WO2018171493A1

WO2018171493A1 - Image processing method and device, and storage medium

Info

Publication number: WO2018171493A1
Application number: PCT/CN2018/079073
Authority: WO
Inventors: 钱梦仁
Original assignee: 腾讯科技（深圳）有限公司
Priority date: 2017-03-21
Filing date: 2018-03-15
Publication date: 2018-09-27
Also published as: CN107038715B; CN107038715A

Abstract

Disclosed in embodiments of the present invention are an image processing method and device. The method comprises: obtaining an original image; performing skin color detection on image pixels in a selected area of the original image to determine characteristic pixels in the selected area of the original image; calculating an average value of characteristic parameter values of the characteristic pixels; correcting a characteristic parameter value of each image pixel of the original image according to a difference value of a preset standard average value and the average value of the characteristic parameter values, the preset standard average value being a numerical value obtained by calculating according to characteristic parameter values of standard characteristic pixels of a preset standard image; and outputting the corrected original image.

Description

Image processing method, device and storage medium

The present application claims priority to Chinese Patent Application No. 200910171010.5, the entire disclosure of which is incorporated herein by reference. .

Technical field

The present invention relates to the field of image processing technologies, and in particular, to a method, an apparatus, and a storage medium for processing an image.

background

With the continuous development of social media, photos are gradually replacing text, becoming the main content way for users to record their lives. Among them, self-portrait life is more. Most of today's mobile phones, tablets and other terminal devices are equipped with a front camera, making self-timer more convenient and self-timer anytime, anywhere. When the existing camera technology processes the photos, the aperture is automatically adjusted according to the ambient light to obtain the exposure that matches the current ambient light.

Technical content

Embodiments of the present invention provide an image processing method and apparatus, which can reduce the probability of an image being overexposed or too dark, and can make the skin color in the image closer to the natural skin color of the user and improve the image quality.

An embodiment of the present invention provides an image processing method, including:

Get the original image;

Performing feature detection on image pixels of selected regions in the original image to determine feature pixels of the selected region in the original image;

Calculating an average value of characteristic parameter values of the feature pixels;

Correcting a feature parameter value of each image pixel of the original image according to a difference between a preset standard average value and an average value of the feature parameter values, the preset standard average value including an image according to a preset standard The value of the characteristic parameter value of the standard feature pixel;

The corrected original image is output.

An embodiment of the present invention further provides an image processing apparatus, the apparatus comprising a processor and a memory, the memory storing computer readable instructions, wherein the processor is:

Get the original image;

An output module for outputting the corrected original image.

The embodiment of the present application further provides a computer readable storage medium storing computer readable instructions, which may cause a processor to execute the methods of the embodiments.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the present invention. Other drawings can also be obtained from those skilled in the art based on these drawings without paying any creative effort.

FIG. 1 is a schematic diagram of an image processing method according to an embodiment of the present invention; FIG.

FIG. 1b is a flowchart of an image processing method according to an embodiment of the present invention;

1c is a flowchart of an image processing method according to another embodiment of the present invention;

FIG. 1 is a schematic diagram of skin color detection in an image processing method according to an embodiment of the invention; FIG.

FIG. 2 is a flowchart of a method for correcting characteristic parameter values of image pixels of an original image in an image processing method according to an embodiment of the present invention; FIG.

2b is a frame diagram of obtaining a standard average value in an image processing method according to an embodiment of the present invention;

FIG. 3 is a flowchart of correcting feature parameter values of each image pixel of an original image in an image processing method according to an embodiment of the present invention; FIG.

4 is a flowchart of an image processing method according to another embodiment of the present invention;

FIG. 5 is a flowchart of an image processing method according to another embodiment of the present invention; FIG.

6 is a flowchart of an image processing method according to still another embodiment of the present invention;

FIG. 7 is a schematic structural diagram of an image processing apparatus according to an embodiment of the present invention; FIG.

