Complex architecture for image processing

• US 11,037,275 B2
• Filed: 05/29/2018
• Issued: 06/15/2021
• Est. Priority Date: 03/17/2016
• Status: Active Grant

• Current Assignee: Tencent Technology Company Limited (Tencent Holdings Limited)
• Original Assignee: Tencent Technology Company Limited (Tencent Holdings Limited)
• Inventors: Wu, Lei
• Primary Examiner(s): Krasnic, Bernard
• Application Number:
  US15/991,818
• Publication Number:
  US 20180276797A1
• Time in Patent Office: 1,113 Days
• Field of Search: None
• CPC Class Codes: G06T 11/001 Texturing; Colouring; Generation of texture or colour inpainting G06T5/77
• CPC Class Codes: G06T 11/60 Editing figures and text; Combining figures or text
• CPC Class Codes: G06T 15/503 Blending, e.g. for anti-aliasing
• CPC Class Codes: G06T 5/50 using two or more images, e.g. averaging or subtraction
• CPC Class Codes: G06T 5/77 Retouching; Inpainting; Scratch removal
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/90 Determination of colour characteristics

Abstract

A system allows for editing images. The system accepts a first image including multiple pixels with corresponding pixel information as an input. The system divides the first image into multiple segmented regions based on the pixel information. The system identifies a segmented image from a first segmented region of the segmented regions. The system accepts a template image comprising an image rendering parameter as another input. The system generates a second image corresponding to the first image based on the template image and the segmented image.

Claims

1. An image processing method for a terminal having a processor and a storage medium having a program stored therein, and a user interface, the method comprising:

inputting, into the terminal via the user interface, a first image including multiple pixels with corresponding pixel information;

dividing, by the terminal, the first image into multiple segmented regions based on the pixel information;

identifying, by the terminal, a segmented image from a first segmented region of the multiple segmented regions;

selecting, by the terminal, a template image comprising an image rendering parameter;

converting the first segmented region of the first image to a first intermediate image corresponding to grayscale information of the first segmented region of the first image;

converting the first segmented region of the first image from RGBA (red, green, blue, alpha) format to HSV (hue, saturation, value) format;

extracting a second intermediate image corresponding to value (V) information of the converted first segmented region of the first image;

generating a third intermediate image by performing histogram equalization on the second intermediate image based on the segmented image; and

generating, by the terminal, a second image corresponding to the first image by;

generating a screen result image by performing screen blending on the first image and the first intermediate image; and

generating a soft light result image by performing soft light blending on at least the screen result image and the third intermediate image.

2. The method according to claim 1, further comprising displaying, on a screen of the terminal, the second image.

3. The method according to claim 1, wherein dividing the first image into the multiple segmented regions comprises dividing the first image based on a maximum flow minimum cut algorithm.

4. The method according to claim 1, wherein the method further comprises:

generating a fourth intermediate image by performing scaling on the template image based on the first segmented region; and

where generating the second image further comprises alpha blending on the soft light result image and the fourth intermediate image.

5. The method according to claim 1 further including establishing a Gaussian mixture model for the first image.

6. The method according to claim 5 further including:

using the Gaussian mixture model to represent pixel information for each pixel point of the first image; and

applying a maximum flow minimum cut algorithm based on the pixel information to determine the multiple segmented regions.

7. A terminal comprising an input circuitry and processing circuitry:

the input circuitry is configured to;

input, into the terminal, a first image including multiple pixels with corresponding pixel information; and

input, into the terminal, a template image, wherein the template image includes an image rendering parameter;

the processing circuitry is configured to;

divide the first image into multiple segmented regions based on the pixel information;

identify a segmented image from a first segmented region of the multiple segmented regions;

convert the first segmented region of the first image to a first intermediate image corresponding to grayscale information of the first segmented region of the first image;

convert the first segmented region of the first image from RGBA (red, green, blue, alpha) format to HSV (hue, saturation, value) format;

extract a second intermediate image corresponding to value (V) information of the converted first segmented region of the first image;

generate a third intermediate image by performing histogram equalization on the second intermediate image based on the segmented image; and

generate a second image corresponding to the first image by;

generating a screen result image by performing screen blending on the first image and the first intermediate image; and

generating a soft light result image by performing soft light blending on at least the screen result image and the third intermediate image.

8. The terminal according to claim 7, wherein:

generating the second image further comprises alpha blending based on the template image and the soft light result image.

9. The terminal according to claim 7, wherein:

the processing circuitry is further configured to divide the first image into the multiple segmented regions based on a maximum flow minimum cut method.

10. The terminal according to claim 7, wherein:

the processing circuitry is further configured to;

generate a fourth intermediate image by performing scaling on the template image based on the first segmented region; and

generate the second image further by performing alpha blending on the soft light result image and the fourth intermediate image.

11. The terminal according to claim 7, further comprising a screen configured to display the second image.

12. The terminal according to claim 7, wherein the processing circuitry is further configured to establish a Gaussian mixture model for the first image.

13. The terminal according to claim 12, wherein the processing circuitry is further configured to:

use the Gaussian mixture model to represent pixel information for each pixel point of the first image; and

apply a maximum flow minimum cut algorithm based on the pixel information to determine the multiple segmented regions.

14. A non-transitory machine readable storage media, storing computer program instructions which when executed by at least one processor perform:

inputting a first image including multiple pixels with corresponding pixel information;

dividing the first image into multiple segmented regions based on the pixel information;

identifying a segmented image from a first segmented region of the multiple segmented regions;

converting the first segmented region of the first image to a first intermediate image corresponding to grayscale information of the first segmented region of the first image;

converting the first segmented region of the first image from RGBA (red, green, blue, alpha) format to HSV (hue, saturation, value) format;

extracting a second intermediate image corresponding to value (V) information of the converted first segmented region of the first image;

generating a third intermediate image by performing histogram equalization on the second intermediate image based on the segmented image;

inputting a template image comprising an image rendering parameter; and

generating a second image corresponding to the first image by;

generating a screen result image by performing screen blending on the first image and the first intermediate image; and

generating a soft light result image by performing soft light blending on at least the screen result image and the third intermediate image.

15. The non-transitory machine readable storage media according to claim 14, wherein generating the second image further comprises alpha blending based on the template image and the soft light result image.

16. The non-transitory machine readable storage media according to claim 15, wherein dividing the first image into the multiple segmented regions comprises dividing the first image based on a maximum flow minimum cut algorithm.

17. The non-transitory machine readable storage media according to claim 14, wherein the at least one processor is further configured to perform:

generating a fourth intermediate image by performing scaling on the template image based on the first segmented region of the first image; and

generating the second image further by performing alpha blending on the soft light result image and the fourth intermediate image.

18. The non-transitory machine readable storage media according to claim 14, wherein the at least one processor is further configured to perform:

displaying, on a screen of a terminal, the second image.

19. The non-transitory machine readable storage media according to claim 14, wherein the at least one processor is further configured to establish a Gaussian mixture model for the first image.

20. The non-transitory machine readable storage media according to claim 19, wherein the at least one processor is further configured to:

use the Gaussian mixture model to represent pixel information for each pixel point of the first image; and

apply a maximum flow minimum cut algorithm based on the pixel information to determine the multiple segmented regions.