Texture-preserving halo suppression for imaging systems

Patent metadata

Field	Value
Publication number	US-12541826-B2
Application number	US-202318490505-A
Country	US
Kind code	B2
Filing date	Oct 19, 2023
Priority date	Jul 24, 2023
Publication date	Feb 3, 2026
Grant date	Feb 3, 2026

Abstract

A method includes obtaining, using at least one processing device of an electronic device, an input image containing blur. The method also includes generating, using the at least one processing device, an edge enhancement mask and a gain mask based on the input image. The method further includes generating, using the at least one processing device, a halo-suppressed edge mask based on the edge enhancement mask and the gain mask. In addition, the method includes generating, using the at least one processing device, a sharpened image based on the input image and the halo-suppressed edge mask.

First claim

What is claimed is:

1 . A method comprising:
obtaining, using at least one processing device of an electronic device, an input image containing blur; generating, using the at least one processing device, an edge enhancement mask and a non-binary gain mask based on the input image, wherein the non-binary gain mask contains pixel-wise gain operator values including at least one of zero or one and at least one value between zero and one; generating, using the at least one processing device, a halo-suppressed edge mask based on the edge enhancement mask and the non-binary gain mask; and generating, using the at least one processing device, a sharpened image based on the input image and the halo-suppressed edge mask; wherein generating the non-binary gain mask comprises:
generating a non-binary initial edge mask based on a gradient computation performed using the input image;
generating a non-binary refined edge mask based on a first morphological operation performed using the non-binary initial edge mask; and
generating a non-binary thickened edge mask based on a second morphological operation performed using the non-binary initial edge mask, and
wherein each of the first morphological operation and the second morphological operation comprises a transform that replaces a value for at least one pixel with one of a minimum pixel value or a maximum pixel value within a window of pixels surrounding the at least one pixel.

2 . The method of claim 1 , wherein generating the non-binary gain mask further comprises:
combining the non-binary refined edge mask and the non-binary thickened edge mask to generate a non-binary gain suppression mask; generating a non-binary gain enhancement mask based on the non-binary gain suppression mask; and generating the non-binary gain mask based on the non-binary gain enhancement mask and the non-binary gain suppression mask.

3 . The method of claim 1 , wherein:
the first morphological operation comprises a first dilation operation followed by an erosion operation; and the second morphological operation comprises a second dilation operation.

4 . The method of claim 2 , wherein generating the non-binary gain mask further comprises:
modulating the non-binary gain suppression mask based on edges in the input image, wherein first multiplicative values for modulating the non-binary gain suppression mask are based on at least one of: edge strengths, edge coherences, and edge orientations; and modulating the non-binary gain enhancement mask based on the edges in the input image, wherein second multiplicative values for modulating the non-binary gain enhancement mask are based on negative enhancements or positive enhancements associated with the edges; and wherein the non-binary gain mask is based on the modulated gain enhancement mask and the modulated gain suppression mask.

5 . The method of claim 1 , wherein generating the edge enhancement mask comprises generating the edge enhancement mask using unsharp-masking (USM) of the input image.

6 . The method of claim 1 , wherein generating the halo-suppressed edge mask comprises performing a pixel-wise multiplication of the edge enhancement mask and the non-binary gain mask.

7 . The method of claim 1 , wherein generating the sharpened image comprises adding the halo-suppressed edge mask to the input image in order to generate the sharpened image.

8 . An electronic device comprising:
at least one processing device configured to:
obtain an input image containing blur;
generate an edge enhancement mask and a non-binary gain mask based on the input image, wherein the non-binary gain mask contains pixel-wise gain operator values including at least one of zero or one and at least one value between zero and one;
generate a halo-suppressed edge mask based on the edge enhancement mask and the non-binary gain mask; and
generate a sharpened image based on the input image and the halo-suppressed edge mask;
wherein, to generate the non-binary gain mask, the at least one processing device is configured to:
generate a non-binary initial edge mask based on a gradient computation performed using the input image;
generate a non-binary refined edge mask based on a first morphological operation performed using the non-binary initial edge mask; and
generate a non-binary thickened edge mask based on a second morphological operation performed using the non-binary initial edge mask, and
wherein each of the first morphological operation and the second morphological operation comprises a transform that replaces a value for at least one pixel with one of a minimum pixel value or a maximum pixel value within a window of pixels surrounding the at least one pixel.

9 . The electronic device of claim 8 , wherein, to generate the non-binary gain mask, the at least one processing device is further configured to:
combine the non-binary refined edge mask and the non-binary thickened edge mask to generate a non-binary gain suppression mask; generate a non-binary gain enhancement mask based on the non-binary gain suppression mask; and generate the non-binary gain mask based on the non-binary gain enhancement mask and the non-binary gain suppression mask.

10 . The electronic device of claim 8 , wherein:
the first morphological operation comprises a first dilation operation followed by an erosion operation; and the second morphological operation comprises a second dilation operation.

11 . The electronic device of claim 9 , wherein, to generate the non-binary gain mask, the at least one processing device is further configured to:
modulate the non-binary gain suppression mask based on edges in the input image, wherein first multiplicative values for modulating the non-binary gain suppression mask are based on at least one of: edge strengths, edge coherences, and edge orientations; and modulate the non-binary gain enhancement mask based on the edges in the input image, wherein second multiplicative values for modulating the non-binary gain enhancement mask are based on negative enhancements or positive enhancements associated with the edges; and wherein the non-binary gain mask is based on the modulated gain enhancement mask and the modulated gain suppression mask.

12 . The electronic device of claim 8 , wherein, to generate the edge enhancement mask, the at least one processing device is configured to use unsharp-masking (USM) of the input image.

13 . The electronic device of claim 8 , wherein, to generate the halo-suppressed edge mask, the at least one processing device is configured to perform a pixel-wise multiplication of the edge enhancement mask and the non-binary gain mask.

14 . The electronic device of claim 8 , wherein, to generate the sharpened image, the at least one processing device is configured to add the halo-suppressed edge mask to the input image.

15 . A non-transitory machine readable medium containing instructions that when executed cause at least one processor of an electronic device to:
obtain an input image containing blur; generate an edge enhancement mask and a non-binary gain mask based on the input image, wherein the non-binary gain mask contains pixel-wise gain operator values including at least one of zero or one and at least one value between zero and one; generate a halo-suppressed edge mask based on the edge enhancement mask and the non-binary gain mask; and generate a sharpened image based on the input image and the halo-suppressed edge mask; wherein the instructions that when executed cause the at least one processor to generate the non-binary gain mask comprise instructions th

Assignees

• Samsung Electronics Co Ltd

Inventors

• Godaliyadda Gunawath Dilshan
• Luu Tyler
• Glotzbach John W
• Sheikh Hamid R

Classifications

• G06T5/75 Primary

  Unsharp masking · CPC title

• G06T2207/20192

  Edge enhancement; Edge preservation · CPC title

• G06T2207/10024

  Color image · CPC title

• G06T5/30

  Erosion or dilatation, e.g. thinning · CPC title