FIG. 8 is a schematic structural diagram of an image processing apparatus according to another embodiment of the present invention; FIG.

FIG. 9 is a schematic structural diagram of a terminal according to an embodiment of the present invention.

Implementation

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, but not all embodiments. All other embodiments obtained by a person skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

Embodiments of the present invention provide an image processing method and a processing device.

The image processing device may be specifically integrated in the terminal, and the terminal may be, for example, a smart phone, a tablet computer, a personal computer, or the like.

For example, as shown in FIG. 1a, the image processing apparatus may acquire an original image, for example, may acquire a real-time image acquired by a camera or other image capturing device, or may acquire a picture selected by the user, or receive a picture sent by another device. Etc., then perform skin color detection on the image pixels of the original image to determine the original skin color pixels in the original image, for example, skin color detection by skin color statistical model, such as RGB color space skin color statistics, or skin color detection by threshold segmentation, etc. Etc., by calculating an average value of the characteristic parameter values of the original skin color pixels, and correcting the characteristic parameter values of the image pixels of the original image according to the difference between the preset standard average value and the calculated average value, thereby outputting The corrected original image.

Therefore, in the embodiment of the present invention, the original image is optimized based on the skin color, that is, the feature parameter value of the skin color pixel of the original image and the feature parameter value of the skin color pixel of the preset standard image are referenced by the preset standard image. For comparison, according to the comparison result, the exposure of the original image can be judged to correct the original image, so that the skin color of the original image is closer to the skin color of the standard image, so that the illumination of the original image is closer to the natural situation, and can be avoided to a certain extent. The phenomenon that the image is overexposed or too dark is conducive to improving the quality of the photo.

Referring to FIG. 1b, FIG. 1b is a flow chart of an embodiment of an image processing method of the present invention. As shown, the image processing method includes the following steps:

Step 101b: Acquire an original image.

The acquiring the original image may be a picture acquired by a camera or other image capturing device in real time, or acquiring a photo selected by a user in an album in a terminal device (such as a mobile phone, etc.), or in a terminal device (such as a mobile phone, etc.) The client (such as WeChat, QQ, etc.) receives pictures sent by other users.

Here, the original image may be a person image, a landscape image, an animal image, or the like.

Step 102b: Perform feature detection on image pixels of the selected area in the original image to determine feature pixels of the selected area in the original image.

In an embodiment of the present application, the selected area in the original image may be a plurality of areas. Correspondingly, the feature pixels of the selected area may be the feature pixels of the plurality of areas.

Wherein, when the original image is a character image, the selected area may be a face area, an entire image, or other areas of the original image that include human skin pixels. When the original image is a landscape image, an animal image, or the like, the selected area may be an entire image, or may be an area occupied by a certain target object in the image.

Here, when the original image is a person image, the image pixels of the selected area in the original image are subjected to feature detection to detect the skin color of the image of the selected area in the original image, and the skin color detection manner may be various. For example, a skin color statistical model is established for skin color detection to determine all skin color pixels of a selected region (eg, a face region) in the original image.

Here, when the original image is a landscape image or an animal image or the like, the image pixels of the selected area in the original image are subjected to feature detection to perform target detection on the original image, and the target detection manner is various, for example, RCNN algorithm (Region Convolutional Neural Network), Fast-RCNN algorithm, Faster-RCNN algorithm, Mask RCNN algorithm, etc., through the target detection, the target in the original image (such as animal, tree, Castle, etc., thereby determining the subject color of the target (when there are multiple targets in the original image, selecting the subject color of the subject target, such as the largest target in the image), for example, the color of the trunk The color of the castle wall, etc.

Step 103b: Calculate an average value of the feature parameter values of the feature pixels.

Here, when the original image is a person image, for example, by determining all the skin color pixels in the face region in step 102B, the average value of the feature parameter values of all skin color pixels is calculated.

