Wrinkle detection method and terminal device

Patent metadata

Field	Value
Publication number	US-11978231-B2
Application number	US-201917295230-A
Country	US
Kind code	B2
Filing date	Nov 13, 2019
Priority date	Nov 19, 2018
Publication date	May 7, 2024
Grant date	May 7, 2024

Abstract

A wrinkle detection method includes: obtaining an original image, where the original image includes a face; adjusting a size of an ROI region on the original image to obtain at least two ROI images of different sizes, where the ROI region is a region in which a wrinkle on the face is located. A terminal device processes all the at least two ROI images of different sizes to obtain at least two binary images, where a white region in each binary image is a region in which a wrinkle is suspected to appear. The terminal device fuses the at least two binary images to obtain a final image, where a white region on the final image is recognized as a wrinkle.

First claim

What is claimed is:

1. A wrinkle detection method, comprises:
obtaining an original image, wherein the original image comprises a face;
adjusting a size of a region on the original image to obtain at least two region of interest (ROI) images of different sizes, wherein in the region a wrinkle on the face is located;
processing the at least two ROI images of different sizes to obtain at least two binary images, wherein a white region in each binary image of the at least two binary images is a region in which the wrinkle is suspected to appear, wherein the processing comprises:
for each ROI image of the at least two ROI images: setting a pixel on the ROI image to be black or white based on a comparison between a pixel value of the pixel and a preset pixel value, wherein the pixel is located in an image block determined based on a preset matrix, and the pixel value is calculated based on the preset matrix; and

fusing the at least two binary images to obtain a final image, wherein a white region in the final image is recognized as the wrinkle.

2. The method of claim 1 , wherein the processing comprises:
repeatedly performing the following steps for each ROI image:
covering the ROI image by using the preset matrix;
determining a pixel value of a pixel that is on the ROI image and that corresponds to each matrix element in the preset matrix;
determining a product of each matrix element and the pixel value of the pixel corresponding to each matrix element;
obtaining a sum of products corresponding to the matrix elements in the preset matrix, wherein the sum is a pixel value at a center position of an image block that is on the ROI image and that is covered by the preset matrix; and
setting the center position to black if the pixel value of the center position of the image block is greater than a preset pixel value, or setting, the center position to white if the pixel value of the center position of the image block is less than or equal to the preset pixel value.

3. The method of claim 1 , wherein the method further comprises:
deleting the white region from M images, wherein M is less than or equal to a preset value, when the M images of the at least two binary images have a white region at a same position.

4. The method of claim 1 , when the wrinkle is a nasolabial fold, wherein the method further comprises:
determining a region in which a beard is located on the final image;
determining n white regions that intersect the region in which the beard is located;
determining a ratio of a quantity of pixels in the region in which the beard is located in a first white region of the n white regions to a quantity of pixels in the first white region; and
deleting the first white region from the final image, when the ratio is greater than or equal to a preset ratio, wherein a remaining white region on the final image is recognized as a nasolabial fold, wherein the preset ratio is 1−n/m, and m is a preset fixed value.

5. The method of claim 1 , when the wrinkle is a nasolabial fold, wherein the method further comprises:
determining a coordinate position of a nose wing in the final image; and
deleting from the final image, a white region that is within a preset distance range from the coordinate position and whose length is greater than a preset length, wherein a remaining white region on the final image is recognized as a nasolabial fold.

6. The method of claim 1 , wherein the method further comprises:
converting the ROI image into a grayscale image;
horizontally adjusting the gray scale grayscale image; and
denoising the horizontally adjusted image.

7. The method of claim 1 , wherein the method further comprises:
determining an evaluation result y of the white regions based on the following formula:
y=w 1* x 1+ w 2* x 2+ w 3* x 3+ w 4* x 4+ w 5* x 5+ w 6* x 6+ b

wherein x1 represents an average width of the white regions, x2 represents an average length of the white regions, x3 represents an average internal and external color contrast of the white regions, x4 represents a ratio of a quantity of pixels of the white regions to a total quantity of pixels of the ROI image, x5 and x6 respectively represent a length and a width of a longest white region in the white regions, and b represents a bias.

8. The method of claim 1 , wherein the method further comprises:
displaying notification information in a viewfinder interface, wherein the notification information is used to notify a position of the wrinkle on the face.

9. The method of claim 1 , wherein the method further comprises:
comparing the wrinkle with a wrinkle in a prestored image; and
performing screen unlocking when the wrinkle is consistent with the wrinkle in the prestored image.

10. The method of claim 1 , wherein the method further comprises:
displaying a payment verification interface; and
comparing the wrinkle with a wrinkle in a prestored image; and
performing a payment process when the wrinkle is consistent with the wrinkle in the prestored image.

11. The method of claim 1 , wherein the method further comprises:
outputting notification information when no wrinkle is detected.

12. An electronic device, comprising:
a non-transitory memory comprising instructions; and
at least one processor coupled to the non-transitory memory, the instructions being executed by the at least one processor to cause the electronic device to:
obtain an original image, wherein the original image comprises a face;
adjust a size of a region on the original image to obtain at least two region of interest (ROI) images of different sizes, wherein in the region a wrinkle on the face is located;
process the at least two ROI images of different sizes to obtain at least two binary images, wherein a white region in each binary image of the at least two binary images is a region in which the wrinkle is suspected to appear, wherein the processing comprises:
for each ROI image of the at least two ROI images: setting a pixel on the ROI image to be black or white based on a comparison between a pixel value of the pixel and a preset pixel value, wherein the pixel is located in an image block determined based on a preset matrix, and the pixel value is calculated based on the preset matrix; and

fuse the at least two binary images to obtain a final image, wherein a white region on the final image is recognized as the wrinkle.

13. The electronic device of claim 12 , the instructions further cause the electronic device to:
repeatedly perform the following steps for each ROI image:
covering the ROI image by using the preset matrix;
determining a pixel value of a pixel that is on the ROI image and that corresponds to each matrix element in the preset matrix;
determining a product of each matrix element and the pixel value of the pixel corresponding to each matrix element;
obtaining a sum of products corresponding to the matrix elements in the preset matrix, wherein the sum is a pixel value at a center position of an image block that is on the ROI image and that is covered by the preset matrix; and
setting the center position to black if the pixel value of the center position of the image block is greater than a preset pixel value, or setting, the center position to white if the pixel value of the center position of the image block is less than or equal to the preset pixel value.

14. The electronic device of claim 12 , the instructions further cause the electronic device to:
delete the white region from M images, wherein M is less than or equal to a preset value, when the M images of the at least two binary images have a white region at a same position.

Assignees

• Honor Device Co Ltd

Inventors

• Hu Hongwei
• Dong Chen
• Ding Xin
• Gao Wenmei

Classifications

• G06T7/73 Primary

  using feature-based methods · CPC title

• G06T7/0012

  Biomedical image inspection · CPC title

• G06V10/25

  Determination of region of interest [ROI] or a volume of interest [VOI] · CPC title

• G06V40/166 Primary

  using acquisition arrangements · CPC title

• G06V40/171

  Local features and components; Facial parts (eye characteristics G06V40/18); Occluding parts, e.g. glasses; Geometrical relationships · CPC title