Nuclear detection via a system of widely distributed low cost detectors having data including gamma intensities, time stamps and geo-positions

• US 7,994,926 B2
• Filed: 07/07/2008
• Issued: 08/09/2011
• Est. Priority Date: 07/05/2007
• Status: Active Grant

• Current Assignee: Purdue Research Foundation (Purdue University)
• Original Assignee: Purdue Research Foundation (Purdue University)
• Inventors: Longman, Andrew, Fischbach, Ephraim, Jenkins, Jere H.
• Primary Examiner(s): HOFSASS, JEFFERY A
• Application Number:
  US12/168,669
• Publication Number:
  US 20090012745A1
• Time in Patent Office: 1,128 Days
• Field of Search: 340/600, 340/500, 340/539.1, 250/336.1
• US Class Current: 340/600
• CPC Class Codes: G01T 7/00 Details of radiation-measuring instruments

Abstract

A radiation detection system includes many receivers to continuously receive radiation emission data from at least some of a sufficient density of dispersed detectors capable of communicating geo-positions and photon emission counts over a network; the data includes gamma intensities, time stamps, and geo-positions. A processor builds digital image data of the received radiation data for a geographic area by treating gamma-ray proton data from each dispersed detector as a pixel in a low-light image. The processor continuously executes a plurality of statistical computational analyses on the digital image data to separate detected radiation signals from random, undesired signal noise, and known signal noise or sources. The statistical computational analyses include match-filter and/or other convolution techniques. An interface reports to a user when the computational analyses result in detection of a radiation signal and reports a location of one or more of the dispersed detectors that contribute to the detection.

Claims

1. A radiation detection system comprising:

a receiver configured to continuously receive radiation emission data from at least some of a sufficient density of dispersed detectors capable of communicating geo-positions and photon emission counts over a wireless network, wherein the radiation emission data includes gamma intensities, time stamps, and geo-positions affiliated therewith reported in real time;

a processor coupled with the receiver and configured to build digital image data of the received radiation emission data for a geographic area by treating emission data from each dispersed detector as an image pixel in a low-light image;

a database coupled with the processor, to store the digital image data;

wherein the processor is configured to continuously execute a plurality of statistical computational analyses on the digital image data to separate detected radiation signals from random, undesired signal noise, wherein the plurality of statistical computational analyses comprise match-filter, other convolution techniques, or a combination thereof; and

an interface configured to report to a user when the plurality of statistical computational analyses result in detection of a radiation signal and to report a location of one or more of the dispersed detectors that contribute to the detection.

2. The system of claim 1, wherein the time stamps are generated from network time protocol (NTP) built into a cellular network that syncs local time of a mobile device clock-calendar with an atomic clock-linked, standards-traceable, Internet Time Service.

3. The system of claim 1, wherein the processor is configured to:

combine the digital image data with previously-saved image data of a matching geographic area to form time-updated digital image data for the geographic area; and

save the updated digital image data in the database.

4. The system of claim 3, further comprising:

a plurality of image processors; and

distributed software code executable by the processor to assign radiation emission data from each dispersed detector to one or more of the plurality of image processors.

5. The system of claim 3, wherein the other convolution techniques comprise image processing and noise significance techniques used in gamma-ray astronomy or charge-coupled device (CCD) cameras, including dark-current pixel significance techniques.

6. The system of claim 3, wherein the plurality of statistical computational analyses comprise:

a pixel significance technique; and

/or a two-dimensional convolution technique that applies, in part, a pixel significance technique.

7. The system of claim 3, wherein the processor is configured to place the digital image data in a form of geometric raster images, wherein each pixel corresponds to photon count within the received radiation emission data.

8. The system of claim 7, wherein the plurality of statistical computation analyses comprises applying physical randomness metrics to the digital image data.

9. The system of claim 8, wherein the physical randomness metrics comprises geometric random inter product (GRIP), wherein the processor is configured to:

calculate a GRIP number for known geographic areas and for tracks through the known geographic areas;

compare each GRIP number with a plurality of pre-calculated, expected GRIP values for the same geographic areas and tracks;

assign discrepancy numbers and confidence levels to each GRIP number based on the comparison;

compare discrepancy numbers and confidence levels to a plurality of acceptable metrics levels; and

begin an alert sequence through the interface in response to, and in correlation with, an unacceptable metric level in discrepancy numbers and confidence levels;

wherein the GRIP number and values, discrepancy numbers, and confidence levels are saved in the database for later comparison during subsequent statistical analyses of a matching geographic area.