Here, when the original image is a landscape image, the tree in the landscape image is determined by step 102B, and the color of the trunk is determined, and the average value of the feature parameter values of all trunk colors is calculated.

Step 104b: correcting, according to a difference between a preset standard average value and an average value of the feature parameter values, a feature parameter value of each image pixel of the original image, where the preset standard average value includes The value calculated from the characteristic parameter values of the standard feature pixels of the standard image.

Step 105b: Output the corrected original image.

Through the above technical solution, the target (person, tree, castle, animal, etc.) in the original image can be made closer to its natural color, so that the target display in the original image is more realistic, and the phenomenon of overexposure or overdark is avoided. Using the technical solution of the present application, the feature pixels of the selected area are obtained by performing feature detection on the image pixels of the selected area in the original image, and the average value of the characteristic parameter values of the feature pixel is calculated according to a preset standard. The difference between the average value and the calculated average value is corrected for the characteristic parameter values of the image pixels of the original image. Therefore, with the technical solution of the present application, the original image can be calculated and corrected according to the feature pixels of the locally selected region in the original image. Compared with the prior art, which uses a uniform filter in the template, a cartoon sticker, and an overall brightness equalization technique to correct all the original images, the present application can improve the correction accuracy of the original image.

Hereinafter, the detection of the skin color by the feature detection will be described. Referring to FIG. 1c, FIG. 1c is a flowchart of an embodiment of the image processing method of the present invention. As shown, the image processing method includes the following steps:

Step S101c: Acquire an original image.

The original image is the image to be processed. The obtained original image may be various. For example, acquiring the original image may include: acquiring a picture acquired by a camera or other image capturing device in real time to obtain an original image; or acquiring a picture selected by the user to obtain an original image; Or, receiving a picture sent by the terminal device to obtain an original image, and the like.

Step S102c: Perform skin color detection on image pixels of the selected area in the original image to determine original skin color pixels of the selected area in the original image.

The skin color pixel is the human skin pixel. In this embodiment, the selected area is a face area, and image pixels of the face area in the original image are selected for skin color detection. Specifically, before step S102c, the following steps are performed: face detection is performed on the original image to determine a face region of the original image. After determining the face region of the original image, performing skin color detection on the image pixels of the original image is specifically: performing skin color detection on the image pixels in the face region of the original image to determine original skin color pixels in the original image. When a human face is not detected in the original image, skin color detection is performed on the image pixels of the entire original image to determine the original skin color pixel of the original image.

Among them, there are various ways of face detection, such as a method based on traditional knowledge, a method based on geometric features, and the like, and a combination of one or several methods may be used for face detection. Specifically, for example, a face detection function provided by a terminal system such as iOS or Android may be used to detect a face by using the face detection function when the image processing apparatus acquires a picture acquired by a camera or other image acquisition device in real time; A face detection algorithm provided by software such as OpenCV (Open Source Computer Vision Library) and Tencent Excellent Picture. The above software is used when the image processing apparatus acquires a picture selected by a user or receives a picture transmitted by a terminal device. The provided face detection algorithm detects the face. Wherein, when the original image is obtained according to the user's selection, the face region may also be actively marked by the user. Specifically, when the user selects the original image, the face in the original image may be marked, thereby marking according to the user The area determines the face area in the original image. There are many ways to detect skin color pixels. For example, skin color detection can be performed by establishing a skin color statistical model, such as based on Y (brightness) Cg (difference between green component and brightness) Cr (difference between red component and brightness) and YCgCb (blue Difference between color component and brightness) Skin color detection in color space.

For example, as shown in FIG. 1d, face detection is performed on the original image 10a, thereby determining the face region 11 in the original image 10a. The skin color detection result 10b is obtained by performing skin color detection on the face area 11, wherein the white area represents the skin color area and the black area represents the non-skin color area in the skin color detection result 10b.

