Component specific image modification using natural language color

Inventors: Zhigang Fan (Webster, NY); Karen Marie Braun (Fairport, NY); Elizabeth Dubiel Wayman (Ontario, NY)

Assignee: Xerox Corporation

G06T11/001 G06F17/20

View Patent ↗

Loading inventors, assignments & file history…

Monitor This Case

Get email alerts when status or documents change.

Order Certified Copies

Most orders are placed with the USPTO same day — all within 24 business hours.

Order via The Patent Place →

Pre-filled with this patent's details

Quick Facts

Patent No.: US 9,076,251
App. No.: 12/850,238
Granted: Jul 7, 2015
Kind: B2

External Links

USPTO Patent Center↗ Full Patent PDF↗

Abstract

Embodiments described herein include a component specific natural language color unit for modifying colors of image components in an image using component specific natural language commands. The CSNLC unit can receive a component specific natural language color command for a color modification of an input image. The component specific natural language color command can include a component identifier and a color modifier. The natural language command can be parsed to attribute the component identifier to an image component in the input image and to attribute the color modifier to a predefined color space associated with the image component. The color modifier can indicate the color modification to be performed. The color modification can be applied to the image component to adjust the color of the image component.

Claims (60)

1. A method for modifying color in an image using component specific natural language color commands comprising:

identifying an input image to be modified;

segmenting the input image into a plurality of image components, the image components being discrete segments of the input image;

assigning each of the plurality of image components a component identity based on content of the image component using pattern recognition, object recognition, and boundaries analysis;

classifying the plurality of image components into a plurality of component types;

receiving a component specific natural language color command for a color modification of the input image, the component specific natural language color command including a component type identifier and a color modifier;

parsing the component specific natural language color command to obtain the component type identifier and the color modifier contained in the component specific natural language color command;

identifying, without user input, the one or more component types corresponding to the parsed component type identifier;

identifying all of the image components classified with each of the identified component types;

attributing the color modifier to a predefined color space associated with each of the identified image components, the color modifier indicating the color modification to be performed; and

applying the color modification to the identified image components to adjust the color of the identified image components.

2. The method of claim 1 , wherein applying the color modification is performed without affecting the plurality of image components other than the identified at least one image component in the image.

3. The method of claim 1 , further comprising:

applying a command syntax to the component specific natural language color command.

4. The method of claim 3 , wherein the command syntax comprises a subject image component, a color, and a subject modifier.

5. The method of claim 3 , wherein the command syntax comprises a magnitude, a direction, and a color.

6. The method of claim 3 , wherein the command syntax applied is retrieved from a dictionary of multiple dictionaries.

7. The method of claim 6 , wherein the dictionary includes selectively one or more of the following color names: black, white, red, green, yellow, blue, brown, pink, orange, purple, and gray.

8. The method of claim 7 , wherein the dictionary further includes compound color names.

9. The method of claim 8 , wherein each compound color names are stored in the dictionary as a single-word color name.

10. The method of claim 7 , wherein the dictionary further includes color modifiers for each color name.

11. The method of claim 7 , wherein each color name in the dictionary corresponds with exactly one finite region of color space in the dictionary.

12. The method of claim 6 , wherein each dictionary of the multiple dictionaries is mapped individually onto color space regions and each dictionary of the multiple dictionaries is associated with a user or group of users associated with a unique command syntax.

13. The method of claim 1 , further comprising:

mapping the component identifier in the component specific natural language color command to the identified at least one image component in the image; and

mapping the color modifier to a color in the color space.

14. The method of claim 1 , wherein assigning each of the plurality of image components a component identifier further includes indentifying a main background, identifying at least one image component disposed on the main background, identifying a local background, identifying at least one image component disposed on the local background, identifying at least one graphic, identifying at least one image, and identifying text as selectively one of normal sized text and small sized text.

15. The method of claim 1 , wherein the input image comprises a computer document.

16. The method of claim 1 , wherein classifying the plurality of image components into a plurality of component types comprises identifying image components as selectively one of text, pictures, and graphics.

17. A non-transient computer readable storage medium storing instructions executable by a computing system including at least one computing device, wherein execution of the instructions implements a method for modifying color in an image using component specific natural language color commands comprising: