Reconstructing linear gradients

Patent metadata

Field	Value
Publication number	US-12450692-B2
Application number	US-202217901583-A
Country	US
Kind code	B2
Filing date	Sep 1, 2022
Priority date	Sep 1, 2022
Publication date	Oct 21, 2025
Grant date	Oct 21, 2025

Abstract

Embodiments are disclosed for reconstructing linear gradients from an input image that can be applied to another image. In particular, in one or more embodiments, the disclosed systems and methods comprise receiving a raster image, the raster image including a representation of a linear color gradient. The disclosed systems and methods further comprise determining a vector representing a direction of the linear color gradient. The disclosed systems and methods further comprise analyzing pixel points along the direction of the linear color gradient to compute color stops of the linear color gradient. The disclosed systems and methods further comprise generating an output color gradient vector with the computed color stops of the linear color gradient, the output color gradient vector to be applied to a vector graphic.

First claim

We claim:

1. A method comprising:
receiving, by a processing device, a raster image, the raster image including a representation of a linear color gradient;
determining a vector representing a direction of the linear color gradient, wherein the direction of the linear color gradient is a combination of a direction of each color channel of a pixel in the linear color gradient;
in response to determining that the direction of each color channel of the pixel in the linear color gradient is aligned, analyzing pixel points along the direction of the linear color gradient to compute color stops of the linear color gradient; and
generating an output color gradient vector with the computed color stops of the linear color gradient, the output color gradient vector to be applied to a vector graphic.

2. The method of claim 1 , wherein analyzing each pixel point along the direction of the linear color gradient to compute the color stops of the linear color gradient comprises:
determining average pixel column color values of the linear color gradient;
generating color channel data for each pixel point along the vector representing the direction of the linear color gradient from the determined average pixel column color values; and
identifying the color stops along the vector representing the direction of the linear color gradient at locations of peaks in the generated color channel data, the locations of peaks indicating changes in color values.

3. The method of claim 2 , wherein determining the average pixel column color values of the linear color gradient comprises:
rotating the raster image to align the vector representing the direction of the linear color gradient with a first axis;
for each pixel point along the first axis,
determining color values associated with pixels in a pixel column parallel to a second axis, the second axis perpendicular to the first axis, and
calculating an average pixel column color value for the pixel column based on the determined color values; and

generating the average pixel column color values of the linear color gradient by aggregating the average pixel column color value for each pixel point.

4. The method of claim 2 , wherein determining the average pixel column color values of the linear color gradient comprises:
for each pixel point along the vector representing the direction of the linear color gradient,
determining color values associated with pixels in a pixel column perpendicular to the vector representing the direction of the linear color gradient, and
calculating an average pixel column color value for the pixel column based on the determined color values; and

generating the average pixel column color values of the linear color gradient by aggregating the average pixel column color value for each pixel point.

5. The method of claim 2 , wherein identifying the color stops along the vector representing the direction of the linear color gradient at the locations of the peaks indicating the changes in the color values further comprises:
applying a Laplacian filter to the color channel data to generate data indicating a rate of change of the linear color gradient for each color channel;
identifying a first set of locations of the peaks above a threshold value;
applying a clustering algorithm to the first set of locations of the peaks; and
generating a second set of locations of the peaks by eliminating locations from the first set of locations that are co-linear to neighboring locations.

6. The method of claim 5 , wherein the threshold value for a color channel is calculated as a median value of all peaks in the color channel data for a corresponding color channel.

7. The method of claim 1 , wherein determining the vector representing the direction of the linear color gradient comprises:
generating color channel data for the raster image, the color channel data including vector fields for each color channel indicating directions of change of a corresponding color;
aggregating the vector fields for each color channel to determine the vector representing the direction of the linear color gradient.

8. A non-transitory computer-readable storage medium storing executable instructions, which when executed by a processing device, cause the processing device to perform operations comprising:
receiving, by a processing device, a raster image, the raster image including a representation of a linear color gradient;
determining a vector representing a direction of the linear color gradient, wherein the direction of the linear color gradient is a combination of a direction of each color channel of a pixel in the linear color gradient;
in response to determining that the direction of each color channel of the pixel in the linear color gradient is aligned, analyzing pixel points along the direction of the linear color gradient to compute color stops of the linear color gradient; and
generating an output color gradient vector with the computed color stops of the linear color gradient, the output color gradient vector to be applied to a vector graphic.

9. The non-transitory computer-readable storage medium of claim 8 , wherein to analyze each pixel point along the direction of the linear color gradient to compute the color stops of the linear color gradient the instructions further cause the processing device to perform operations comprising:
determining average pixel column color values of the linear color gradient;
generating color channel data for each pixel point along the vector representing the direction of the linear color gradient from the determined average pixel column color values; and
identifying the color stops along the vector representing the direction of the linear color gradient at locations of peaks in the generated color channel data, the locations of peaks indicating changes in color values.

10. The non-transitory computer-readable storage medium of claim 9 , wherein to determine the average pixel column color values of the linear color gradient the instructions further cause the processing device to perform operations comprising:
rotating the raster image to align the vector representing the direction of the linear color gradient with a first axis;
for each pixel point along the first axis,
determining color values associated with pixels in a pixel column parallel to a second axis, the second axis perpendicular to the first axis, and
calculating an average pixel column color value for the pixel column based on the determined color values; and

generating the average pixel column color values of the linear color gradient by aggregating the average pixel column color value for each pixel point.

11. The non-transitory computer-readable storage medium of claim 9 , wherein to determine the average pixel column color values of the linear color gradient the instructions further cause the processing device to perform operations comprising:
for each pixel point along the vector representing the direction of the linear color gradient,
determining color values associated with pixels in a pixel column perpendicular to the vector representing the direction of the linear color gradient, and
calculating an average pixel column color value for the pixel column based on the determined color values; and

generating the average pixel column color values of the linear color gradient by aggregating the average pixel column color value for each pixel point.

12. The non-transitory computer-readable storage medium of claim 9 , wherein to identify the color stops along the vector representing the direction of the linear color gradient at the locations of the peaks indicating the changes in the color values the inst

Assignees

• Adobe Inc

Inventors

• Chakraborty Souymodip
• Batra Vineet
• Lukác Michal
• Fisher Matthew David
• Phogat Ankit

Classifications

• G06T5/20

  using local operators · CPC title

• G06T2207/20076

  Probabilistic image processing · CPC title

• G06T2207/10024

  Color image · CPC title

• G06T3/60 Primary

  Rotation of whole images or parts thereof · CPC title

• G06T7/90 Primary

  Determination of colour characteristics · CPC title