10. The system of claim 7, wherein the plurality of statistical computational analyses comprises a gradient analysis, wherein the processor is configured to:

find a maximum two-dimensional or three-dimensional spatial rate of change between a plurality of time-wise successive images retrieved from the database;

signal gradient directions with spatial rates of change by assigning intensity and confidence level numbers to the pixels within the digital image data that indicate to what extent a concentrated source of detected radiation exists, and where it is located; and

alert the user through the interface in response to the intensity and confidence levels meeting a certain combination of intensity and confidence levels that indicate to the user a potentially dangerous concentration of radioactive material.

11. The system of claim 7, wherein the plurality of statistical computation analyses comprises algebraic analysis, wherein the processor is configured to:

identify any nearest neighbor groupings of the dispersed detectors;

select a plurality of dispersed detectors from each grouping that provide gamma intensities above a determined noise threshold, wherein the selected plurality of dispersed detectors from each grouping comprises a cluster;

calculates algebraic circular intersections, where the algebraic circular intersection indicate a source location or a point of origin location, where the source location or point of origin location are based upon the signal intensities observed in each member detector of the cluster;

assign weights to each of the algebraic circular intersections by treating the groupings of clusters statistically and spatially, and applying GRIP or other distribution measuring algorithms thereto;

assign intensity and confidence levels to each cluster; and

alert the user through the interface in response to the intensity and confidence levels being a combination that calls for importance investigation of a potentially dangerous concentration of radioactive material.

12. The system of claim 1, wherein the dispersed detectors comprise cellular phones, personal digital assistants (PDA'"'"'s), pagers, laptops, Bluetooth devices, global positioning system (GPS) navigation devices, mobile devices, automobiles, fixed-position detectors, or a combination thereof.

13. The system of claim 12, wherein some of the dispersed detectors comprise gamma detectors that include at least one of a Geiger-Mü

ller detector tube, a silicon wafer detector, a scintillator detector, or a combination thereof.

14. A distributed radiation-detection network comprising:

a plurality of distributed recipient servers coupled with a network to continuously receive radiation emission data from at least some of a sufficient density of dispersed detectors capable of communicating geo-positions and photon emission counts over the network, wherein the radiation emission data includes gamma intensities, time stamps, and geo-positions affiliated therewith provided in real time;

a statistics server coupled with at least some of the recipient servers, the statistics server having a processor configured to build digital image data of the received radiation emission data for a geographic area by treating emission data from each dispersed detector as an image pixel in a low-light image of the geographic area, the processor configured to continuously execute a plurality of statistical computational analyses on the digital image data to separate detected radiation signals from historically normal levels of noise, wherein the plurality of statistical computational analyses comprise a pixel significance technique, a two-dimensional convolution technique that applies, in part, a pixel significance technique, or a combination thereof;

a database coupled with the statistics server to continuously store updated digital image data as it is processed by the plurality of statistical computational analyses; and

a mapping server coupled with the statistics server and configured to continuously receive the processed digital image data from the statistics server, the mapping server having a graphical user interface to overlay the processed digital image data on top of a geographical-matching map for viewing by a first responder, wherein a statistically significant new source of radiation within the geographic area is displayed to the first responder in real time.

15. The network of claim 14, wherein the time stamps are generated from network time protocol (NTP) built into a cellular network that syncs local time of a mobile device clock-calendar with an atomic clock-linked, standards-traceable, Internet Time Service.

16. The network of claim 14, wherein the statistically significant new source of radiation comprises one or more of bomb locations, intensities, percent confidence metrics, and a combination thereof.

17. The network of claim 14, wherein the plurality of statistical computational analyses comprise convolution techniques, including image processing and noise significance techniques selected from a group consisting of match-filter and dark-current pixel significance techniques.

18. The network of claim 14, wherein known nuclear sources that are on the move within the geographical area each carry a detection node traceable by the plurality of recipient servers, wherein the processor is configured to remove the known nuclear sources from the digital image data.

19. The network of claim 14, wherein the statistics server comprises:

a plurality of image processors; and

distributed software code executable by the processor to assign radiation emission data from each dispersed detector to one or more of the plurality of image processors.

20. The network of claim 14, wherein the processor is configured to place the digital image data in a form of geometric raster images, wherein each pixel corresponds to photon frequency within the received radiation emission data such that the digital image data includes multiple low-light images captured as layers of increasing photon energy.

21. The network of claim 20, wherein the plurality of statistical computation analyses comprises applying physical randomness metrics to the digital image data.

