Method and apparatus for creating a color name dictionary and for querying an image by color name

• US 6,081,276 A
• Filed: 11/14/1996
• Issued: 06/27/2000
• Est. Priority Date: 11/14/1996
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Delp, Helen R.
• Primary Examiner(s): Nguyen, Phu K.
• Assistant Examiner(s): Vo, Cliff N.
• Application Number:
  US08/749,150
• Time in Patent Office: 1,321 Days
• Field of Search: 345/431, 345/418, 345/419, 345/429, 345/150, 345/152, 345/153
• US Class Current: 345/589
• CPC Class Codes: G01J 3/46 Measurement of colour; Colour measuring devices, e.g. colorimeters measuring colour temperature G01J5/60
• CPC Class Codes: G01J 3/462 Computing operations in or between colour spaces; Colour management systems
• CPC Class Codes: G06F 16/5838 using colour

Abstract

A method and apparatus are provided for creating a color name dictionary. A predetermined color space is identified. The predetermined color space is divided into a plurality of color space segments. A color name is assigned to each of the plurality of color space segments. A method and apparatus are provided for querying an image by color name. An image is acquired from an image source, such as an image database. A pixel is acquired from the acquired image. The acquired pixel is compared with a color name volume. A color name volume containing the acquired pixel is identified.

Claims

1. A method for creating a color name dictionary comprising the steps of:

identifying a predetermined color space;

dividing said predetermined color space into a plurality of color space segments;

assigning an associated color name to each of said plurality of color space segments;

each said associated color name including a word or a phrase that constitutes a distinctive designation for said associated color space segment.

2. A method for creating a color name dictionary as recited in claim 1 wherein the step of identifying a predetermined color space includes the step of identifying a three-dimensional predetermined color space.

3. A method for creating a color name dictionary as recited in claim 2 wherein the step of dividing said predetermined color space into a plurality of color space segments includes the step of dividing a first dimension of said three-dimensional predetermined color space into a predefined number of hue segments.

4. A method for creating a color name dictionary as recited in claim 3 further includes the step of dividing a second dimension of said three-dimensional predetermined color space into a predefined number of chroma segments.

5. A method for creating a color name dictionary as recited in claim 4 further includes the step of dividing a third dimension of said three-dimensional predetermined color space into a predefined number of lightness segments.

6. A method for creating a color name dictionary as recited in claim 1 wherein the step of assigning a color name to each of said plurality of color space segments includes the step of storing a color name with boundary values defining each of said plurality of color space segments.

7. A method for creating a color name dictionary as recited in claim 1 wherein the step of dividing said predetermined color space into a plurality of color space segments includes the steps of identifying a set of color name boundaries in a three-dimensional color space;

said identified set of color name boundaries including six boundary values to define concentric cylindrical segments.

8. A method for creating a color name dictionary as recited in claim 7 wherein the step of identifying a set of color name boundaries in a three-dimensional color space includes the steps of defining a color name volume by a low radius boundary value and a high radius boundary value, a low angle boundary value in the X-Y plane and a high angle boundary value in the X-Y plane, and a low boundary value Z coordinate and a high boundary value Z coordinate.

9. A method for creating a color name dictionary as recited in claim 1 wherein the step of dividing said predetermined color space into a plurality of color space segments includes the steps of identifying a set of color name boundaries in a three-dimensional color space;

said identified set of color name boundaries including six boundary values to define spherical segments.

10. A method for creating a color name dictionary as recited in claim 1 wherein the step of dividing said predetermined color space into a plurality of color space segments includes the steps of identifying a set of color name boundaries in a three-dimensional color space;

said identified set of color name boundaries including six boundary values to define curvilinear segments.

11. A method for creating a color name dictionary as recited in claim 1 wherein the step of identifying a predetermined color space includes the step of identifying a generally visually uniform color space.

28. A method for querying a point in color space by color name as recited in claim 3 includes the steps of acquiring a next pixel within an image region and repeating said comparing and identifying steps;

and incrementing a count value for said identified color name.

12. A method for querying an image by color name comprising the steps of:

acquiring an image;

acquiring a pixel from said acquired image;

comparing said acquired pixel with a color name volume;

identifying at color name volume containing said acquired pixel; and

identifying a color name for said identified color name volume;

said color name including a word or a phrase that constitutes a distinctive designation for said identified color name volume.

13. A method for querying an image by color name as recited in claim 12 wherein said step of acquiring a pixel from said acquired image includes the step of transforming pixel coordinates from a device-dependent color space into a device-independent color space of a color name dictionary.

14. A method for querying an image by color name as recited in claim 12 wherein said step of comparing said acquired pixel with a color name volume includes the steps of acquiring volume coordinate values for said color name volume.

15. A method for querying an image by color name as recited in claim 12 wherein said step of comparing said acquired pixel with a color name volume includes the steps of receiving pixel coordinates in a device-independent color space of a color name dictionary and comparing said received pixel coordinates with boundary values of said color name volume.

16. A method for querying an image by color name as recited in claim 12 further includes the step of incrementing a pixel count for an identified color name volume.

17. A method for querying an image by color name as recited in claim 16 further includes the step of generating a color name histogram.

18. A method for querying an image by color name as recited in claim 17 further includes the step of sorting color name records contained in said color name histogram responsive to a predefined type of user query.

19. A method for querying an image by color name as recited in claim 17 further includes the step of sorting color name records contained in said color name histogram by a fractional count responsive to a predefined type of user query.

20. A method for querying an image by color name as recited in claim 17 wherein said step of acquiring a pixel from said acquired image includes the steps of identifying an image region within said acquired image.

21. A method for querying an image by color name as recited in claim 20 further includes the step of sorting color name records contained in said color name histogram responsive to a predefined type of user query.

22. A method for querying an image by color name as recited in claim 20 further includes the step of sorting color name records contained in said color name histogram by said image region and fractional count responsive to a predefined type of user query.

23. A computer program product for use in a computer system having a central processor for creating a color name dictionary, the computer program product comprising:

a recording medium;

means, recorded on the recording medium, for identifying a predetermined color space;

means, recorded on the recording medium, for dividing said predetermined color space into a plurality of color space segments;

means, recorded on the recording medium, for assigning an associated color name to each of said plurality of color space segments;

each said associated color name including a word or a phrase that constitutes a distinctive designation for said associated color space segment.

24. A computer program product for use in a computer system as recited in claim 23 wherein said means, recorded on the recording medium, for dividing said predetermined color space into a plurality of color space segments include means, recorded on the recording medium, for identifying a set of color name boundaries in a three-dimensional color space;

said identified set of color name boundaries including six boundary values to define concentric cylindrical segments.

25. A computer program product for use in a computer system having a central processor for querying an image by color name, the computer program product comprising:

a recording medium;

means, recorded on the recording medium, for acquiring an image;

means, recorded on the recording medium, for acquiring a pixel from said acquired image;

means, recorded on the recording medium, for comparing said acquired pixel with a color name volume;

means, recorded on the recording medium, for identifying a color name volume containing said acquired pixel; and

means, recorded on the recording medium, for identifying a color name for said identified color name volume;

said color name including a word or a phrase that constitutes a distinctive designation for said identified color name volume.

26. A computer program product for use in a computer system as recited in claim 25 further includes means, recorded on the recording medium, for generating a color name histogram.

27. A method for querying a point in color space by color name comprising the steps of:

acquiring a pixel;

comparing said acquired pixel with a color name volume;

identifying a color name volume containing said acquired pixel; and

identifying a color name for said identified color name volume;

said color name including a word or a phrase that constitutes a distinctive designation for said identified color name volume.