Hue-based color naming for an image

• US 9,785,649 B1
• Filed: 09/02/2014
• Issued: 10/10/2017
• Est. Priority Date: 09/02/2014
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Dorner, Charles Shearer, Blackburn, Jenny Ann, Manolis, Eva, Ong, Timothy Andrew, Sayre, III, Paul Barnhart
• Primary Examiner(s): Krasnic, Bernard
• Application Number:
  US14/475,401
• Time in Patent Office: 1,134 Days
• Field of Search: None
• CPC Class Codes: G06F 16/54 Browsing; Visualisation therefor
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 10/56 relating to colour

Abstract

Systems and methods are provided for associating colors or color names with a color image based on hue. A set of hue values can be extracted from color data, such as a color palette, corresponding to a color image. A representative hue can be derived from the set of hue values. Colors included in the color data that are attributable to the representative hue can be identified and one or more colors can be derived therefrom as representative of a subject depicted by the color image. Color names matching the representative colors can be identified and associated with the color image. In particular, a common color name can be determined therefrom and associated with the subject of the color image.

Claims

1. A computer-implemented method for determining color names for a color image based at least in part on hue, the computer-implemented method comprising:

under control of a hardware computing device configured with specific computer executable instructions,

obtaining color data corresponding to a color image;

generating a hue distribution from the color data, wherein the hue distribution comprises a plurality of distinct hue values each associated with a hue weight;

modifying the hue distribution based, at least in part, on merging two or more hue values in the hue distribution in accordance with respective hue weights associated with the two or more hue values and in accordance with a hue distance, wherein the hue distance is defined as the shortest arc or smallest angle between the two or more hue values on a corresponding color circle;

determining a representative hue based, at least in part, on the modified hue distribution;

identifying one or more color values from the color data, wherein the identified one or more color values are attributable to the representative hue;

determining a representative color from the identified one or more color values;

identifying one or more color names corresponding to the representative color; and

associating the one or more color names with the color image.

2. The computer-implemented method of claim 1, wherein the color data includes a color palette generated based, at least in part, on the color image.

3. The computer-implemented method of claim 2, wherein the color palette includes a plurality of palette color values in a color space, wherein each palette color value of the plurality of palette color values is associated with a palette color weight.

4. The computer-implemented method of claim 3, wherein the one or more color values are identified from the plurality of palette color values.

5. The computer-implemented method of claim 1, wherein the color data includes RED GREEN BLUE (“

RGB”

) color values represented in an RGB color space and wherein generating the hue distribution comprises at least partially converting representation of the color values from the RGB color space to a Hue-Saturation-Brightness (“

HSB”

) color space.

6. A computer-implemented method comprising:

under control of a hardware computing device configured with specific computer executable instructions,

generating a hue distribution from color data corresponding to a color image, wherein the hue distribution comprises a plurality of distinct hue values each associated with a hue weight;

modifying the hue distribution based, at least in part, on merging two or more hue values in the hue distribution in accordance with respective hue weights associated with the two or more hue values and in accordance with a hue distance, wherein the hue distance is defined as the shortest arc or smallest angle between the two or more hue values on a corresponding color circle;

determining a representative hue based, at least in part, on the modified hue distribution;

determining a representative color from the color data based, at least in part, on the representative hue; and

associating one or more color names with the color image based, at least in part, on the representative color.

7. The computer-implemented method of claim 6, wherein the representative hue corresponds to at least one of a hue value or a range of hue values.

8. The computer-implemented method of claim 6 further comprising:

binning the plurality of hue values in accordance with a set of standardized hue values;

wherein the generating the hue distribution is based, at least in part, on the binned plurality of hue values.

9. The computer-implemented method of claim 8, wherein the binning of the plurality of hue values is consistent for a set of color images based, at least in part, on metadata associated with the set of color images.

10. The computer-implemented method of claim 6, wherein the representative hue is further determined based, at least in part, on at least one of a dominance corresponding to the modified hue distribution or a concentration corresponding to the modified hue distribution.

11. A system comprising:

a data store configured to at least store computer-executable instructions; and

a hardware processor in communication with the data store, the hardware processor configured to execute the computer-executable instructions to at least;

generate a hue distribution from color data corresponding to a color image, wherein the hue distribution comprises a plurality of distinct hue values each associated with a hue weight;

modify the hue distribution based, at least in part, on merging two or more hue values in the hue distribution in accordance with respective hue weights associated with the two or more hue values and in accordance with a hue distance, wherein the hue distance is defined as the shortest arc or smallest angle between the two or more hue values on a corresponding color circle;

determine a first representative hue based, at least in part, on the modified hue distribution;

determine a first representative color from the color data based, at least in part, on the first representative hue; and

associate one or more color names with the color image based, at least in part, on the first representative color.

12. The system of claim 11, wherein the hardware processor is further configured to execute the computer-executable instructions to at least identify at least one second representative hue based, at least in part, on the modified hue distribution.

13. The system of claim 12, wherein the hardware processor is further configured to execute the computer-executable instructions to at least generate a hue palette corresponding to the color image, wherein the hue palette includes the first and second representative hues.

14. The system of claim 11, wherein the hardware processor is further configured to execute computer-executable instructions to at least determine at least one second representative color from the color data based, at least in part, on the first representative hue.

15. The system of claim 14, wherein the at least one second representative color is determined in accordance with a divergence measure of color values included in the color data that are attributable to the first representative hue.

16. A non-transitory computer readable storage medium storing computer executable instructions that when executed by a processor perform operations comprising:

generating a hue distribution from color data corresponding to a color image, wherein the hue distribution comprises a plurality of distinct hue values each associated with a hue weight;

modifying the hue distribution based, at least in part, on merging two or more hue values in the hue distribution in accordance with respective hue weights associated with the two or more hue values and in accordance with a hue distance, wherein the hue distance is defined as the shortest arc or smallest angle between the two or more hue values on a corresponding color circle;

determining a representative hue based, at least in part, on the modified hue distribution;

determining a representative color from the color data based, at least in part, on the representative hue; and

associating one or more color names with the color image based, at least in part, on the representative color.

17. The non-transitory computer readable storage medium of claim 16, wherein the color data corresponding to the color image includes a color palette corresponding to the color image.

18. The non-transitory computer readable storage medium of claim 16, wherein the operations further comprise determining a contextual criterion associated with the color image.

19. The non-transitory computer readable storage medium of claim 18, wherein the contextual criterion corresponds to at least one of a subject depicted by the color image, category of the subject depicted, source contributing to the color image, geographical region, age, gender, ethnic group, religion, or culture.

20. The non-transitory computer readable storage medium of claim 18, wherein at least one of determining the representative hue, determining the representative color, or associating the one or more color names with the color image is based, at least in part, on the contextual criterion.