Method, apparatus and system for dynamic range estimation of imaged scenes

• US 20080056704A1
• Filed: 08/30/2006
• Published: 03/06/2008
• Est. Priority Date: 08/30/2006
• Status: Active Grant

• Current Assignee: Micron Technology, Inc.
• Original Assignee: Micron Technology, Inc.
• Inventors: Ovsiannikov, Ilia
• Granted Patent: US 7,944,485 B2
• Time in Patent Office: Days
• US Class Current: 396/234
• CPC Class Codes: G03B 7/08 Control effected solely on the basis of the response, to the intensity of the light received by the camera, of a built-in light-sensitive device

Abstract

A method, apparatus, and system for dynamic range estimation of imaged scenes for automatic exposure control. For a given exposure time setting, certain areas of a scene may be brighter than what a camera can capture. In cameras, including those experiencing substantial lens vignetting, a gain stage may be used to extend dynamic range and extract auto-exposure data from the extended dynamic range. Alternatively, dynamic range can be extended using pre-capture image information taken under reduced exposure conditions.

Claims

1. An imaging device comprising:

a pixel array;

a readout circuit for reading out digitized pixel signals from the pixel array;

a pixel processing pipeline for processing the digitized pixel signals, the processing pipeline comprising a histogram stage adapted to use un-clipped pixel data to generate automatic exposure information, and

an automatic exposure circuit for adjusting image exposure parameters based on the generated automatic exposure information.

2. The imaging device of claim 1, wherein the digitized pixel signals have an N-bit width and the pixel processing pipeline comprises at least one gain stage for applying a gain to at least some of the pixel signals, the at least one gain stage providing the un-clipped pixel data with a bit width of N+M, where M is an additional number of bits, and the histogram stage provides automatic exposure information having the N+M-bit width.

3. The imaging device of claim 2, further comprising at least one pixel signal processing stage, wherein the at least one pixel signal processing stage processes the pixel signals output from the at least one gain stage with a bit width of N+M.

4. The imaging device of claim 2, further comprising at least two pixel signal processing stages, wherein a first of the pixel signal processing stages processes the pixel signals output from the at least one gain stage with a bit width of N+M and a second of the pixel signal processing stages processes clipped pixel signals having a bit width of N.

5. The imaging device of claim 4, wherein the first pixel signal processing stage is a demosaicing stage and the second pixel signal processing stage is a gamma correction stage.

6. The imaging device of claim 4, wherein the first pixel signal processing stage is a color correction stage and the second pixel signal processing stage is a gamma correction stage.

7. The imaging device of claim 2, wherein the at least one gain stage is a white point correction stage processing the pixel signals to form un-clipped data having the N+M-bit width.

8. The imaging device of claim 7, wherein the at least one gain stage further comprises a lens vignetting correction stage that processes un-clipped data having the N+M-bit width.

9. The imaging device of claim 1, wherein pixel signals from a pre-capture image are under-exposed to provide un-clipped pixel data to the histogram stage.

10. The imaging device of claim 9, further comprising at least one gain stage for applying a gain to the under-exposed processed pixel signals independent of the histogram stage.

11. A digital camera comprising:

a lens; and

an imaging device positioned to receive an image through the lens, the imaging device comprising;

an image sensor, the image sensor adapted to output digitized pixel signals having a first bit-width;

an image processing unit for processing un-clipped pixel data corresponding to the digital pixel signals using a histogram to obtain automatic exposure information; and

an auto-exposure circuit for adjusting an exposure value for the image sensor based on the histogram automatic exposure information.

12. The digital camera of claim 11, wherein the digitized pixel signals have an N-bit width and the image processing unit comprises at least one gain stage for applying a gain to at least some of the pixel signals, the at least one gain stage providing the un-clipped pixel data with a bit width of N+M, and the histogram provides automatic exposure information having the N+M-bit width.

13. The digital camera of claim 12, further comprising at least one pixel signal processing stage, wherein the at least one pixel signal processing stage processes the pixel signals output from the at least one gain stage with a bit width of N+M.

14. The digital camera of claim 12, further comprising at least two pixel signal processing stages, wherein a first of the pixel signal processing stages processes the pixel signals output from the at least one gain stage with a bit width of N+M and a second of the pixel signal processing stages processes clipped pixel signals having a bit width of N.

15. The digital camera of claim 11, wherein pixel signals from a pre-capture image are under-exposed to provide un-clipped pixel data to the histogram stage.

16. The digital camera of claim 15, further comprising at least one gain stage for applying a gain to the under-exposed processed pixel signals independent of the histogram.

17. The digital camera of claim 11, wherein the histogram uses only pixel data from one color component in the image.

18. The digital camera of claim 11, wherein the histogram comprises three histograms, each histogram for each color component in the image.

19. The digital camera of claim 11, wherein the histogram uses combined color components in the image.

20. An automatic exposure control method comprising:

inputting pixel signals representing a scene, the signals having an N-bit width;

applying a gain to the pixel signals to generate modified pixel signals having an N+M-bit width;

mapping the modified pixel signals using a histogram;

determining a high and a low limit of the scene dynamic range using N+M-bit information; and

adjusting an exposure value for the scene based on the high and low limits.

21. The method of claim 20, wherein determining the low limit comprises:

examining the histogram to determine a lower histogram value of the darkest pixels in an image; and

assigning the lower histogram value to the low limit.

22. The method of claim 21, wherein the histogram is mapped after performing a lens vignetting correction operation on the modified pixel signals.

23. The method of claim 21, wherein the histogram is mapped after performing a color correction operation on the modified pixel signals.

24. The method of claim 20, wherein determining the high limit comprises:

mapping a separate histogram for each color component with information from the modified pixel signals having an N+M-bit width;

examining green color pixel values in the histogram;

determining the green color pixel value in the histogram corresponding to the brightest value in the histogram; and

assigning the green color pixel value in the histogram corresponding to the brightest value in the histogram to the high limit.

25. The method of claim 24, wherein the histogram is mapped after performing a lens vignetting correction operation on the modified pixel signals.

26. The method of claim 24, wherein the histogram is mapped after performing a color correction operation on the modified pixel signals.

27. The method of claim 20, wherein determining the high limit comprises:

mapping a histogram with information from the modified pixel signals having an N+M-bit width;

examining the histogram to determine a higher histogram value of the brightest pixels in the image; and

assigning the higher histogram value to the high limit.

28. The method of claim 20, wherein determining the high limit comprises:

mapping a histogram for a green color component with information from the modified pixel signals having an N+M-bit width;

determining a green color pixel value in the histogram corresponding to the brightest value in the histogram; and

assigning the green color pixel value to the high limit.

29. An automatic exposure method comprising:

underexposing an image of a scene by a predetermined amount to ensure that image data does not become clipped;

inputting underexposed image pixel data to an image processing pipeline;

processing the underexposed image pixel data using a histogram to obtain histogram information from the image pixel data;

examining the histogram to determine a high limit; and

adjusting an exposure parameter based on the determined high limit.

30. The method of claim 29, wherein the histogram is mapped after performing a color correction operation on the image pixel data.