tree of shapes model.

32. The method of claim 20 wherein the model comprises an artificial neural network.

33. The method of claim 20 wherein the model comprises a fuzzy logic model.

34. The method of claim 20 wherein the model comprises a hierarchical data model.

35. The method of claim 20 further comprising displaying a linkage between micro-body assemblage data of the target object and data identifying a geographical location of an identified micro-body assemblage.

36. A computer-implemented method of dynamic indexing of spectral data of a spectral database tree having a root node, the method implemented on processor search manager apparatus comprising a processor and memory, the method comprising:

storing spectral data in a database of said memory, the spectral data being represented by an attribute vector having a plurality of M dimensions, where M is at least three;

defining a root node of a spectral database tree and a test for the root node responsive to a query to said processor search manager apparatus related to an object, the test defining an indexing surface in a three dimensional sub-space for separating the spectral data into at least two clusters of spectral data;

applying a test for another node of the spectral database tree, the processor outputting reference evidence data responsive to said query comprising one of a location and an event related to the query object.