22. The network of claim 21, wherein the physical randomness metrics comprises geometric random inter product (GRIP), wherein the processor is configured to:

calculate a GRIP number for known geographic areas and for tracks through the known geographic areas;

compare each GRIP number with a plurality of pre-calculated, expected GRIP values for the same geographic areas and tracks;

assign discrepancy numbers and confidence levels to each GRIP number based on the comparison;

compare discrepancy numbers and confidence levels to a plurality of acceptable metrics levels; and

begin an alert sequence through the interface in response to, and in correlation with, an unacceptable metric level in discrepancy numbers and confidence levels;

wherein the GRIP number and values, discrepancy numbers, and confidence levels are saved in the database for later comparison during subsequent statistical analyses of a matching geographic area.

23. The network of claim 20, wherein the plurality of statistical computational analyses comprises a gradient analysis, wherein the processor is configured to:

find a maximum two-dimensional or three-dimensional spatial rate of change between a plurality of time-wise successive images retrieved from the database;

signal gradient directions with spatial rates of change by assigning intensity and confidence level numbers to the pixels within the digital image data that indicate to what extent a concentrated source of detected radiation exists, and where it is located; and

alert the user through the interface in response to the intensity and confidence levels meeting a certain combination of intensity and confidence levels that indicate to the user a potentially dangerous concentration of radioactive material.

24. The network of claim 20, wherein the plurality of statistical computation analyses comprises algebraic analysis, wherein the processor is configured to:

identify any nearest neighbor groupings of the dispersed detectors;

select a plurality of dispersed detectors from each grouping that provide gamma intensities above a determined noise threshold, wherein the selected plurality of dispersed detectors from each grouping comprises a cluster;

calculate a circular intersection for each cluster;

assign weights to each circular intersection by treating the groupings of clusters statistically and spatially, and applying GRIP or other distribution algorithms thereto;

assign intensity and confidence levels to each cluster; and

alert the user through the interface in response to the intensity and confidence levels being a combination that calls for importance investigation of a potentially dangerous concentration of radioactive material.

25. The network of claim 20, wherein the dispersed detectors comprise cellular phones, personal digital assistants (PDA'"'"'s), pagers, laptops, Bluetooth devices, global positioning system (GPS) navigation devices, mobile devices, automobiles, fixed-position detectors, or a combination thereof.

26. The network of claim 25, wherein some of the dispersed detectors comprise a Geiger-Mü

ller detector tube, a silicon wafer, a scintillator, or a combination thereof.

27. A computer-implemented method for detection of radiation within a geographical area, comprising:

continuously receiving by a receiver radiation emission data from at least some of a sufficient density of dispersed detectors capable of communicating geo-positions and photon emission counts over a wireless network, wherein the radiation emission data includes gamma intensities, time stamps, and geo-positions affiliated therewith reported in real time;

building digital image data for a geographical area by a processor coupled with the receiver, wherein the digital image data is built though combining the radiation emission data from each dispersed detector as an image pixel in a low-light image;

storing the digital image data in a database coupled with the processor;

continuously executing, by the processor, a plurality of statistical computational analyses on the digital image data to separate detected radiation signals from random, undesired signal noise, wherein the plurality of statistical computational analyses comprise match-filter, other convolution techniques, or a combination thereof; and

displaying, through an interface coupled with the processor, a report to a user when the plurality of statistical computational analyses result in detection of a radiation signal, including a location of one or more of the dispersed detectors that contribute to the detection.

28. The method of claim 27, wherein the time stamps are generated from network time protocol (NTP) built into a cellular network that syncs local time of a mobile device clock-calendar with an atomic clock-linked, standards-traceable, Internet Time Service.

29. The method of claim 27, further comprising:

combining the digital image data with previously-saved image data of a matching geographic area to form time-updated digital image data for the geographic area; and

saving the updated digital image data in the database.

30. The method of claim 29, wherein the other convolution techniques comprise image processing and noise significance techniques used in gamma-ray astronomy or charge-coupled device (CCD) cameras, including dark-current pixel significance techniques.

31. The method of claim 29, wherein the plurality of statistical computational analyses comprise:

a pixel significance technique; and

/or a two-dimensional convolution technique that applies, in part, a pixel significance technique.

32. The method of claim 29, further comprising:

placing the digital image data in a form of geometric raster images, wherein each pixel corresponds to photon count within the received radiation emission data.

33. The method of claim 32, wherein the plurality of statistical computation analyses comprises applying physical randomness metrics to the digital image data.