In this embodiment, by detecting the skin color for the face region, the amount of calculation can be reduced, the cost of skin color detection can be reduced, and the noise point can be reduced, compared to the skin color detection of the full image region.

Step S103c: Calculate an average value of the feature parameter values of the original skin color pixels.

The feature parameter value refers to the value of the parameter of the pixel, and may be, for example, a grayscale value, a chrominance value, or the like of the pixel. The feature parameter values of the original skin color pixels may be one or more, and when there are multiple feature parameter values, the average value of each feature parameter value is calculated separately. The calculating process of the average value of a certain feature parameter value may specifically include: acquiring feature parameter values of each original skin color pixel; counting the number of original skin color pixels; and calculating a sum of the same feature parameter values of all the original skin color pixels. Calculating the ratio of the sum to the number of original skin color pixels, thereby obtaining an average of the same feature parameter values of all the original skin color pixels.

Step S104c: correcting the feature parameter values of each image pixel of the original image according to the difference between the preset standard average value and the average value of the feature parameters, and the preset standard average value is according to the preset standard image. The characteristic parameter values of the standard skin color pixels are calculated.

Step S105c: Output the corrected original image.

Among them, the skin color of the user is displayed in the standard image, and the image with normal exposure and brightness is relatively close to the natural condition, and the skin color in the standard image is the skin color closest to the natural skin color of the user. In this embodiment, according to the difference between the preset standard average value and the average value of the feature parameter values of the original skin color pixel, it can be determined that the skin color of the original image is darker or brighter than the skin color of the standard image, and further Adjusting the original image according to the difference can make the skin color of the original image closer to the skin color of the standard image (that is, the user's natural skin color), so that the portrait image of the original image is more normal, avoiding overexposure or excessive darkness. . Moreover, based on the skin color detection analysis of the original image, the whole image of the original image is adjusted, so that the brightness of the whole image is closer to the natural situation, the brightness of the light is more balanced, and the image is prevented from being overexposed or too dark, thereby improving the image quality.

Wherein, when the original image is a real-time image acquired by an image capturing device such as a camera, in step S105c, the corrected original image is output and displayed as a real-time image. Specifically, for example, when the user performs photographing (for example, self-photographing) using a camera in the terminal, the image processing apparatus in the terminal automatically adjusts the original image so that the real-time screen seen by the user is already the adjusted image, thereby Make the photos taken by users better. In this case, the user is non-perceived, thereby further enhancing the user experience. When the original image is an image selected by the user, for example, an image selected by the user in the album, the corrected original image is output and saved in the album in step S105c. Specifically, for example, the user selects one image in the album in the terminal, and the image processing device in the terminal adjusts the image, so that the image is rendered better. In this case, the user can clearly perceive the difference before and after the image adjustment.

In this embodiment, the method further includes the step of calculating a standard average value. Specifically, referring to FIG. 2a, and in combination with FIG. 2b, each image of the original image is obtained according to a difference between the calculated average value and the preset standard value. Before the pixel's characteristic parameter values are corrected, the following steps are also included:

Step S201: Acquire at least two standard images.

The standard image can be selected by the user, and the user can select an image that is closer to the skin color of the user as a standard image, and obtain a standard image according to the user's selection.

Step S202: Perform skin color detection on image pixels of each standard image to determine standard skin color pixels of the standard image.

In this embodiment, before step S202, the method further includes: performing face detection on each standard image to determine a face region of the standard image. In step S202, the skin color detection is performed on the image pixels of each standard image, specifically, the skin color detection is performed on the image pixels in the face region of the standard image to determine standard skin color pixels in the standard image. When no face is detected in the standard image, skin color detection is performed on the image pixels of the entire standard image to determine standard skin color pixels. In the above manner, the amount of calculation can be reduced, the calculation performance can be improved, and noise points can be reduced.

Step S203: Calculate an average value of the feature parameter values of the standard skin color pixels of each standard image.

