Dark current gradient estimation using optically black pixels

Patent metadata

Field	Value
Publication number	US-9628730-B2
Application number	US-201514606562-A
Country	US
Kind code	B2
Filing date	Jan 27, 2015
Priority date	Jan 27, 2015
Publication date	Apr 18, 2017
Grant date	Apr 18, 2017

Abstract

A method for removing background signal values from readout signal values generated by image pixels in a pixel array may include generating background signal values using optically black pixels formed around a border of the pixel array. The background signal values may be separated into blocks. Block average values may be calculated based on the background signal values included in each of the blocks. Corner regions of the pixel array may be free of image pixels and optically black pixels. Corner background signal values may be extrapolated based on the block average values. Image pixel background signal values may be interpolated based on the block average values and the corner background signal values. The image pixel background signal values may be subtracted from the readout values.

First claim

What is claimed is:

1. A method for correcting for background noise in a frame of image data, the method comprising:
receiving the frame of image data from an image sensor having an array of image pixels and a plurality of optically black pixels, wherein the frame of image data includes first background signal values from the optically black pixels;
grouping the first background signal values into blocks;
calculating block average values based on the first background signal values included in each respective block;
determining second background signal values for the image pixels in the array based on the block averages, wherein the second background signal values indicate an amount of background noise associated with the image pixels;
generating corner background signal values for corner regions of the pixel array based on the block average values, wherein the corner regions of the pixel array are free of image pixels and optically black pixels; and
applying the second background signal values to the frame of image data.

2. The method defined in claim 1 , wherein the second background signal values are determined using bilinear interpolation of the first background signal values.

3. The method defined in claim 1 , wherein the corner background signal values are extrapolated based on the block average values.

4. The method defined in claim 1 , wherein the second background signal values are determined based on the block average values and the corner background signal values.

5. The method defined in claim 1 , wherein the optically black pixels are formed in a border around the array having first and second opposing edges and third and fourth opposing edges.

6. The method defined in claim 5 , wherein the first background signal values from the optically black pixels formed in each of the respective first, second, third, and fourth edges are grouped into at least respective first, second, third, and fourth blocks, and wherein respective first, second, third, and fourth block average values are calculated for each of the respective first, second, third, and fourth blocks.

7. The method defined in claim 6 , further comprising:
for at least one of the image pixels, generating a first component background signal value based on the first and second block average values, and generating a second component background signal value based on the third and fourth block average values.

8. The method defined in claim 7 , wherein the second background signal value is determined by subtracting a corner background signal value of the corner background signal values from a sum of the first and second composite background signal values.

9. The method defined in claim 1 , wherein the corner regions of the pixel array include traces.

10. A method of removing background signals generated by image pixels in a pixel array, the method comprising:
determining a background signal estimate for an image pixel in the pixel array based on background signal values from optically black pixels arranged in a border around the pixel array, wherein the pixel array has first and second opposing edges connected by third and fourth opposing edges, wherein the background signal values from the optically black pixels formed in the first edge of the pixel array are grouped into at least respective first, second, third and fourth blocks, and wherein respective first, second, third and fourth block average values are calculated for each of the respective first, second, third and fourth blocks;
receiving an image pixel readout value generated by the image pixel in response to light that is received at the pixel array, and
using the background signal estimate to remove the background signals from the image pixel readout value and obtain a background signal-compensated image pixel value.

11. The method defined in claim 10 , wherein determining the background signal estimate comprises:
determining the background signal values using the optically black pixels, wherein the background signal values indicate dark current generated by the optically black pixels;
grouping the background signal values from adjacent optically black pixels into blocks;
calculating block averages based on the background signal values included in the respective blocks; and
determining the background signal estimate for the image pixel based on the block averages.

12. The method defined in claim 11 , wherein the image pixel readout value includes an image signal value component generated in response to the light that is received at the pixel array and a background signal value component generated in response to dark current generated by the image pixel, and wherein the background signal estimate is an estimate of the background signal value component in the image pixel readout value.

13. The method defined in claim 11 , wherein determining the background signal estimate comprises:
interpolating a first estimate component based on a first set of the block averages;
interpolating a second estimate component based on a second set of the block averages; and
interpolating a third estimate component based on a third set of the block averages, wherein the first, second, and third sets of block averages are different, and wherein the background signal estimate is determined based on the first, second, and third estimate components.

14. The method defined in claim 13 , wherein the third set of block averages is used to extrapolate corner background signal values for corners of the pixel array, and wherein the third estimate component is generated based on the corner background signal values.

15. The method defined in claim 10 , wherein the background signal estimate is determined using bilinear interpolation of the background signal values.

16. A method for removing background signals from image data generated by pixels in an array, the method comprising:
with a first plurality of pixels, generating a frame of image data in response to light received at the array;
with a second plurality of pixels that do not receive light, generating background signal values;
grouping the background signal values into blocks, wherein the blocks include background signal values generated by the second plurality of pixels;
calculating block averages for each of the respective blocks based on the background signal values in each of the respective blocks;
interpolating additional background signal values for the first plurality of pixels based on the block averages; and
applying the additional background signal values to the frame of image data.

17. The method defined in claim 16 , wherein the image data includes readout signal values, and wherein applying the additional background signal values to the frame of image data comprises subtracting the additional background signal values from readout signal values generated by the first plurality of pixels.

18. The method defined in claim 16 , wherein the additional background signal values are applied to a plurality of additional frames of image data generated after the frame of image data.

Assignees

• Semiconductor Components Ind Llc

Inventors

• Vanhoff Barry

Classifications

• H04N25/134 Primary

  based on three different wavelength filter elements · CPC title

• H04N23/57

  Mechanical or electrical details of cameras or camera modules specially adapted for being embedded in other devices · CPC title

• H04N25/633

  by using optical black pixels · CPC title

• H04N5/361 Primary

  Electricity · mapped topic

• H04N5/2257

  Electricity · mapped topic