34. The method of claim 33, wherein the physical randomness metrics comprises geometric random inter product (GRIP), the method further comprising:

calculating a GRIP number for known geographic areas and for tracks through the known geographic areas;

comparing each GRIP number with a plurality of pre-calculated, expected GRIP values for the same geographic areas and tracks;

assigning discrepancy numbers and confidence levels to each GRIP number based on the comparison;

comparing discrepancy numbers and confidence levels to a plurality of acceptable metrics levels; and

beginning an alert sequence through the interface in response to, and in correlation with, an unacceptable metric level in discrepancy numbers and confidence levels;

wherein the GRIP number and values, discrepancy numbers, and confidence levels are saved in the database for later comparison during subsequent statistical analyses of a matching geographic area.

35. The method of claim 32, wherein the plurality of statistical computational analyses comprises a gradient analysis, the method further comprising:

finding a maximum two-dimensional or three-dimensional spatial rate of change between a plurality of time-wise successive images retrieved from the database;

signaling gradient directions with spatial rates of change by assigning intensity and confidence level numbers to the pixels within the digital image data that indicate to what extent a concentrated source of detected radiation exists, and where it is located; and

alerting the user through the interface in response to the intensity and confidence levels meeting a certain combination of intensity and confidence levels that indicate to the user a potentially dangerous concentration of radioactive material.

36. The method of claim 32, wherein the plurality of statistical computation analyses comprises algebraic analysis, the method further comprising:

identifying any nearest neighbor groupings of the dispersed detectors;

selecting a plurality of dispersed detectors from each grouping that provide gamma intensities above a determined noise threshold, wherein the selected plurality of dispersed detectors from each grouping comprises a cluster;

calculating a circular intersection for each cluster;

assigning weights to each circular intersection by treating the groupings of clusters statistically and spatially, and applying GRIP or other distribution algorithms thereto,

assigning intensity and confidence levels to each cluster; and

alerting the user through the interface in response to the intensity and confidence levels being a combination that calls for importance investigation of a potentially dangerous concentration of radioactive material.

37. The method of claim 27, wherein the dispersed detectors comprise cellular phones, personal digital assistants (PDA'"'"'s), pagers, laptops, Bluetooth devices, global positioning system (GPS) navigation devices, mobile devices, automobiles, fixed-position detectors, or a combination thereof, and wherein some of the dispersed detectors include a Geiger-Mü

ller detector tube, a silicon wafer, a scintillator, or a combination thereof.

38. A computer-implemented method for detection of radiation within a geographical area, comprising:

continuously receiving radiation emission data by a plurality of distributed, recipient servers coupled with a network from at least some of a sufficient density of dispersed detectors capable of communicating geo-positions and photon emission counts over the network, wherein the radiation emission data includes gamma intensities, time stamps, and geo-positions affiliated therewith;

building digital image data of the received radiation emission data with a processor of a second server that is coupled with at least some of the recipient servers, the digital image data being built for a geographic area by treating emission data from each dispersed detector as an image pixel in a low-light image thereof;

continuously executing, by the processor, a plurality of statistical computational analyses on the digital image data to separate detected radiation signals from historically normal levels of noise, wherein the plurality of statistical computational analyses comprise a pixel significance technique, a two-dimensional convolution technique that applies, in part, a pixel significance technique, or a combination thereof;

continuously combining the digital image data with previously-saved image data of a matching geographic area to form time-updated digital image data for the geographic area;

storing the updated digital image data in a database coupled with the second server; and

continuously mapping, with the second server, the processed digital image data for display through a graphical user interface (GUI) such that it overlays a geographical-matching map for viewing by a first responder, wherein a statistically significant new source of radiation is displayed to the first responder in real time.

39. The method of claim 38, wherein the time stamps are generated from network time protocol (NTP) built into a cellular network that syncs local time of a mobile device clock-calendar with an atomic clock-linked, standards-traceable, Internet Time Service.

40. The method of claim 38, wherein the statistically significant new source of radiation comprises one or more of bomb locations, intensities, percent confidence metrics, and a combination thereof.

41. The method of claim 38, wherein the plurality of statistical computational analyses comprise convolution techniques, including image processing and noise significance techniques selected from a group consisting of a match-filter and dark-current pixel significance techniques.

42. The method of claim 38, further comprising:

including a detection node traceable by the plurality of recipient servers with known nuclear sources that are on the move within the geographical area; and

removing the known nuclear sources from the digital image data.

43. The method of claim 38, wherein the plurality of statistical computation analyses comprises applying physical randomness metrics to the digital image data.