This step is similar to the calculation method of step S103b. Specifically, for the average value of each feature parameter value of the standard skin color pixel of each standard image, the calculation process may be as follows: acquiring the feature parameter value of the standard skin color pixel of the standard image ; count the number of standard skin color pixels; calculate the sum of the same feature parameter values of all standard skin color pixels, calculate the ratio of the sum to the number of standard skin color pixels, thereby obtaining the same feature parameter of all standard skin color pixels of each standard image The average of the values.

Step S204: averaging the average values of the characteristic parameter values of the standard skin color pixels of all standard images to obtain a standard average value.

The calculation step of the standard average includes: counting the total number of standard image images; calculating the sum of the average values of the same feature parameter values of the standard skin color pixels of all standard images; calculating the ratio of the sum to the total number of standard images, thereby obtaining a standard skin color The standard average of the same characteristic parameter values for the pixel.

By using the average of the characteristic parameter values of the standard skin color pixels of a plurality of standard images to calculate the standard average value, the standard average value can be made closer to the natural skin color of the user, and thus, based on the standard average value, according to the characteristics of the original skin color pixel The difference between the average of the parameter values and the standard average adjusts the original image to make the skin tone in the original image closer to the user's natural skin tone.

Of course, in other embodiments, a standard image may also be selected to calculate a standard average value, in which the standard average is the average of the characteristic parameter values of the standard skin color pixels of the standard image.

Wherein, as shown in FIG. 3, correcting the characteristic parameter values of each image pixel of the original image according to the difference between the calculated average value and the preset standard average value includes the following sub-steps:

Sub-step S301: Calculating the difference between the preset standard average value and the average value of the feature parameter values of the original skin color pixels.

The difference specifically refers to the difference between the standard average value minus the average value of the characteristic parameter values of the original skin color pixels.

Sub-step S302: Acquire feature parameter values of each image pixel of the original image.

Sub-step S303: calculating a correction coefficient of each image pixel according to an average value of the feature parameter values of the original skin color pixel and a feature parameter value of each image pixel.

Wherein, when the feature parameter value of the image pixel is less than or equal to the average value of the feature parameter values of the original skin color pixel, the correction coefficient of the image pixel is a ratio of the feature parameter value of the image pixel to the average value of the feature parameter values of the original skin color pixel.

When the feature parameter value of the image pixel is greater than the average value of the feature parameter value of the original skin color pixel, calculating a first difference between the preset constant and the feature parameter value of the image pixel, and calculating a preset constant and a feature of the original skin color pixel A second difference between the average values of the parameter values, the correction coefficient of the image being the ratio of the first difference to the second difference.

Sub-step S304: correcting the feature parameter values of the image pixels of the original image according to the difference between the preset standard average value and the average value of the feature parameter values of the original skin color pixels, and the correction coefficient of the image pixels.

Specifically, the feature parameter values of the image pixels of the original image are corrected according to the following formula:

C ₁ =C ₀ +ΔC*k (1)

Wherein, in the above formula 1, C ₁ represents a characteristic parameter value of the corrected image pixel, and C ₀ represents a characteristic parameter value of the image pixel before the correction, that is, a characteristic parameter value of the original image pixel acquired in step S302, ΔC The difference between the standard average of the corresponding feature parameter values representing the standard skin color pixels and the average of the corresponding feature parameter values of the original skin color pixels, and k represents the correction coefficient of the image pixels.

Wherein, when the feature parameter value of the image pixel is less than or equal to the average value of the feature parameter values of the original skin color pixel, the correction coefficient k is:

k=C ₀ /C _a , where C ₀ ≤C _a (b)

When the feature parameter value of the image pixel is greater than the average value of the feature parameter value of the original skin color pixel, the correction coefficient k is:

k = (nC ₀ ) / (nC _a ), where C ₀ > C _a , n is a preset constant (3).

