Systems and methods for detecting defects in coatings utilizing color-based thermal mismatch

• US 8,204,294 B2
• Filed: 11/25/2009
• Issued: 06/19/2012
• Est. Priority Date: 11/25/2009
• Status: Active Grant

• Current Assignee: Toyota Motor Engineering & Manufacturing North America Incorporated (Toyota Motor Corporation), University Of Kentucky Research Foundation (University of Kentucky)
• Original Assignee: Toyota Motor Engineering & Manufacturing North America Incorporated (Toyota Motor Corporation), University Of Kentucky Research Foundation (University of Kentucky)
• Inventors: Alloo, Richard, Saito, Kozo, Gharaibeh, Belal, Chuah, Keng, Akafuah, Nelson, Salaimeh, Ahmad
• Primary Examiner(s): Punnoose, Roy M
• Application Number:
  US12/626,381
• Publication Number:
  US 20110123093A1
• Time in Patent Office: 937 Days
• Field of Search: 382/141, 382/152, 382/164, 356/402
• US Class Current: 382/141
• CPC Class Codes: G01N 25/72 Investigating presence of flaws

Abstract

A method of analyzing a thermal image of a coated substrate to determine the presence of defects includes determining a defect temperature range based on a color of the coated substrate and the maximum temperature of the coated substrate in the thermal image. Thereafter, the thermal image is processed by determining a signal value of a pixel of interest based on a temperature of the pixel of interest, temperatures of pixels in a kernel of pixels surrounding the pixel of interest, and the color of the coated substrate. The signal value of the pixel of interest is then compared to the lower temperature threshold of the defect temperature range, wherein the pixel of interest is a defect location when the signal value of the pixel of interest is greater than or equal to the lower temperature threshold.

Claims

1. A method of analyzing a thermal image of a coated substrate to determine the presence of defects, the method comprising:

determining a color of the coated substrate;

determining a maximum temperature range of a coated substrate from a thermal image;

determining a defect temperature range of the coated substrate based on the color and the maximum temperature;

identifying a pixel of interest; and

processing the thermal image by;

a) determining a signal value of the pixel of interest based on;

a temperature of the pixel of interest;

temperatures of pixels in a kernel of pixels surrounding the pixel of interest; and

the color of the coated substrate; and

b) comparing the signal value to a lower temperature threshold and an upper temperature threshold of the defect temperature range, wherein the pixel of interest is a defect location when the signal value of the pixel of interest is greater than or equal to the lower temperature threshold and less than the upper temperature threshold.

2. The method of claim 1 wherein processing the thermal image further comprises:

c) setting the signal value of the pixel of interest to a defect value when the pixel of interest is a defect; and

d) setting the signal value of the pixel of interest to a non-defect value when the pixel of interest is not a defect.

3. The method of claim 2 further comprising repeating steps a), b), c), and d) for each pixel in the thermal image.

4. The method of claim 3 further comprising presenting a processed thermal image comprising pixels having defect values and pixels having non-defect values.

5. A method for detecting defects in a coated substrate comprising:

determining a color of the coated substrate;

manipulating a temperature of the coated substrate;

acquiring a thermal image of the coated substrate;

determining a maximum temperature of the coated substrate from the thermal image;

determining a defect temperature range based on the color of the coated substrate and the maximum temperature of the coated substrate; and

processing the thermal image by;

a) determining a signal value of a pixel of interest based on;

a temperature of the pixel of interest;

temperatures of pixels in a kernel of pixels surrounding the pixel of interest; and

the color of the coated substrate; and

b) comparing the signal value of the pixel of interest to a lower temperature threshold and an upper temperature threshold of the defect temperature range, wherein the pixel of interest is a defect location when the signal value of the pixel of interest is greater than or equal to the lower temperature threshold and less than the upper temperature threshold.

