Detection of Defective Pixels in an Image Sensor

• US 20110043668A1
• Filed: 08/24/2009
• Published: 02/24/2011
• Est. Priority Date: 08/24/2009
• Status: Active Grant

• Current Assignee: Lifesize, Inc. (Enghouse Systems Limited)
• Original Assignee: Lifesize, Inc. (Enghouse Systems Limited)
• Inventors: Flosdorf, Stephen P., McKinnon, Patrick R., Li, Jingqiang
• Granted Patent: US 8,154,632 B2
• Time in Patent Office: Days
• US Class Current: 348/246
• CPC Class Codes: H04N 25/683 by defect estimation performed on the scene signal, e.g. real time or on the fly detection

Abstract

A system and method for detecting defective pixels in a sensor. A plurality of pixel values of the sensor may be detected. The values may include those of a first pixel and each nearest neighboring pixel to the first pixel. A second pixel may have the highest value of the neighboring pixels. A third pixel may have the next highest value of the neighboring pixels. A first function may be performed on the second pixel value, producing a first output value. A second function may be performed on the third pixel value, producing a second output value. If the first pixel value is higher than the first output value, or, if the first pixel value is higher than the second output value and the second pixel value is higher than the second output value, it may be determined that the first pixel is defective.

Claims

1. A method for detecting hot pixels in a sensor which senses a plurality of pixels, comprising:

detecting a plurality of pixel values, the plurality of pixel values comprising values of a first pixel and each nearest neighboring pixel to the first pixel;

storing the plurality of pixel values in a memory;

reading a value of the first pixel from the memory;

reading a highest value among nearest neighboring pixels to the first pixel from the memory, wherein the highest value is a value of a second pixel;

reading a next highest value among the nearest neighboring pixels to the first pixel from the memory, wherein the next highest value is a value of a third pixel;

performing a first function on the value of the second pixel, wherein the first function produces a first output value;

if the value of the first pixel is higher than the first output value, determining that the first pixel is hot;

performing a second function on the value of the third pixel, wherein the second function produces a second output value;

if the value of the first pixel is higher than the second output value and the value of the second pixel is also higher than the second output value, determining that the first pixel is hot.

2. The method of claim 1,

Wherein the first function and the second function are the same function.

3. The method of claim 1,

Wherein the first function comprises a first multiplier and a first offset, wherein the second function comprises a second multiplier and a second offset.

4. The method of claim 1,

wherein the first function and the second function are variable based at least in part on the gain of the sensor.

5. The method of claim 1,

wherein the first pixel represents a first color;

wherein the nearest neighboring pixels comprise the nearest neighboring pixels which also sense the first color.

6. The method of claim 1,

wherein the pixels are arranged in a square grid;

wherein the nearest neighboring pixels comprise the eight nearest neighboring pixels.

7. The method of claim 1,

wherein the pixels are arranged in a Bayer pattern;

wherein the nearest neighboring pixels comprise the nearest neighboring pixels in the same color plane as the first pixel.

8. The method of claim 1,

wherein the plurality of pixel values are luminosity values.

9. The method of claim 1,

wherein the first pixel represents a first color;

wherein the first function and the second function are based at least in part on the first color.

10. The method of claim 1, further comprising:

if the first pixel is defective and at least one of the nearest neighboring pixels is also defective, determining that the sensor has a defective cluster comprising the first pixel and the at least one nearest neighboring pixel.

11. A method for detecting cold pixels in a sensor which senses a plurality of pixels, comprising:

detecting a plurality of pixel values, the plurality of pixel values comprising values of a first pixel and each nearest neighboring pixel to the first pixel;

storing the plurality of pixel values in a memory;

reading a value of the first pixel from the memory;

reading a lowest value among nearest neighboring pixels to the first pixel from the memory, wherein the lowest value is a value of a second pixel;

reading a next lowest value among the nearest neighboring pixels to the first pixel from the memory, wherein the next lowest value is a value of a third pixel;

performing a first function on the value of the second pixel, wherein the first function produces a first output value;

if the value of the first pixel is lower than the first output value, determining that the first pixel is cold;

performing a second function on the value of the third pixel, wherein the second function produces a second output value;

if the value of the first pixel is lower than the second output value and the value of the second pixel is lower than the second output value, determining that the first pixel is cold.

12. The method of claim 11,

Wherein the first function and the second function are the same function.

13. The method of claim 11,

Wherein the first function comprises a first multiplier and a first offset, wherein the second function comprises a second multiplier and a second offset.

14. The method of claim 11,

wherein the first function and the second function are variable based at least in part on the gain of the sensor.

15. The method of claim 11,

wherein the first pixel represents a first color;

wherein the nearest neighboring pixels comprise the nearest neighboring pixels which also sense the first color.

16. The method of claim 11,

wherein the pixels are arranged in a square grid;

wherein the nearest neighboring pixels comprise the eight nearest neighboring pixels.

17. The method of claim 11,

wherein the pixels are arranged in a Bayer pattern;

wherein the nearest neighboring pixels comprise the nearest neighboring pixels in the same color plane as the first pixel.

18. The method of claim 11,

wherein the plurality of pixel values are luminosity values.

19. The method of claim 11,

wherein the first pixel represents a first color;

wherein the first function and the second function are based at least in part on the first color.

20. The method of claim 11, further comprising:

if the first pixel is defective and at least one of the nearest neighboring pixels is also defective, determining that the sensor has a defective cluster comprising the first pixel and the at least one nearest neighboring pixel.

21. A method for detecting defective pixels in a sensor which senses a plurality of pixels, comprising:

detecting a plurality of pixel values, the plurality of pixel values comprising values of a first pixel and each nearest neighboring pixel to the first pixel;

storing the plurality of pixel values in a memory;

reading a value of the first pixel from the memory;

reading a highest value among the nearest neighboring pixels to the first pixel from the memory, wherein the highest value is a value of a second pixel;

reading a next highest value among the nearest neighboring pixels to the first pixel from the memory, wherein the next highest value is a value of a third pixel;

performing a first function on the value of the second pixel, wherein the first function produces a first output value;

if the value of the first pixel is higher than the first output value, determining that the first pixel is hot;

performing a second function on the value of the third pixel, wherein the second function produces a second output value;

if the value of the first pixel is higher than the second output value and the value of the second pixel is also higher than the second output value, determining that the first pixel is hot.

reading a lowest value among the nearest neighboring pixels to the first pixel from the memory, wherein the lowest value is a value of a fourth pixel;

reading a next lowest value among the nearest neighboring pixels to the first pixel from the memory, wherein the next lowest value is a value of a fifth pixel;

performing a third function on the value of the fourth pixel, wherein the third function produces a third output value;

if the value of the first pixel is lower than the third output value, determining that the first pixel is cold;

performing a fourth function on the value of the fifth pixel, wherein the fourth function produces a fourth output value;

if the value of the first pixel is lower than the fourth output value and the value of the fourth pixel is lower than the fourth output value, determining that the first pixel is cold.

22. The method of claim 21, further comprising:

if the value of the first pixel is not higher than the first output value; and

if the value of the first pixel is not higher than the second output value and the value of the second pixel is not also higher than the second output value; and

if the value of the first pixel is not lower than the third output value; and

if the value of the first pixel is not lower than the fourth output value and the value of the fourth pixel is not also lower than the fourth output value;

then;

determining that the first pixel is not defective.

23. The method of claim 21, further comprising:

if the value of the first pixel is higher than the second output value and the value of the second pixel is also higher than the second output value, determining that the second pixel is also defective;

if the value of the first pixel is lower than the fourth output value and the value of the fourth pixel is also lower than the fourth output value, determining that the fourth pixel is also defective.

24. The method of claim 21,

wherein one or more of the first function, the second function, the third function, and the fourth function are variable based at least in part on the gain of the sensor.

25. A system for determining if a pixel is defective, comprising:

an input for receiving a plurality of pixel values;

a memory, coupled to the input, for storing the plurality of pixel values; and

logic, coupled to the memory, configured to;

read a value of a first pixel from the memory;

read a highest value among nearest neighboring pixels to the first pixel from the memory, wherein the highest value is a value of a second pixel;

read a next highest value among the nearest neighboring pixels to the first pixel from the memory, wherein the next highest value is a value of a third pixel;

perform a first function on the value of the second pixel, wherein the first function produces a first output value;

if the value of the first pixel is higher than the first output value, determine that the first pixel is hot;

perform a second function on the value of the third pixel, wherein the second function produces a second output value;

if the value of the first pixel is higher than the second output value and the value of the second pixel is also higher than the second output value, determine that the first pixel is hot.

26. A system for determining if a pixel is defective, comprising:

an input for receiving a plurality of pixel values;

a memory, coupled to the input, for storing the plurality of pixel values; and

logic, coupled to the memory, configured to;

read a value of a first pixel from the memory;

read a lowest value among the nearest neighboring pixels to the first pixel from the memory, wherein the lowest value is a value of a second pixel;

read a next lowest value among the nearest neighboring pixels to the first pixel from the memory, wherein the next lowest value is a value of a third pixel;

perform a first function on the value of the second pixel, wherein the first function produces a first output value;

if the value of the first pixel is lower than the first output value, determine that the first pixel is cold;

perform a second function on the value of the third pixel, wherein the second function produces a second output value;

if the value of the first pixel is lower than the second output value and the value of the second pixel is lower than the second output value, determine that the first pixel is cold.