44. The method of claim 43, wherein the physical randomness metrics comprises geometric random inter product (GRIP), the method further comprising:

calculating a GRIP number for known geographic areas and for tracks through the known geographic areas;

comparing each GRIP number with a plurality of pre-calculated, expected GRIP values for the same geographic areas and tracks;

assigning discrepancy numbers and confidence levels to each GRIP number based on the comparison;

comparing discrepancy numbers and confidence levels to a plurality of acceptable metrics levels; and

beginning an alert sequence through the interface in response to, and in correlation with, an unacceptable metric level in discrepancy numbers and confidence levels;

wherein the GRIP number and values, discrepancy numbers, and confidence levels are saved in the database for later comparison during subsequent statistical analyses of a matching geographic area.

45. The method of claim 38, wherein the plurality of statistical computational analyses comprises a gradient analysis, the method further comprising:

finding a maximum two-dimensional or three-dimensional spatial rate of change between a plurality of time-wise successive images retrieved from the database;

signaling gradient directions with spatial rates of change by assigning intensity and confidence level numbers to the pixels within the digital image data that indicate to what extent a concentrated source of detected radiation exists, and where it is located; and

alerting the user through the interface in response to the intensity and confidence levels meeting a certain combination of intensity and confidence levels that indicate to the user a potentially dangerous concentration of radioactive material.

46. The method of claim 38, wherein the plurality of statistical computation analyses comprises algebraic analysis, the method further comprising:

identifying any nearest neighbor groupings of the dispersed detectors;

selecting a plurality of dispersed detectors from each grouping that provide gamma intensities above a determined noise threshold, wherein the selected plurality of dispersed detectors from each grouping comprises a cluster;

calculating a circular intersection for each cluster;

assigning weights to each circular intersection by treating the groupings of clusters statistically and spatially, and applying GRIP or other distribution algorithms thereto;

assigning intensity and confidence levels to each cluster; and

alerting the user through the interface in response to the intensity and confidence levels being a combination that calls for importance investigation of a potentially dangerous concentration of radioactive material.

47. The method of claim 38, wherein the dispersed detectors comprise cellular phones, personal digital assistants (PDA'"'"'s), pagers, laptops, Bluetooth devices, global positioning system (GPS) navigation devices, mobile devices, automobiles, fixed-position detectors, or a combination thereof, and wherein some of the dispersed detectors include a Geiger-Mü

ller detector tube, a silicon wafer, a scintillator, or a combination thereof.

48. A machine-readable storage medium comprising programming code operable to be executed by a processor, the programming code comprising instructions operable to perform a method comprising:

continuously receiving by a receiver radiation emission data from at least some of a sufficient density of dispersed detectors capable of communicating geo-positions and photon emission counts over a wireless network, wherein the radiation emission data includes gamma intensities, time stamps, and geo-positions affiliated therewith reported in real time;

building digital image data for a geographical area by a processor coupled with the receiver, wherein the digital image data is built though combining the radiation emission data from each dispersed detector as an image pixel in a low-light image;

storing the digital image data in a database coupled with the processor;

continuously executing, by the processor, a plurality of statistical computational analyses on the digital image data to separate detected radiation signals from random, undesired signal noise, wherein the plurality of statistical computational analyses comprise match-filter, other convolution techniques, or a combination thereof; and

displaying, through an interface coupled with the processor, a report to a user when the plurality of statistical computational analyses result in detection of a radiation signal, including a location of one or more of the dispersed detectors that contribute to the detection.

49. A machine-readable storage medium comprising programming code operable to be executed by a processor, the programming code comprising instructions operable to perform a method comprising:

continuously receiving radiation emission data by a plurality of distributed, recipient servers coupled with a network from at least some of a sufficient density of dispersed detectors capable of communicating geo-positions and photon emission counts over the network, wherein the radiation emission data includes gamma intensities, time stamps, and geo-positions affiliated therewith;

building digital image data of the received radiation emission data with a processor of a second server that is coupled with at least some of the recipient servers, the digital image data being built for a geographic area by treating emission data from each dispersed detector as an image pixel in a low-light image thereof;

continuously executing, by the processor, a plurality of statistical computational analyses on the digital image data to separate detected radiation signals from historically normal levels of noise, wherein the plurality of statistical computational analyses comprise a pixel significance technique, a two-dimensional convolution technique that applies, in part, a pixel significance technique, or a combination thereof;

continuously combining the digital image data with previously-saved image data of a matching geographic area to form time-updated digital image data for the geographic area;

storing the updated digital image data in a database coupled with the second server; and

continuously mapping, with the second server, the processed digital image data for display through a graphical user interface (GUI) such that it overlays a geographical-matching map for viewing by a first responder, wherein a statistically significant new source of radiation is displayed to the first responder in real time.