System for generating thermographic images using thermographic signal reconstruction

• US 6,751,342 B2
• Filed: 04/15/2002
• Issued: 06/15/2004
• Est. Priority Date: 12/02/1999
• Status: Expired due to Term

• Current Assignee: Thermal Wave Imaging, Inc.
• Original Assignee: Thermal Wave Imaging, Inc.
• Inventors: Shepard, Steven M.
• Primary Examiner(s): AHMED, SAMIR ANWAR
• Application Number:
  US10/123,726
• Publication Number:
  US 20020172410A1
• Time in Patent Office: 792 Days
• Field of Search: 382/141, 382/143, 382/149, 382/152, 348/86, 348/125, 700/95, 700/212, 250/330, 250/332, 250/358.1, 250/341.6, 250/339.02, 250/339.14, 250/341.1, 356/51, 374/5, 374/10, 374/120, 374/121, 374/124, 209/577
• US Class Current: 382/141
• CPC Class Codes: G01N 25/72 Investigating presence of flaws
• CPC Class Codes: G06T 1/0007 Image acquisition
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/10048 Infrared image
• CPC Class Codes: G06T 2207/30148 Semiconductor; IC; Wafer
• CPC Class Codes: G06T 2207/30164 Workpiece; Machine component
• CPC Class Codes: G06T 7/001 using an image reference approach
• CPC Class Codes: G06T 7/40 Analysis of texture depth or shape recovery from texture G06T7/529
• CPC Class Codes: G06T 7/42 using transform domain methods

Abstract

A method and system for non-destructive, reference-free thermographic detection of sub-surface defects uses an infrared camera to capture multiple, spatially different images of a sample that has been heated and allow to cool to equilibrium temperature. The temperature-time data obtained for each pixel in each image is converted into the logarithmic domain and a least squares fit is conducted on the data to generate a polynomial expression corresponding to the temperature-time data for a given pixel. This polynomial expression can be transformed into the original time domain to obtain temperature-time data with improved signal-to-noise characteristics.

Claims

1. A system for determining a time response to a monotonically changing characteristic of a sample, comprising:

a camera that obtains at least two different images of the sample over time, each image having a plurality of pixels, each pixel having an amplitude corresponding to the monotonically changing characteristic of the sample;

a controller coupled to at least one of the sample or the camera for moving the camera or the sample spatially to obtain said at least two different images of the sample; and

a processor that receives the plurality of images and generates a data array for each of the spatially different images, the data array corresponding to a logarithm of the pixel amplitude at a given time and a logarithm of the given time,

wherein the processor fits a polynomial representing a reconstructed function to at least a portion of the data array, the polynomial having at least one polynomial coefficient, such that each pixel in said different images is represented by a coefficient array containing said at least one polynomial coefficient, and

wherein the processor generates from coefficient arrays at least two reconstructed images corresponding to said at least two different images.

2. The system of claim 1, wherein the two different images are spatially different images separated from one another along a pre-determined inspection route and wherein said processor forms an image mosaic from the reconstructed images.

3. The system of claim 2, wherein the processor includes an acquisition module that generates time and date for information stamp data for each of the spatially different images based on each image'"'"'s position on the inspection route, and wherein the stamp data is used to position each of said reconstructed images in the image mosaic.

4. The system of claim 3, wherein the stamp data is at least one of time and data information or position data.

5. The system of claim 1, wherein the processor generates the coefficient array for each pixel in each of said different images as each image is obtained by the camera.

6. The system of claim 1, wherein the processor obtains said at least two of different images before generating the coefficient array for each of the different images.

7. The system of claim 1, wherein the logarithm of the pixel amplitude or the logarithm of the given time includes applying a natural logarithm or scaling using a T⁻

2 factor where T=time.

8. The system of claim 1, wherein the processor calculates a first derivative of the reconstructed function for a specified time value.

9. The system of claim 8, wherein the processor determines whether the first derivative of the reconstructed function for each pixel over time deviates from −

0.5 by a predetermined threshold.

10. The system of claim 8, wherein the processor calculates a second derivative of the reconstructed function for a specified time value.

11. The system of claim 10, wherein the processor determines whether the second derivative of the reconstructed function for each pixel has a zero crossing, wherein the presence of the zero crossing indicates a defective pixel and the absence of the zero crossing indicates a defect-free pixel.

