METHOD OF SHUTTERLESS NON-UNIFORMITY CORRECTION FOR INFRARED IMAGERS

• US 20190075256A1
• Filed: 08/17/2018
• Published: 03/07/2019
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: DRS Network & Imaging Systems, LLC (Leonardo S.p.A.)
• Original Assignee: DRS Network & Imaging Systems, LLC (Leonardo S.p.A.)
• Inventors: Bailey, John R., Allred, Rustin W.
• Granted Patent: US 10,462,388 B2
• Time in Patent Office: Days
• CPC Class Codes: G06T 2207/10048 Infrared image
• CPC Class Codes: G06T 5/00 Image enhancement or restoration
• CPC Class Codes: G06T 5/77 Retouching; Inpainting; Scratch removal
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• 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

A method of correcting an infrared image including a plurality of pixels arranged in an input image array, a first pixel in the plurality of pixels having a first pixel value and one or more neighbor pixel with one or more neighbor pixel values. The first pixel and the one or more neighbor pixels are associated with an object in the image. The method includes providing a correction array having a plurality of correction pixel values, generating a corrected image array by adding the first pixel value to a correction pixel value in the correction array, and detecting edges in the corrected image array. The method also includes masking the detected edges in the corrected image array, updating the correction array, for each correction pixel value in the correction array and providing an output image array based on the correction array and the input image array.

Claims

1. (canceled)

2. A method of correcting an infrared image, the method comprising:

providing a processor;

receiving the infrared image from a camera, the infrared image comprising a plurality of pixels arranged in an input image array, a first pixel in the plurality of pixels having a first pixel value and one or more neighbor pixels with one or more neighbor pixel values, wherein the first pixel and the one or more neighbor pixels are associated with an object in the infrared image, the one of more neighbor pixels being adjacent to the first pixel in the input image array;

processing the infrared image to generate a processed image;

determining, by the processor, that the processed image is a first frame;

initializing a correction table to an initialized correction table;

initializing a mean image counter to zero;

determining, by the processor, that the camera is moving;

determining that a motion of the camera is greater than or equal to a motion threshold;

updating the correction table using a motion-based algorithm; and

providing an output image based on the updated correction table using the motion-based algorithm and the processed image.

3. The method of claim 2, further comprising:

determining that the motion of the camera is less than the motion threshold;

initializing a mean image to the processed image; and

initializing the mean image counter to zero.

4. The method of claim 3, further comprising:

determining that the mean image counter is greater than a count threshold;

updating the correction table using a motionless algorithm; and

providing the output image based on the updated correction table using the motionless algorithm and the processed image.

5. The method of claim 3, further comprising:

determining that the mean image counter is less than or equal to a count threshold;

providing the output image based on the initialized correction table and the processed image.

6. The method of claim 2, wherein updating the correction table using the motion-based algorithm comprises:

for each correction pixel value, determining that the first pixel value is not an edge;

for each of the one or more neighbor pixel values, determining that a neighbor pixel value is not an edge;

computing a difference between the first pixel value and the one or more neighbor pixel values; and

determining a correction pixel value based on the difference.

7. The method of claim 6, wherein updating the correction table using the motion-based algorithm further comprises:

adding the difference to a previous neighbor correction value to update a neighbor correction value; and

subtracting the difference from a previous correction value to update a correction value.

8. The method of claim 2, wherein updating the correction table using the motion-based algorithm comprises:

for each correction pixel value, determining correction values for the one or more neighbor pixels;

computing a difference between the first pixel value and the one or more neighbor pixel values; and

updating the correction pixel value using the difference.

9. The method of claim 8, wherein updating the correction table using the motion-based algorithm further comprises:

determining that an absolute value of the difference is within a range.

10. The method of claim 9, wherein the range is defined by a lower threshold and an upper threshold that are predetermined values.

11. The method of claim 9, wherein updating the correction table using the motion-based algorithm further comprises:

adding a fraction of the difference to a previous neighbor correction value to update a neighbor correction value; and

subtracting the fraction of the difference from a previous correction value to update a correction value.

12. The method of claim 11, wherein the fraction of the difference comprises dividing the difference by a denominator, wherein the denominator is a predetermined value.

13. The method of claim 2, wherein the motion threshold is defined by a plurality of pixel units.

14. A method of correcting an infrared image, the method comprising:

providing a processor;

receiving the infrared image from a camera, the infrared image comprising a plurality of pixels arranged in an input image array, a first pixel in the plurality of pixels having a first pixel value and one or more neighbor pixels with one or more neighbor pixel values, wherein the first pixel and the one or more neighbor pixels are associated with an object in the infrared image, the one of more neighbor pixels being adjacent to the first pixel in the input image array;

processing the infrared image to generate a processed image;

determining, by the processor, that the processed image is a first frame;

initializing a correction table to an initialized correction table;

initializing a mean image counter to zero;

determining, by the processor, that the camera is not moving;

computing a percentage change in a grayscale value of each pixel relative to a corresponding pixel in a mean image;

marking pixels having a percentage change greater than a grayscale threshold as moving pixels; and

counting a number of moving pixels.

15. The method of claim 14, further comprising:

determining that the number of moving pixels is greater a change threshold;

initializing the mean image to the processed image;

initializing the mean image counter to zero; and

providing an output image based on the initialized correction table and the processed image.

16. The method of claim 14, further comprising:

determining that the number of moving pixels is less than or equal to a change threshold;

initializing the mean image to a sum of the processed image multiplied by a constant and a fraction of a previous mean image; and

incrementing the mean image counter,

wherein the constant is less than unity, and a sum of the constant and the fraction is equal to unity.

17. The method of claim 16, further comprising:

determining that the mean image counter is greater than a count threshold;

updating the correction table using a motionless algorithm; and

providing an output image based on the updated correction table using the motionless algorithm and the processed image.

18. The method of claim 17, wherein updating the correction table using the motionless algorithm comprises:

for each correction pixel value, determining correction values for the one or more neighbor pixels;

computing a difference between the first pixel value and the one or more neighbor pixel values; and

updating the correction pixel value using the difference.

19. The method of claim 18, wherein updating the correction table using the motionless algorithm further comprises:

determining that an absolute value of the difference is within a range.

20. The method of claim 18, wherein updating the correction table using the motionless algorithm further comprises:

adding a fraction of the difference to a previous neighbor correction value to update a neighbor correction value; and

subtracting the fraction of the difference from a previous correction value to update a correction value, wherein the fraction of the difference comprises dividing the difference by a denominator, wherein the denominator is a predetermined value.

21. The method of claim 16, further comprising:

determining that the mean image counter is less than or equal to a count threshold;

providing an output image based on the initialized correction table and the processed image.