Dynamic identification and correction of defective pixels

• US 8,098,304 B2
• Filed: 02/26/2008
• Issued: 01/17/2012
• Est. Priority Date: 07/07/2003
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Zoran Corporation (Qualcomm, Inc.)
• Inventors: Pinto, Victor, Shaposhnik, Diana
• Primary Examiner(s): SELBY, GEVELL V
• Application Number:
  US12/037,699
• Publication Number:
  US 20080143856A1
• Time in Patent Office: 1,421 Days
• Field of Search: 348/246, 348/247, 348/251, 348/245, 382/254
• US Class Current: 348/246
• CPC Class Codes: H04N 17/002 for television cameras
• CPC Class Codes: H04N 23/843 Demosaicing, e.g. interpolating colour pixel values
• CPC Class Codes: H04N 25/134 based on three different wavelength filter elements
• CPC Class Codes: H04N 25/683 by defect estimation performed on the scene signal, e.g. real time or on the fly detection

Abstract

Rather than storing a table within a digital camera or other optical device of defective pixels within its image sensor, defective pixels are dynamically identified by comparing the output values of individual pixels under test with those of neighboring pixels. If the neighbors are all above or below the pixel under test by a defined threshold amount, the output of the pixel under test is replaced with a value calculated from the outputs of neighboring pixels.

Claims

1. A method of distinguishing high quality elements from defective elements in an array of photo-sensitive elements, the method comprising:

calculating a plurality of difference values for a first element of the array of photo-sensitive elements by comparing the output of the first element to the outputs of a corresponding plurality of elements neighboring the first element;

determining whether the sign values of the plurality of calculated difference values fall within at least a first, second or third group wherein the first group being the sign values are all positive, the second group being the sign values are all negative, and the third group being the sign values are neither all positive nor all negative;

in response to a determination that the sign values of the plurality of difference values fall within the first and second groups comparing each of the plurality of difference values with at least one defect determination threshold, and

in response to a determination that the sign values of the plurality of difference values fall within the third group identifying the first element to be of high quality.

2. The method of claim 1, where each of the elements of the array provide an output sensitive to one of a plurality of colors and each of difference values is calculated between the output of the first element and a respective neighboring element sensitive to the same color as the first element.

3. The method of claim 1, where the elements of the array are arranged into rows and columns and the plurality of elements neighboring the first element includes elements from a different row than that to which the first element belongs.

4. A method of identifying and correcting defective ones of an array of photo-sensitive pixels, comprising:

directing an object field of varying light intensity across the array;

calculating a plurality of difference values for a first element of the array of photo-sensitive elements by comparing the output of the first element to the outputs of a corresponding plurality of elements neighboring the first element;

determining whether the sign values of the plurality of calculated difference values fall within at least a first, second or third group wherein the first group being the sign values are all positive, the second group being the sign values are all negative, and the third group being the sign values are neither all positive nor all negative;

in response to a determination that the sign values of the plurality of difference values fall within the first and second groups, performing a process including;

determining whether one or more of the difference values are in excess of the threshold,

in response to a determination of one or more of the difference values not exceeding the threshold, utilizing the output of the first pixel for data of the object field, and

in response to a determination of all of the difference values exceeding the threshold, calculating a pixel value of the first pixel from at least some of the neighboring pixels and utilizing the calculated pixel value for data of the object field; and

in response to a determination that the sign values of the plurality of difference values fall within the third group, utilizing the output of the first pixel for data of the object field without comparing the difference values with a defect determination threshold.

5. The method of claim 4, wherein said threshold includes either of at least first or second quantities that are different from each other depending upon whether said same sign is positive or negative.

6. The method of claim 4, wherein said threshold includes either of at least first or second quantities that are different from each other depending upon a distance between the first pixel and the neighboring pixel with which the corresponding difference was computed.

7. The method of claim 4, wherein a value of said threshold is dependent upon both (a) whether said same sign is positive or negative and (b) a distance between the first pixel and the neighboring pixel with which the corresponding difference was computed.

8. The method of claim 4, where each of the pixels of the array provide output sensitive to one of a plurality of colors and each of difference values is calculated between the output of the first pixel and a respective neighboring pixel sensitive to the same color as the first pixel.

9. The method of claim 4, where the pixels of the array are arranged into rows and columns and the plurality of pixels neighboring the first pixel includes pixels from a different row than that to which the first pixel belongs.