6. The method of claim 5 wherein the signal value of the pixel of interest is determined by:

determining a mean temperature of the kernel of pixels;

determining a standard deviation of the temperature of the kernel of pixels from the mean temperature;

determining a temperature difference between the temperature of the pixel of interest and the mean temperature; and

comparing the temperature difference to a product of the standard deviation and a standard deviation factor, wherein;

when the temperature difference is greater than or equal to the product of the standard deviation and the standard deviation factor, the signal value of the pixel of interest is equal to the product of the temperature of the pixel of interest and a highlight factor;

when the temperature difference is less than the product of the standard deviation and the standard deviation factor, the signal value of the pixel of interest is equal to the product of the temperature of the pixel of interest and a dim factor; and

the standard deviation factor, the highlight factor and the dim factor are dependent on the color of the coated substrate.

7. The method of claim 5 wherein a size of the kernel of pixels surrounding the pixel of interest is dependent on the color of the coated substrate.

8. The method of claim 5 wherein processing the thermal image further comprises:

c) setting the signal value of the pixel of interest to a defect value when the pixel of interest is a defect; and

d) setting the signal value of the pixel of interest to a non-defect value when the pixel of interest is not a defect.

9. The method of claim 5 further comprising repeating steps a), b), c), and d) for each pixel in the thermal image.

10. The method of claim 9 further comprising presenting a processed thermal image comprising pixels having defect values and pixels having non-defect values.

11. The method of claim 10 further comprising filtering the processed thermal image to remove defects having a size below a defect size threshold.

12. A defect detection system for detecting a defect in a coated substrate comprising a thermal detector electrically coupled to a controller and at least one temperature manipulation device coupled to the controller, wherein the controller is programmed to:

receive an input indicative of a color of the coated substrate;

manipulate a temperature of the coated substrate with the temperature manipulation device;

acquire a thermal image of the coated substrate with the thermal detector;

determine a maximum temperature of the coated substrate from the thermal image;

determine a defect temperature range based on the color of the coated substrate and the maximum temperature of the coated substrate;

identify a pixel of interest; and

process the thermal image by;

a) determining a signal value of the pixel of interest based on;

a temperature of the pixel of interest;

temperatures of pixels in a kernel of pixels surrounding the pixel of interest; and

the color of the coated substrate; and

b) comparing the signal value of the pixel of interest to a lower temperature threshold and an upper temperature threshold of the defect temperature range, wherein the pixel of interest is a defect location when the signal value of the pixel of interest is greater than or equal to the lower temperature threshold and less than the upper temperature threshold.

13. The defect detection system of claim 12 wherein the defect detection system further comprises an object detector electrically coupled to the controller.

14. The defect detection system of claim 13 wherein the object detector is operable to output a signal to the controller indicative of the color of the coated substrate.

15. The defect detection system of claim 12 wherein the controller is further programmed determine the signal value of the pixel of interest by:

determining a mean temperature of the kernel of pixels;

determining a standard deviation of the temperature of the kernel of pixels from the mean temperature;

determining a temperature difference between the temperature of the pixel of interest and the mean temperature; and

comparing the temperature difference to a product of the standard deviation and a standard deviation factor, wherein;

when the temperature difference is greater than or equal to the product of the standard deviation and the standard deviation factor, the signal value of the pixel of interest is equal to the product of the temperature of the pixel of interest and a highlight factor;

when the temperature difference is less than the product of the standard deviation and the standard deviation factor, the signal value of the pixel of interest is equal to the product of the temperature of the pixel of interest and a dim factor; and

the standard deviation factor, the highlight factor and the dim factor are dependent on the color of the coated substrate.

16. The defect detection system of claim 12 wherein the controller is further programmed to determine a size of the kernel of pixels surrounding the pixel of interest based on the color of the coated substrate.

17. The defect detection system of claim 12 wherein the controller is further programmed to process the thermal image by:

c) setting the signal value of the pixel of interest to a defect value when the pixel of interest is a defect; and

d) setting the signal value of the pixel of interest to a non-defect value when the pixel of interest is not a defect.

18. The defect detection system of claim 17 wherein the controller is further programmed to repeat steps a), b), c), and d) for each pixel in the thermal image.

19. The defect detection system of claim 18 wherein the controller is further programmed to output a processed thermal image comprising pixels having defect values and pixels having non-defect values.

20. The defect detection system of claim 19 wherein the controller is further programmed to filter the processed thermal image to remove defects having a size below a defect size threshold.