The embodiments of the present invention will be further described below in conjunction with specific characteristic parameter values.

In an embodiment of the invention, the feature parameter value includes pixel values of three primary color components of red, green, and blue in the first color space, wherein the first color space is red, green, and blue RGB (Red, Green) , Blue (blue)) color space. In this embodiment, the pixel value of each primary color component of the pixel ranges from 0 to 1.

The standard average value of the pixel values of the red, green, and blue primary color components of the standard skin color pixel of the standard image may be calculated in advance according to the above steps S201 to S204.

Referring to FIG. 4, the image processing method of this embodiment specifically includes the following steps:

Step S401: Acquire an original image.

Step S402: Perform skin color detection on the image pixels of the original image to determine original skin color pixels in the original image.

Step S403: Obtain pixel values of three primary color components of red, green, and blue of each original skin color pixel in the original image, and calculate an average value of pixel values of three primary color components of red, green, and blue of all original skin color pixels in the original image. The average of the feature parameter values of the original skin tone pixels.

Specifically, the total amount of the original skin color pixels, that is, the total amount of the respective primary color components, is first counted. Taking the red component as an example, the pixel values of the red components of the original skin color pixels are obtained, the sum of the pixel values of the red components of all the original skin color pixels is calculated, and then the sum of the pixel values of the red components of all the original skin color pixels and the original skin color pixels are calculated. The ratio of the total amount, resulting in an average of the pixel values of the red component of the original skin tone pixel. The average of the pixel values of the green component and the blue component can be calculated in a similar manner.

Step S404: calculating a difference between a standard average value of the pixel values of the three primary color components of the standard skin color pixels in the preset standard image and an average value of the pixel values of the corresponding primary color components of the original skin color pixel in the original image. .

Specifically, the standard average of the pixel values of the red component of the standard skin color pixel is subtracted from the average of the pixel values of the red component of the original skin color pixel, and the standard average of the pixel values of the green component of the standard skin color pixel, respectively. The difference between the average value of the pixel values of the green component of the original skin color pixel, the standard average value of the pixel values of the blue component of the standard skin color pixel, and the difference of the average value of the pixel values of the blue component of the original skin color pixel are subtracted .

Step S405: Acquire pixel values of three primary color components of red, green, and blue of each image pixel to obtain characteristic parameter values of each image pixel.

Step S406: Calculate red, green, and blue of each image pixel according to an average value of pixel values of three primary color components of the original skin color pixel, and pixel values of three primary color components of red, green, and blue of each image pixel. A correction factor for the pixel values of the three primary color components.

Wherein, the correction coefficients of the pixel values of the three primary color components of the red, green and blue of each image pixel in the original image are calculated according to the above formula 2 and formula 3. Taking the red component as an example, when the pixel value of the red component of the image pixel is less than or equal to the average value of the pixel value of the red component of the original skin color pixel, the correction coefficient of the pixel value of the red component of the image pixel is calculated by using the formula 2, where the formula C ₀ in the two represents the pixel value of the red component of the image pixel, and C _a represents the average value of the pixel values of the red component of the original skin color pixel. When the pixel value of the red component of the image pixel is greater than the average value of the pixel value of the red component of the original skin color pixel, the correction coefficient of the red component of the image pixel is calculated by using Equation 3, where n in Equation 3 is 1, that is, when the pixel When the value ranges from 0 to 1, the preset constant is 1. Thus, according to the comparison result of the average value of the pixel value of the red component of the image pixel and the pixel value of the red component of the original skin color pixel, C ₀ , C _a and n are substituted into the above formula 2 or formula 3 to calculate the image pixel. The correction factor for the pixel value of the red component.

The correction coefficients of the pixel values of the green component and the blue component can be obtained in a similar manner, and will not be described herein.

