Processing an image of a person's face

• US 8,559,712 B2
• Filed: 09/13/2010
• Issued: 10/15/2013
• Est. Priority Date: 09/13/2010
• Status: Expired due to Fees

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Bloom, Daniel, Staudacher, David, Dalton, Dan L.
• Primary Examiner(s): Krasnic, Bernard
• Application Number:
  US12/880,524
• Publication Number:
  US 20120063648A1
• Time in Patent Office: 1,128 Days
• Field of Search: 382/103, 382/162, 382/167, 382/274
• US Class Current: 382/167
• CPC Class Codes: G06V 40/165 using facial parts and geometric relationships

Abstract

An image contains a representation of a person'"'"'s face, where the image has plural points each comprising multiple components of a color space. For each of the plural points, at least two of the color space components are combined to produce a respective aggregate value. The image is processed to convert at least some of the points of the image based on the calculated aggregate values.

Claims

1. A method comprising:

receiving an image containing a representation of a person'"'"'s face, where the face has a target feature, wherein the image has points each including a Cb component and a Cr component in a YCbCr space;

for each of the points of the image, combining, by a system having a processor, values of the Cb and Cr components to provide an aggregate value for the corresponding point, wherein combining the values comprises calculating a product based on the Cb and Cr components;

lightening midtones of values of a Y component in the YCbCr space to provide pre-processed values of the Y component;

darkening midtones of the aggregate values derived from calculating the product based on the Cb and Cr components to provide pre-processed aggregate values;

combining the pre-processed values of the Y component with the pre-processed aggregate values; and

producing a converted image based on values for the corresponding points derived from combining the pre-processed values of the Y component and the pre-processed aggregate values.

2. The method of claim 1, wherein producing the converted image comprises producing the converted image that highlights lips or a mouth of the person'"'"'s face.

3. The method of claim 1, wherein providing the aggregate values for the corresponding points comprises providing the aggregate values for corresponding pixels of the image.

4. An article of manufacture comprising at least one non-transitory machine-readable storage medium storing instructions that upon execution cause a system having a processor to:

receive an image containing a representation of a person'"'"'s face, where the face has a target feature, wherein the image has points each including a Cb component and a Cr component in a YCbCr space;

for each of the points of the image, combine values of the Cb and Cr components to provide an aggregate value for the corresponding point, wherein combining the values comprises calculating a product based on the Cb and Cr components;

lighten midtones of the values of a Y component in the YCbCr space to provide pre-processed values of the Y component;

darken midtones of the aggregate values derived from calculating the product based on the Cb and Cr components to provide pre-processed aggregate values;

combine the pre-processed values of the Y component with the pre-processed aggregate values; and

produce a converted image based on values for the corresponding points derived from combining the pre-processed values of the Y component and the pre-processed aggregate values.

5. The article of manufacture of claim 4, wherein producing the converted image comprises:

clipping a predefined portion of the points of the image corresponding to regions of the face other than the target feature.

6. The article of manufacture of claim 5, wherein clipping the predefined portion of the points comprises setting the aggregate values for the predefined portion of the points to a predefined value to highlight a difference of the regions from the target feature.

7. An electronic device comprising:

at least one processor; and

machine-readable instructions executable on the at least one processor to;

receive an image containing a representation of a person'"'"'s face, wherein the image has plural pixels each comprising multiple components of a YCbCr space;

for each of the plural pixels, calculate a product based on a value of the Cb component and a value of the Cr component to produce a respective aggregate value; and

process the image to convert at least some of the pixels of the image based on the calculated aggregate values, to form a converted image that highlights lips or a mouth of the person'"'"'s face, wherein the converting includes inverting the calculated aggregate values for the at least some of the pixels, wherein processing the image to form the converted image comprises;

modifying the aggregate values of a predefined portion of the pixels of the image, the modifying including the inverting; and

after the modifying, combining the aggregate values of the pixels of the image with values of a Y component of the YCbCr space associated with the pixels;

pre-processing the aggregate values and the values of the Y component, wherein pre-processing the aggregate values comprises darkening midtones of the aggregate values, and wherein pre-processing the values of the Y component comprises lightening midtones of the values of the Y component,

wherein combining the aggregate values and values of the Y component comprises combining the pre-processed aggregate values and the pre-processed values of the Y component.

8. An electronic device comprising:

at least one processor; and

machine-readable instructions executable on the at least one processor to;

receive an image containing a representation of a person'"'"'s face, where the face has a target feature, wherein the image has points each including a Cb component and a Cr component in a YCbCr space;

for each of the points of the image, combine values of the Cb and Cr components to provide an aggregate value for the corresponding point, wherein combining the values comprises calculating a product based on the Cb and Cr components;

lighten midtones of the values of a Y component in the YCbCr space to provide pre-processed values of the Y component;

darken midtones of the aggregate values derived from calculating the product based on the Cb and Cr components to provide pre-processed aggregate values;

combine the pre-processed values of the Y component with the pre-processed aggregate values; and

produce a converted image based on values for the corresponding points derived from combining the pre-processed values of the Y component and the pre-processed aggregate values.