COUPLET CORRECTION IN IMAGE SENSORS

• US 20070285535A1
• Filed: 05/25/2007
• Published: 12/13/2007
• Est. Priority Date: 05/29/2006
• Status: Active Grant

• Current Assignee: STMicroelectronics Limited (STMicroelectronics NV)
• Original Assignee: STMicroelectronics Limited (STMicroelectronics NV)
• Inventors: McGillvray, Gary, Storrar, David, Brown, Ewen, DUNCAN, Ed
• Granted Patent: US 8,036,483 B2
• Time in Patent Office: Days
• US Class Current: 348/246
• CPC Class Codes: H04N 25/68 applied to defects

Abstract

A digital filtering algorithm preprocesses couplet defects prior to their being operated on by a main filtering algorithm that focuses on singlet defects. The preprocessing involves the replacement of one of the defective pixel values in the couplet with a pixel value that is known to be non-defective. In preferred embodiments, the selection of the pixel value in the couplet to be replaced is customized depending on the direction of scanning of the pixel array, and can take into account edge effects of border pixels.

Claims

1-28. -28. (canceled)

29. A method of correcting for pixel defects in an output of an image sensor, the method comprising:

performing a preliminary filtering operation which operates on couplet defects wherein two adjacent pixels in a given image plane are defective, the preliminary filtering operation comprising replacing a signal value of a first pixel in the couplet with a corrected signal value; and

performing a main filtering operation on a signal resulting from the preliminary filtering operation.

30. The method of claim 29, wherein the corrected signal value in the preliminary filtering operation is a value of a neighboring non-defective pixel.

31. The method of claim 29, wherein performing the preliminary filtering operation comprises identifying a scanning direction of a readout process, and based on the identifying, choosing a pixel value from a pixel preceding the first pixel in the direction of scanning as the corrected signal value.

32. The method of claim 31, wherein the pixel preceding the first pixel in the direction of scanning is defined by a coordinate shift.

33. The method of claim 32, wherein the coordinate shift with respect to the first pixel corresponds to a translation in an opposite direction along a row to the direction in which the rows are sequentially being scanned.

34. The method of claim 31, wherein performing the preliminary filtering operation further comprises, for a couplet in which the first pixel is on a border of an origin side in the direction of scanning, choosing a pixel value from a pixel proceeding the first pixel in the direction of scanning as the corrected signal value.

35. The method of claim 34, wherein the pixel proceeding the first pixel in the direction of scanning is defined by a coordinate shift.

36. The method of claim 35, wherein the coordinate shift with respect to the first pixel corresponds to a translation along the direction in which rows are being scanned.

37. The method of claim 29, wherein the couplet defects are identified at manufacture and their location is stored in a memory.

38. The method of claim 37, wherein the memory comprises a non-volatile memory.

39. The method of claim 29, wherein the main filtering operation comprises:

obtaining a central pixel value and a plurality of neighboring pixel values;

determining the average value of each opposing pair of pixel values;

comparing each average value with the central pixel value; and

replacing the central pixel value with the average value which is closest to the central pixel value.

40. A signal processing system for use with an image sensor having a pixel array, the signal processing system comprising:

a preliminary filter for receiving image data and for performing a preliminary filtering operation which operates on couplet defects wherein two adjacent pixels in a given image plane are defective, the preliminary filtering operation comprising replacing a signal value of a first pixel in the couplet with a corrected signal value; and

a main filter for performing a main filtering operation on a signal resulting from the preliminary filtering operation.

41. The system of claim 40, wherein the corrected signal value in the preliminary filtering operation is a value of a neighboring non-defective pixel.

42. The system of claim 40, wherein said preliminary filter identifies a scanning direction of a readout process, and chooses a pixel value from a pixel preceding the first pixel in the direction of scanning as the corrected signal value.

43. The system of claim 42, wherein choosing the pixel value comprises defining a coordinate shift.

44. The system of claim 43, wherein the coordinate shift with respect to the first pixel corresponds to a translation in an opposite direction along a row to the direction in which the rows are sequentially being scanned.

45. The system of claim 42, wherein said preliminary filter identifies whether a couplet is on a border of the origin side in the direction of scanning, and chooses a pixel value from a pixel proceeding the first pixel in the direction of scanning as the corrected signal value.

46. The system of claim 45, wherein choosing the pixel value from a pixel proceeding the first pixel in the direction of scanning defines a coordinate shift.

47. The system of claim 46, wherein the coordinate shift with respect to the first pixel corresponds to a translation along the direction in which rows are being scanned.

48. The system of claim 40, further comprising a memory for storing locations of couplet defects.

49. The system of claim 48, wherein said memory comprises a non-volatile memory forming part of the image sensor.

50. The system of claim 40, wherein said main filter is configured for:

obtaining a central pixel value and a plurality of neighboring pixel values;

determining an average value of each opposing pair of pixel values;

comparing each average value with the central pixel value; and

replacing the central pixel value with the average value which is closest to the central pixel value.

51. An image sensor apparatus comprising:

a pixel array;

a controller for operating said pixel array to produce an image data stream; and

a signal processing system for processing the image data stream and comprising

a preliminary filter for receiving the image data stream and for performing a preliminary filtering operation which operates on couplet defects wherein two adjacent pixels in a given image plane are defective, the preliminary filtering operation comprising replacing a signal value of a first pixel in the couplet with a corrected signal value, and

a main filter for performing a main filtering operation on a signal resulting from the preliminary filtering operation.

52. The image sensor apparatus of claim 51, wherein the corrected signal value in the preliminary filtering operation is a value of a neighboring non-defective pixel.

53. The image sensor apparatus of claim 51, wherein said preliminary filter identifies a scanning direction of a readout process, and chooses a pixel value from a pixel preceding the first pixel in the direction of scanning as the corrected signal value.

54. The image sensor apparatus of claim 53, wherein choosing the pixel value comprises defining a coordinate shift, and wherein the coordinate shift with respect to the first pixel corresponds to a translation in an opposite direction along a row to the direction in which the rows are sequentially being scanned.

55. The image sensor apparatus of claim 53, wherein said preliminary filter identifies whether a couplet is on a border of the origin side in the direction of scanning, and chooses a pixel value from a pixel proceeding the first pixel in the direction of scanning as the corrected signal value.

56. The image sensor apparatus of claim 55, wherein choosing the pixel value from a pixel proceeding the first pixel in the direction of scanning defines a coordinate shift, and wherein the coordinate shift with respect to the first pixel corresponds to a translation along the direction in which rows are being scanned.

57. The image sensor apparatus of claim 51, further comprising a memory for storing locations of couplet defects, and wherein said memory comprises a non-volatile memory forming part of the image sensor.

59. The image sensor apparatus of claim 51, wherein said pixel array, said controller and said signal processing system are formed as a single chip.

60. The image sensor apparatus of claim 51, wherein said pixel array is part of at least one of the following:

a mobile device, a webcam, an optical mouse and a video camera.

58. The image sensor apparatus of claim S1, wherein said main filter is configured for:

obtaining a central pixel value and a plurality of neighboring pixel values;

determining an average value of each opposing pair of pixel values;

comparing each average value with the central pixel value; and

replacing the central pixel value with the average value which is closest to the central pixel value.