Method and apparatus for predicting object properties using similarity based information retrieval and modeling

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 and thermal 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, an evidence tree display showing a target object linked by a pathway to a predicted property comprises a similarity value, a speculation value and a model-based rank.

What we claim is: 
     
         1 . 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; and   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, a manufacturer and an event related to the query object, the query object being a target object.   
     
     
         2 . The computer-implemented method of  claim 1 , further comprising:
 using Euclidean distance to determine similarity between spectra data samples.   
     
     
         3 . The computer-implemented method of  claim 1 , further comprising:
 using a cosine distance metric to determine similarity between spectra data samples.   
     
     
         4 . The computer-implemented method of  claim 1 , further comprising:
 using a squared chord distance to determine similarity between spectra data samples.   
     
     
         5 . The computer-implemented method of  claim 1 , further comprising:
 the method for dynamic indexing for a plurality of objects having an association between first and second objects and displaying the association as one of a line and a curved line joining the first and second objects.   
     
     
         6 . The computer-implemented method of  claim 1  comprising displaying the target object at a location central to the display, the target object displayed as a symbol and linked by one of a line and a curved line to another object, one of the lines forming a part of a pathway having a similarity value, a speculation level and a model-based rank, the rank being an ordinal value balancing the similarity with the speculation level. 
     
     
         7 . The computer-implemented method of  claim 1  further comprising recursively applying another test for another node to define a second indexing surface in said three dimensional subspace for separating the spectral data into at least three clusters of spectral data and displaying the first indexing surface and the second indexing surface and first, second and third clusters of spectral data. 
     
     
         8 . Computer apparatus for the dynamic indexing of spectral data, the spectral data being of a spectral database tree having a root node, the computer apparatus comprising search manager apparatus, the search manager apparatus comprising a processor, memory associated with the processor and an output device, the computer apparatus comprising
 a memory for storing the spectral data of the spectral database, the spectral data being represented by a attribute vector having a plurality of M dimensions, where M is at least three;   the processor for defining a root node of the spectral database tree and a test for the root node responsive to a query received at the processor search manager apparatus, the query 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;   the processor for applying a test for another node of the spectral database tree, the processor outputting reference evidence data responsive to the query via the output device, the query for one of a location, a manufacturer and an event related to the query object, the query object being a target object.   
     
     
         9 . The computer apparatus of  claim 8 , the processor using Euclidean distance to determine similarity between spectral data samples. 
     
     
         10 . The computer apparatus of  claim 8 , the processor using a cosine distance metric to determine similarity between spectral data samples. 
     
     
         11 . The computer apparatus of  claim 8 , the output device for outputting an association between first and second objects as one of a line and a curved line joining the first and second objects. 
     
     
         12 . The computer apparatus of  claim 8 , the output device for displaying the target object at a location central to a display as a symbol and the target object being linked by one of a line and a curved line to another object, the one of a line and a curved line forming a part of a pathway having a similarity value, a speculation level and a model-based rank calculated by the processor, the rank being an ordinal value balancing the similarity with the speculation level. 
     
     
         13 . The computer apparatus of  claim 12 , the display output device outputting the display further comprising a plurality of circles, a first circle comprising the first and second objects linked by the one of the line and the curved line, the second object further being linked to a third object of a second circle. 
     
     
         14 . A computer-implemented method of generating an evidence tree of spectral data, the method implemented on processor search manager apparatus comprising a processor and memory, the method using a model comprising:
 creating a display showing a pathway of linkages between a target object comprising a symbol and a second object and a third object, the target object, the second object and the third object being different from one another, the target object linked to the second object and the second object linked to the third object, each link represented by one of a curved line and a line, the pathway comprising a similarity value, a speculation level and a model-based rank, each of the similarity value, the speculation value and the model-based rank being stored in the memory for the pathway, the rank being an ordinal value for balancing the similarity with the speculation level; and   the dynamic indexing for predicting a property of the target object using the model, the model comprising a property of the third object.   
     
     
         15 . The computer-implemented method of  claim 14 , the display further comprising a plurality of circles, a first circle comprising the first target object and second object linked by one of the line and the curved line, the second object further being linked to a third object of a second circle. 
     
     
         16 . The computer-implemented method of  claim 15 , the display further comprising a third circle having a fourth object, the second and third circles being displayed within a fourth circle concentric with the first circle. 
     
     
         17 . The computer-implemented method of  claim 14  further comprising displaying a forensic measurement as a rectangular symbol, the forensic measurement comprising one of a frequency spectral signature and a thermal signature associated with a target object. 
     
     
         18 . The computer-implemented method of  claim 15  further comprising displaying the first, second and third objects as three circular graphs. 
     
     
         19 . The computer-implemented method of  claim 14 , further comprising displaying an outer concentric circle for a displayed forensic measurement comprising a predicted property, the predicted property comprising a one of a frequency spectral signature and a thermal signature associated with the target object. 
     
     
         20 . The computer-implemented method of  claim 14 , further comprising displaying an evidence tree as a circular graph responsive to an input via an evidence interface of attribute data of a target object, the circular graph representing the target object as a symbol, the target object located at the center of a plurality of concentric circles, a first concentric circle comprising attribute data of the target object, a second, next concentric circle having an object having attribute data located as a symbol on the second, next concentric circle.