Wrinkle detection apparatus and wrinkle detection method

Patent metadata

Field	Value
Publication number	US-9443131-B2
Application number	US-201514715309-A
Country	US
Kind code	B2
Filing date	May 18, 2015
Priority date	Jun 9, 2014
Publication date	Sep 13, 2016
Grant date	Sep 13, 2016

Abstract

A wrinkle detection apparatus is an apparatus for detecting a wrinkle area of skin included in an image. The wrinkle detection apparatus includes: an image obtaining unit that obtains the image including the skin; an area estimation unit that estimates a plurality of image areas, each of the plurality of image areas having a different gloss level of the skin; a parameter determination unit that determines one or more parameter values for each of the plurality of estimated image areas, the one or more parameter values being used to detect the wrinkle area; and a wrinkle detection processing unit that detects the wrinkle area in the image by using the determined one or more parameter values for each of the plurality of image areas.

First claim

What is claimed is:

1. A wrinkle detection apparatus comprising:
a memory that stores instructions; and
a processor that, when executing the instructions stored in the memory, performs operations comprising:
obtaining an image including skin;
estimating a plurality of image areas in the image, each of the plurality of image areas belonging to either a first area or a second area other than the first area, the first area extending from a lower eyelid of a left eye to above a left cheek and extending from a lower eyelid of a right eye to above a right cheek;
determining first parameter values for each of the estimated image areas belonging to the first area and second parameter values for each of the estimated image areas belonging to the second area,
the first parameter values including a first threshold value, a first filter angle for a Gabor filter, and a first filter wavelength for the Gabor filter,
the second parameter values including a second threshold value, a second filter angle for the Gabor filter, and a second filter wavelength for the Gabor filter, and
the first threshold value being smaller than the second threshold value, the first filter angle having a narrower angle range than the second filter angle, and the first filter wavelength being shorter than the second filter wavelength; and

detecting a wrinkle area in the image by calculating first gradient values in the first area by using the Gabor filter adopting the first filter angle and the first filter wavelength, calculating second gradient values in the second area by using the Gabor filter adopting the second filter angle and the second filter wavelength, comparing each of the first gradient values with the first threshold value, and comparing each of the second gradient values with the second threshold value.

2. A computer-implemented wrinkle detection method comprising:
obtaining, by a processor, an image including skin;
estimating, by a processor, a plurality of image areas in the image, each of the plurality of image areas belonging to either a first area or a second area other than the first area, the first area extending from a lower eyelid of a left eye to above a left cheek and extending from a lower eyelid of a right eye to above a right cheek;
determining, by a processor, first parameter values for each of the estimated image areas belonging to the first area and second parameter values for each of the estimated image areas belonging to the second area,
the first parameter values including a first threshold value, a first filter angle for a Gabor filter, and a first filter wavelength for the Gabor filter,
the second parameter values including a second threshold value, a second filter angle for the Gabor filter, and a second filter wavelength for the Gabor filter, and
the first threshold value being smaller than the second threshold value, the first filter angle having a narrower angle range than the second filter angle, and the first filter wavelength being shorter than the second filter wavelength; and

detecting, by a processor, the wrinkle area in the image by calculating first gradient values in the first area by using the Gabor filter adopting the first filter angle and the first filter wavelength, calculating second gradient values in the second area by using the Gabor filter adopting the second filter angle and the second filter wavelength, comparing each of the first gradient values with the first threshold value, and comparing each of the second gradient values with the second threshold value.

3. The wrinkle detection apparatus according to claim 1 , wherein
each of the first gradient values and the second gradient values indicates a degree of change in a pixel value in the image,
in the first gradient values, a gradient value for an edge area is bigger than a gradient value for a non-edge area, and
in the second gradient values, a gradient value for an edge area is bigger than a gradient value for a non-edge area.

4. The wrinkle detection method according to claim 2 , wherein
each of the first gradient values and the second gradient values indicates a degree of change in a pixel value in the image,
in the first gradient values, a gradient value for an edge area is bigger than a gradient value for a non-edge area, and
in the second gradient values, a gradient value for an edge area is bigger than a gradient value for a non-edge area.

Assignees

• Panasonic Ip Man Co Ltd

Inventors

• Yamanashi Tomofumi

Classifications

• G06V40/16 Primary

  Human faces, e.g. facial parts, sketches or expressions · CPC title

• G06T5/10

  using non-spatial domain filtering · CPC title

• G06K9/6202

  Physics · mapped topic

• G06K9/00221 Primary

  Physics · mapped topic

• G06K9/00

  Physics · mapped topic