System and methods for determining nonuniformity correction parameters in detector-array imaging

• US 7,294,817 B2
• Filed: 05/05/2005
• Issued: 11/13/2007
• Est. Priority Date: 05/06/2004
• Status: Expired due to Fees

• Current Assignee: Siemens Energy Incorporated (Siemens AG)
• Original Assignee: Siemens Power Generation, Inc.
• Inventors: Ramesh, Visvanathan, Lemieux, Dennis H., Brummel, Hans-Gerd, Voigt, Matthias, Zarzycki, Martin
• Primary Examiner(s): Epps; Georgia
• Assistant Examiner(s): Lee; Patrick J.
• Application Number:
  US11/122,568
• Publication Number:
  US 20050247867A1
• Time in Patent Office: 922 Days
• Field of Search: 250/208.1, 348/164, 348/166, 348/615, 382/275, 341/139, 264/515, 264/517
• US Class Current: 250/208.1
• CPC Class Codes: H04N 25/674 based on the scene itself, e.g. defocusing
• CPC Class Codes: H04N 25/68 applied to defects
• CPC Class Codes: H04N 5/33 Transforming infrared radiation cameras or camera modules for generating image signals from infrared radiation H04N23/20; circuitry of SSIS for transforming infrared radiation into image signals H04N25/20

Abstract

An imaging system (110) for imaging a scene with a detector array (104) having an array of imaging elements is provided. The imaging system (110) includes an image estimation module (202) for generating a plurality estimates of uncorrupted images based upon a plurality of noisy images generated by the detector array (104). The imaging system (110) further includes a parameter determination module (204) for determining one or more nonuniformity correction parameters based upon the estimates of uncorrupted images.

Claims

1. A method for determining nonuniformity correction parameters for a detector array, the method comprising the steps of:

for each element of the detector array, generating a plurality estimated and observed intensity pairs, wherein the estimated intensities for each element is calculated based upon corresponding observed intensities for that element;

generating a plurality of estimates of uncorrupted images, each of the estimates being generated by fitting a parametric patch model to the plurality estimated and observed intensity pairs for a neighborhood of noisy image samples, each uncorrupted image uniquely corresponding to one of a plurality of noisy images; and

determining at least one nonuniformity correction parameter by fitting a parametric curve to the plurality of estimates and the plurality of noisy images corresponding to each of the estimates;

identifying whether one or more elements of the detector array is faulty based upon the fitted parametric model and, if at least one detector is determined to be faulty, suppressing the output of the at least one faulty detector.

2. The method of claim 1, wherein the parametric patch model comprises a second-order model.

3. The method of claim 1, wherein fitting the parametric patch model comprises minimizing a predefined merit function using a least-squares procedure.

4. The method of claim 3, wherein the predefined merit function expresses a mathematical relationship between the plurality of noisy images and a plurality of parametric patch models, the mathematical relationship being defined by a coefficient matrix and an inverse covariance matrix.

5. The method of claim 1, wherein fitting the parametric curve comprises minimizing a predefined merit function using a least-squares procedure.

6. The method of claim 5, wherein the predefined merit function expresses a mathematical relationship between the plurality of estimates and the plurality of noisy images corresponding to each of the estimates, the mathematical relationship being defined by coefficient matrix and an inverse covariance matrix.

7. A method of updating nonuniformity correction parameters for a detector array, the method comprising the steps of:

(a) providing a plurality of noisy image samples;

(b) for each element of the detector array, generating a plurality estimated and observed intensity pairs, wherein the estimated intensities for each element is calculated based upon corresponding observed intensities for that element, and generating a plurality of estimates of uncorrupted images by fitting a curve to the plurality of estimated and observed intensity pairs for the plurality of noisy image samples and an initial set of correction parameters, each of the estimates of uncorrupted images corresponding to a unique one of the plurality of noisy images;

(c) generating a plurality of noise-reduced estimates of uncorrupted images using a set of spatial-neighborhood observations, each of the noise-reduced estimates of uncorrupted images corresponding to a unique one of the estimates of uncorrupted images; and

(d) determining an updated set of correction parameters based upon a predefined mathematical relation between each of the noise-reduced estimates of uncorrupted images and corresponding one of the estimates of uncorrupted images, identifying whether one or more elements of the detector array is faulty based upon the fitted curve, and, if at least one detector is determined to be faulty, suppressing the output of the at least one faulty detector.

8. The method of claim 7, further comprising the step of:

(e) setting the initial set of correction parameters equal to the updated set of correction parameters and repeating steps (a) through (d) to determine a different set of update correction parameters if the updated set of correction parameters and a previously determined updated set of correction parameters does not converge.

9. The method of claim 7, wherein the plurality of correction parameters comprises an offset parameter and a gain parameter.

10. The method of claim 7, wherein the step of generating a plurality of noise-reduced estimates of uncorrupted images comprises computing a weighted average of the set of spatial-neighborhood observations.

11. The method of claim 7, wherein the step of determining an updated set of correction parameters is based upon a linear relationship between each of the noise-reduced estimates of uncorrupted images and corresponding one of the estimates of uncorrupted images.

12. The method of claim 11, wherein the linear relationship is determined using a robust least-squares procedure.

13. The method of claim 7, wherein the step of determining the updated set of correction parameters further comprises determining the updated set of correction parameters based upon a predetermined set of damping coefficients.

14. The method of claim 13, wherein the damping coefficients define a recursive relation between a currently determined set of updated correction parameters and a previously determined set of updated correction parameters.

15. The method of claim 7, further comprising the step of:

(f) determining when the element of the array fails to function accurately.

16. The method of claim 15, further comprising the step of:

(g) suppressing a contribution of the element of the array to the determination of correction parameters if the element of the array is determined to be functioning inaccurately.

17. An imaging system for imaging a scene with a detector array having an array of imaging elements, the system comprising:

an image estimation module for generating a plurality estimates of uncorrupted images based upon a plurality of noisy images generated by the detector array, wherein the image estimation module is configured to generate a plurality estimated and observed intensity pairs for each element of the detector array, the estimated intensities for each element being calculated based upon a corresponding observed intensities for that element, and to generate the estimates of the uncorrupted images by fitting a curve to the plurality of estimated and observed intensity pairs; and

a parameter determination module for determining at least one nonuniformity correction parameter based upon the estimates of uncorrupted images.

18. The imaging system of claim 17, wherein the image estimation module generates the plurality of estimates of uncorrupted images by fitting a parametric patch model into a neighborhood of noisy image samples such that each estimate and uniquely corresponds to one of a plurality of noisy images;

and wherein the parameter determination module determines at least one nonuniformity correction parameter by fitting a parametric curve to the plurality of estimates and the plurality of noisy images corresponding to each of the estimates.

19. The imaging system of claim 17, wherein the image estimation module generates a plurality of estimates of uncorrupted images based upon a plurality of noisy image samples and an initial set of correction parameters, each of the estimates of uncorrupted images corresponding to a unique one of the plurality of noisy images;

and wherein the image estimation module further generates a plurality of noise-reduced estimates of uncorrupted images using a set of spatial-neighborhood observations, each of the noise-reduced estimates of uncorrupted images corresponding to a unique one of the estimates of uncorrupted images.

20. The imaging system of claim 19, wherein the parameter determination module determines an updated set of correction parameters based upon a predefined mathematical relation between each of the noise-reduced estimates of uncorrupted images and corresponding ones of the estimates of uncorrupted images.