Measuring Linear Separations in Digital Radiographs

Abstract

Digital pixel data is obtained from radiographic imaging of one or more objects, and corresponds to an imaged area containing a feature to be measured. A data profile for a region around the measured feature is created from the digital pixel data. A reference profile is then created from the data profile. The reference profile represents an expected data profile for a reference condition of the objects, and accounts for the point spread function of the imager. The difference between the data profile and the reference profile is calculated. Based on that difference, the degree by which the actual condition of the objects varies from the reference condition is determined. The calculated difference can be compared to a lookup table mapping previously calculated differences to degrees of variation from the reference condition. The calculated difference can also be used as an input to an experimentally derived formula.

1 . A method, comprising:
 analyzing image data for an object having first and second regions, wherein
 an actual geometry of the first region varies by an unknown amount from an ideal geometry for the first region, and 
 the analysis is based on data derived from the first region ideal geometry and from a point spread function for an imager used to create the image data; and 
   determining at least one aspect of the second region based on the analysis.   
   
   
       2 . The method of  claim 1 , wherein
 the second region includes an interface between the first region and a third region,   the at least one aspect comprises a separation of the first region from the third region, and   determining at least one aspect of the second region comprises determining the separation to a resolution that is finer than a standard deviation of the imager point spread function.   
   
   
       3 . The method of  claim 2 , wherein a density of the first region is substantially dissimilar from a density of the third region. 
   
   
       4 . The method of  claim 1 , wherein
 the image data comprises an actual data profile,   analyzing the image data comprises approximating a location for a portion of the first region based on data derived from multiple hypothetical data profiles for the object, and   each of the multiple hypothetical data profiles assumes the first region has the ideal geometry.   
   
   
       5 . The method of  claim 4 , wherein
 the second region includes an interface between the first region and a third region,   each of the hypothetical data profiles assumes a different separation between the first and third regions.   
   
   
       6 . The method of  claim 4 , wherein
 the second region includes an interface between the first region and a third region,   analyzing the image data further comprises creating a hypothetical reference data profile for a portion of the object that includes portions of the first, second and third regions,   the hypothetical reference data profile assumes the second region is in a reference condition, and   a portion of the hypothetical reference data profile corresponding the second region is created using the imager point spread function.   
   
   
       7 . The method of  claim 1 , further comprising creating the image data based on successive sets of overlapping images. 
   
   
       8 . The method of  claim 1 , wherein the object is an artillery shell, the first region is a steel portion of the shell, and the second region includes an interface between that steel portion and an explosive. 
   
   
       9 . A computer-readable medium having stored thereon data representing sequences of instructions which, when executed by a processor, cause the processor to perform steps comprising:
 analyzing image data for an object having first and second regions, wherein
 an actual geometry of the first region varies by an unknown amount from an ideal geometry for the first region, and 
 the analysis is based on data derived from the first region ideal geometry and from a point spread function for an imager used to create the image data; and 
   determining at least one aspect of the second region based on the analysis.   
   
   
       10 . The computer-readable medium of  claim 9 , wherein
 the second region includes an interface between the first region and a third region,   the at least one aspect comprises a separation of the first region from the third region, and   determining at least one aspect of the second region comprises determining the separation to a resolution that is finer than a standard deviation of the imager point spread function.   
   
   
       11 . The computer-readable medium of  claim 10 , wherein a density of the first region is substantially dissimilar from a density of the third region. 
   
   
       12 . The computer-readable medium of  claim 9 , wherein
 the image data comprises an actual data profile,   analyzing the image data comprises approximating a location for a portion of the first region based on data derived from multiple hypothetical data profiles for the object, and
 each of the multiple hypothetical data profiles assumes the first region has the ideal geometry. 
   
   
   
       13 . The computer-readable medium of  claim 12 , wherein
 the second region includes an interface between the first region and a third region,   each of the hypothetical data profiles assumes a different separation between the first and third regions.   
   
   
       14 . The computer-readable medium of  claim 12 , wherein
 the second region includes an interface between the first region and a third region,   analyzing the image data further comprises creating a hypothetical reference data profile for a portion of the object that includes portions of the first, second and third regions,   the hypothetical reference data profile assumes the second region is in a reference condition, and   a portion of the hypothetical reference data profile corresponding the second region is created using the imager point spread function.   
   
   
       15 . The computer-readable medium of  claim 9 , comprising additional data representing sequences of instructions which, when executed by a processor, cause the processor to perform an additional step comprising creating the image data based on successive sets of overlapping images. 
   
   
       16 . The method of  claim 1 , wherein the object is an artillery shell, the first region is a steel portion of the shell, and the second region includes an interface between that steel portion and an explosive. 
   
   
       17 . A method, comprising:
 receiving image data for an object having an unknown internal dimension, wherein
 the unknown internal dimension is based on at least one reference point, 
 the at least one reference point has a location that varies from an ideal location by an unknown amount, and 
 the unknown internal dimension is of the same order of magnitude as a standard deviation for a point spread function for an imager used to create the image data; and 
 determining the unknown internal dimension from the image data, to a resolution that is finer than the standard deviation of the point spread function, using data derived from the ideal location and from the point spread function. 
   
   
   
       18 . The method of  claim 17 , wherein the unknown internal dimension is a separation between materials of substantially dissimilar densities. 
   
   
       19 . The method of  claim 17 , wherein the step of receiving image data includes receiving data generated by creating successive sets of overlapping images. 
   
   
       20 . A device comprising:
 an imaging array; and   at least one processor receiving image data generated by the array and configured to perform steps that include
 analyzing image data for an object having first and second regions, wherein an actual geometry of the first region varies by an unknown amount from an ideal geometry for the first region, and wherein the analysis is based on data derived from the first region ideal geometry and from a point spread function corresponding to the imaging array, and 
 determining at least one aspect of the second region based on the analysis.