27. A computer readable memory medium comprising program instructions for detecting defective pixels in a sensor, wherein the program instructions are executable to:

detect a plurality of pixel values, the plurality of pixel values comprising values of a first pixel and each nearest neighboring pixel to the first pixel;

store the plurality of pixel values;

read a value of the first pixel;

read a highest value among nearest neighboring pixels to the first pixel, wherein the highest value is a value of a second pixel;

read a next highest value among the nearest neighboring pixels to the first pixel, wherein the next highest value is a value of a third pixel;

perform a first function on the value of the second pixel, wherein the first function produces a first output value;

if the value of the first pixel is higher than the first output value, determine that the first pixel is hot;

perform a second function on the value of the third pixel, wherein the second function produces a second output value;

if the value of the first pixel is higher than the second output value and the value of the second pixel is also higher than the second output value, determine that the first pixel is hot.

28. A computer readable memory medium comprising program instructions for detecting defective pixels in a sensor, wherein the program instructions are executable to:

detect a plurality of pixel values, the plurality of pixel values comprising values of a first pixel and each nearest neighboring pixel to the first pixel;

store the plurality of pixel values;

read a value of the first pixel;

read a lowest value among the nearest neighboring pixels to the first pixel, wherein the lowest value is a value of a second pixel;

read a next lowest value among the nearest neighboring pixels to the first pixel, wherein the next lowest value is a value of a third pixel;

perform a first function on the value of the second pixel, wherein the first function produces a first output value;

if the value of the first pixel is lower than the first output value, determine that the first pixel is cold;

perform a second function on the value of the third pixel, wherein the second function produces a second output value;

if the value of the first pixel is lower than the second output value and the value of the second pixel is lower than the second output value, determine that the first pixel is cold.

29. A camera, comprising:

an image sensor which senses a plurality of pixels;

a memory coupled to the image sensor;

logic coupled to the image sensor and the memory, wherein the logic is configured to;

detect a plurality of pixel values of at least a subset of the plurality of light-sensitive pixels, the plurality of pixel values comprising values of a first pixel and each nearest neighboring pixel to the first pixel;

store the plurality of pixel values in the memory;

read a value of the first pixel from the memory;

read a highest value among nearest neighboring pixels to the first pixel from the memory, wherein the highest value is a value of a second pixel;

read a next highest value among the nearest neighboring pixels to the first pixel from the memory, wherein the next highest value is a value of a third pixel;

perform a first function on the value of the second pixel, wherein the first function produces a first output value;

if the value of the first pixel is higher than the first output value, determine that the first pixel is hot;

perform a second function on the value of the third pixel, wherein the second function produces a second output value;

if the value of the first pixel is higher than the second output value and the value of the second pixel is also higher than the second output value, determine that the first pixel is hot.

30. A camera, comprising:

an image sensor which senses a plurality of pixels;

a memory coupled to the image sensor;

logic coupled to the image sensor and the memory, wherein the logic is configured to;

detect a plurality of pixel values of at least a subset of the plurality of light-sensitive pixels, the plurality of pixel values comprising values of a first pixel and each nearest neighboring pixel to the first pixel;

store the plurality of pixel values in the memory;

read a value of the first pixel from the memory;

read a lowest value among the nearest neighboring pixels to the first pixel from the memory, wherein the lowest value is a value of a second pixel;

read a next lowest value among the nearest neighboring pixels to the first pixel from the memory, wherein the next lowest value is a value of a third pixel;

perform a first function on the value of the second pixel, wherein the first function produces a first output value;

if the value of the first pixel is lower than the first output value, determine that the first pixel is cold;

perform a second function on the value of the third pixel, wherein the second function produces a second output value;

if the value of the first pixel is lower than the second output value and the value of the second pixel is lower than the second output value, determine that the first pixel is cold.

31. A method for detecting defective pixels, the method comprising:

performing a first comparison of a value of a first pixel with a value of a second pixel, wherein the first comparison is for determining if the first pixel has a first type of defect;

performing a second comparison, wherein the second comparison compares the values of the first pixel and the second pixel with a value of a third pixel, wherein the second comparison is also for determining if the first pixel has the first type of defect;

determining if the first pixel has the first type of defect based on the first comparison and the second comparison.

32. The method of claim 31,

wherein the first comparison uses a first function to bias the first comparison.

33. The method of claim 32, wherein the first function is of the form f(x)=ax+b, wherein the first function is performed on the value of the second pixel, thereby producing a first output value, wherein the first comparison comprises a comparison of the value of the first pixel with the first output value.

34. The method of claim 31,

wherein the first comparison uses a first function to bias the first comparison;

wherein the second comparison uses a second function to bias the second comparison.

35. The method of claim 31,

wherein the first type of defect is insufficient sensitivity.

36. The method of claim 31,

wherein the first type of defect is excessive sensitivity.

37. The method of claim 31,

wherein the first, second and third pixels are comprised in a plurality of pixels arranged in a Bayer pattern, wherein the second and third pixels are adjacent to the first pixel in a color plane.