12. The system of claim 11, wherein the processor converts the logarithm of the earliest zero crossing time of the second derivative of the reconstructed function into a linear time domain for defective pixels, assigning a reserved value for defect-free pixels, and using the reserved value and the converted zero crossing time for each pixel to generate a defect map array that uses a first color value for defective pixels and a second color value for defect-free pixels.

13. The system of claim 1, wherein the processor generates a reference polynomial and subtracts the reconstructed function from the reference polynomial to generate a contrast curve, wherein a peak in the contrast curve indicates the presence of a subsurface defect.

14. The system of claim 1, wherein the processor fits a first polynomial to a first portion of the data array and a second polynomial to a second portion of the data array to generate the reconstructed function.

15. The system of claim 1, wherein the processor divides the data array into an early region, an intermediate region, and a late region and fits a first polynomial to the early region, a second polynomial to the intermediate region, and a third polynomial to the late region to generate the reconstructed function.

16. The system of claim 1, further comprising a memory that stores said coefficient array.

17. The system of claim 16, wherein the coefficient array is stored in the memory as a file having a size that is independent of the number of images obtained by the camera.

18. A method for determining a time response to a monotonic change in a characteristic of a sample, comprising the acts of:

obtaining a plurality of different images of the sample over time, each different image having a plurality of pixels each having an amplitude corresponding to the monotonically changing characteristic of the sample;

generating a data array for each of the plurality of pixels, the data array corresponding to the logarithm of the pixel amplitude at a given time and the logarithm of the given time;

fitting a polynomial representing a reconstructed function to the data array associated with at least a portion of the plurality of pixels, the polynomial having at least two polynomial coefficients, such that each pixel in said portion of plurality of pixels is represented by a coefficient array containing said at least two polynomial coefficients;

generating a plurality of reconstructed images corresponding to the plurality of different images.

19. The method of claim 18, wherein the obtaining different images step includes the step of obtaining spatially different images along a pre-determined inspection route.

20. The method of claim 18, further comprising the act of generating stamp data for each of the spatially different images based on each image'"'"'s position on the inspection route, wherein the forming act uses the stamp data to position the reconstructed images in the image mosaic.

21. The system of claim 20, wherein the stamp data is at least one of time and data information or position data.

22. The method of claim 18, wherein the fitting act generates the coefficient array for each pixel in each of said different images as each image is obtained.

23. The method of claim 18, wherein the obtaining act obtains a plurality of the different images before the fitting act generates the coefficient array for each of the different images.

24. The method of claim 18, further comprising the act of calculating an instantaneous first derivative of the reconstructed function for a specified time value.

25. The method of claim 24, further comprising the act of determining whether the first derivative of the reconstructed function for each pixel over time deviates from a slope of −

0.5.

26. The method of claim 24, further comprising the act of calculating an instantaneous second derivative of the reconstructed function for a specified time value.

27. The method of claim 26, further comprising the act of determining whether the second derivative of the reconstructed function for each pixel has a zero crossing, wherein the presence of the zero crossing indicates a defective pixel and the absence of the zero crossing indicates a defect-free pixel.

28. The method of claim 27, further comprising the act of converting the logarithm of the zero crossing time of the second derivative of the reconstructed function into a linear time domain for defective pixels, assigning a reserved value for defect-free pixels, and using the reserved value and the converted zero crossing time for each pixel to generate a defect map array that uses a first color value for defective pixels and a second color value for defect-free pixels.

29. The method of claim 18, further comprising the act of generating a reference polynomial and subtracts the reconstructed function associated with each pixel in said portion of plurality of pixels from the reference polynomial to generate a contrast curve, wherein a peak in the contrast curve indicates the presence of a subsurface defect.

30. The method of claim 18, further comprising the step of fitting a first polynomial to a first portion of the data array and a second polynomial to a second portion of the data array to generate the reconstructed function.

31. The method of claim 18, further comprising the act of dividing the data array into an early region, an intermediate region, and a late region and fits a first polynomial to the early region, a second polynomial to the intermediate region, and a third polynomial to the late region to generate the reconstructed function.

32. The method of claim 18, further comprising the act of storing the coefficient array in a memory.

33. The method of claim 32, wherein the coefficient array is stored in the memory in a file having a size that is independent of the number of images obtained by the camera.