Noise estimation using user-configurable information

Patent metadata

Field	Value
Publication number	US-12375820-B2
Application number	US-202217739291-A
Country	US
Kind code	B2
Filing date	May 9, 2022
Priority date	Jun 14, 2019
Publication date	Jul 29, 2025
Grant date	Jul 29, 2025

Abstract

In some examples, a method comprises receiving pixel data from an image capture device having a color filter, wherein the pixel data represents a portion of an image. The method further includes performing wavelet decomposition on the pixel data to produce decomposed pixel data and determining a local intensity of the pixel data. The method also includes determining a noise threshold value based on the local intensity and a noise intensity function that is based on the color filter; determining a noise value for the pixel data based on the decomposed pixel data and the noise threshold value; and correcting the pixel data based on the noise value to produce an output image.

First claim

What is claimed is:

1. A system comprising:
an image capture device configured to provide image data that includes a set of channels;
a memory configured to store correlation information that specifies a noise correlation value for each channel of the set of channels; and
an image processor coupled to the image capture device and the memory and configured to:
receive the image data;
based on the image data, determine a respective local intensity for each channel of the set of channels;
sum the respective local intensities of a subset of the set of channels based on the correlation information to produce a sum of local intensities;
determine a noise threshold based on the sum of local intensities;
perform wavelet decomposition on the image data to produce respective decomposed image data for each channel of the set of channels;
sum the respective decomposed image data of the subset of the set of channels based on the correlation information to produce an intermediate sum; and
apply a noise suppression function to each channel of the subset of the set of channels based on the noise threshold, wherein the noise suppression function is based on the intermediate sum.

2. The system of claim 1 , wherein:
the image capture device includes a color filter;
each channel of the set of channels of the image data is associated with a respective color of the color filter; and
the correlation information is associated with the color filter.

3. The system of claim 1 , wherein:
the memory is further configured to store a noise-intensity function; and
the image processor is configured to determine the noise threshold further based on the noise-intensity function.

4. The system of claim 1 , wherein, to determine the noise threshold, the image processor is further configured to:
apply an averaging value that is based on a number of channels in the subset of the set of channels to the sum of local intensities to produce an intermediate intensity value; and
apply a noise-intensity function to the intermediate intensity value.

5. The system of claim 1 , wherein the image processor is configured to:
determine a gain value associated with the image capture device; and
apply an inverse of the gain value to the sum of local intensities prior to the determining of the noise threshold.

6. The system of claim 5 , wherein the gain value is a lens shading correction compensation gain value.

7. The system of claim 1 , wherein:
the image data is associated with a set of images;
the image processor is further configured to merge the image data across images of the set of images to produce high dynamic range image data; and
the respective local intensity for each channel of the set of channels is based on the high dynamic range image data.

8. A system comprising:
an image capture device configured to provide image data that includes a set of channels including a first channel;
a memory configured to store correlation information that specifies a noise correlation value for each channel of the set of channels; and
an image processor coupled to the image capture device and the memory and configured to:
receive the image data;
based on the image data, determine a respective local intensity for each channel of the set of channels;
sum the respective local intensities of a subset of the set of channels based on the correlation information to produce a sum of local intensities, wherein the subset of the set of channels does not include the first channel;
determine a first noise threshold based on the sum of local intensities;
apply a noise suppression function to each channel of the subset of the set of channels based on the first noise threshold;
determine a local intensity of the first channel;
determine a second noise threshold based on the local intensity of the first channel that is independent of a remainder of the set of channels; and
apply the noise suppression function to the first channel based on the second noise threshold.

9. A method comprising:
receiving image data that includes a set of channels;
determining a respective local intensity for each channel of the set of channels;
receiving correlation information associated with the set of channels;
determining a subset of the set of channels to sum based on the correlation information;
summing the respective local intensities of the subset of the set of channels to produce a sum of local intensities;
determining a noise threshold based on the sum of local intensities
producing respective decomposed image data for each channel of the set of channels;
summing the respective decomposed image data of the subset of the set of channels based on the correlation information to produce an intermediate sum; and
applying a noise suppression function to each channel of the subset of the set of channels based on the noise threshold, wherein the noise suppression function is based on the intermediate sum.

10. The method of claim 9 , wherein:
the image data is associated with a color filter; and
the correlation information is associated with the color filter.

11. The method of claim 9 , wherein the producing of the respective decomposed image data for each channel of the set of channels includes:
performing wavelet decomposition on the image data to produce the respective decomposed image data for each channel of the set of channels.

12. The method of claim 9 further comprising:
determining a lens shading correction gain value; and
applying an inverse of the lens shading correction gain value to the sum of local intensities prior to the determining of the noise threshold.

13. The method of claim 9 , wherein:
the image data is associated with a set of images;
the method further comprises merging the image data across images of the set of images to produce high dynamic range image data; and
the respective local intensity for each channel of the set of channels is based on the high dynamic range image data.

14. The method of claim 9 , wherein the determining of the noise threshold includes applying an averaging value that is based on a number of channels in the subset of the set of channels to the sum of local intensities to produce an intermediate intensity value, wherein the noise threshold is based on the intermediate intensity value.

15. A method comprising:
receiving image data that includes a set of channels that includes a first channel;
determining a respective local intensity for each channel of the set of channels;
receiving correlation information associated with the set of channels;
determining a subset of the set of channels to sum based on the correlation information, wherein the subset of the set of channels does not include the first channel;
summing the respective local intensities of the subset of the set of channels to produce a sum of local intensities;
determining a first noise threshold based on the sum of local intensities; and
applying a noise suppression function to each channel of the subset of the set of channels based on the first noise threshold;
wherein the method further comprises:
determining a local intensity of the first channel;
determining a second noise threshold based on the local intensity of the first channel that is independent of a remainder of the set of channels; and
applying the noise suppression function to the first channel based on the second noise threshold.

16. A device comprising:
an image processor; and
a non-transitory computer-readable memory coupled to the image processor that stores instructions that, when executed, cause the image

Assignees

• Texas Instruments Inc

Inventors

• Hua Gang
• Mody Mihir Narendra
• Allu Rajasekhar Reddy
• Nandan Niraj
• Dabral Shashank

Classifications

• H04N25/61

  the noise originating only from the lens unit, e.g. flare, shading, vignetting or "cos4" · CPC title

• H04N23/84

  for processing colour signals · CPC title

• G06T2207/20064

  Wavelet transform [DWT] · CPC title

• H04N25/135

  based on four or more different wavelength filter elements · CPC title

• H04N25/133

  including elements passing panchromatic light, e.g. filters passing white light · CPC title