Step S407: The difference between the standard average value of the pixel values of the red, green, and blue primary color components of the standard skin color pixel in the preset standard image and the average value of the pixel values of the corresponding primary color component of the original skin color pixel in the original image. And a correction coefficient of the pixel values of the three primary color components of the red, green, and blue components of each image pixel, and correcting the pixel values of the three primary color components of the red, green, and blue color of each image pixel.

Wherein, the pixel values of the three primary color components of the red, green and blue of each image pixel of the pair of original images are corrected according to the above formula. Specifically, taking the red component as an example, in Equation 1, C ₁ represents the pixel value of the red component of the corrected image pixel, C ₀ represents the pixel value of the red component of the image pixel before correction, and ΔC represents the standard skin color pixel. The difference between the standard average of the pixel values of the red component and the average of the pixel values of the red component of the original skin color pixel, k represents the correction coefficient of the image pixel, thereby substituting C ₀ , ΔC, and k into the above formula 1 The pixel value of the red component of the corrected image pixel can be calculated.

The correction process of the pixel values of the green component and the blue component of the image pixel is thus derived. In the above manner, the pixel values of the respective primary color components corrected for each image pixel can be obtained, and the original image can be corrected.

Step S408: Output the corrected original image.

In this embodiment, by correcting the pixel values of the respective primary color components of the image pixels of the original image, the brightness of the corrected original image can be made closer to the natural situation, and the displayed skin color is closer to the natural skin color of the user. It is beneficial to reduce the phenomenon of excessive darkness or overexposure of images and improve image quality.

In another embodiment of the present invention, the feature parameter value includes a luminance value, a first chrominance value, and a second chrominance value on the second color space, wherein the second color space is YUV (Luminance, Chroma) Degree)) Color space, for convenience of description, the brightness value is represented by a Y value, the first chromaticity value is represented by a U value, and the second chromaticity value is represented by a V value.

The Y value, the U value, and the standard average value of the V value of the standard skin color pixel of the standard image may be calculated in advance according to the above steps S201 to S204.

Referring to FIG. 5, the image processing method of this embodiment specifically includes the following steps:

Step S501: Acquire an original image.

Step S502: Perform skin color detection on image pixels of the original image to determine original skin color pixels in the original image.

Step S503: Obtain pixel values of three primary color components of red, green, and blue in the RGB color space of each original skin color pixel in the original image, according to pixel values of three primary color components of red, green, and blue of each original skin color pixel in the original image. Calculate the Y value, U value, and V value of each original skin color pixel in the YUV color space, and calculate the average value of the Y values of all the original skin color pixels, the average value of the U value, and the average value of the V values to obtain the original skin color pixels. The average of the characteristic parameter values.

The pixel value of each primary color component of each original skin color pixel ranges from 0 to 1.

Calculate the Y value, U value, and V value of each original skin color pixel in the YUV color space according to the following formula:

Wherein, Y ₁ , U ₁ , and V ₁ respectively represent Y values, U values, and V values of the original skin color pixels in the YUV color space, and R ₁ , G ₁ , and B ₁ respectively represent the original skin color pixels in the RGB color space. The pixel values of the primary components of red, green, and blue.

After calculating the Y value, the U value, and the V value of each original skin color pixel, the total amount of the original skin color pixels is counted, and the sum of the Y values of all the original skin color pixels, the sum of the U values, and the sum of the V values are calculated, and then respectively calculated. The ratio of the sum of the Y values to the total amount of the original skin color pixels, the ratio of the sum of the U values to the total amount of the original skin color pixels, and the ratio of the sum of the V values to the total amount of the original skin color pixels, thereby respectively obtaining the Y of the original skin color pixel The average of the values, the average of the U values, and the average of the V values.

Step S504: calculating a difference between a standard average value of the Y value, the U value, and the V value of the standard skin color pixel in the preset standard image and an average value of the corresponding Y value, the U value, and the V value of the original skin color pixel in the original image. .

Specifically, the difference between the standard average value of the Y value of the standard skin color pixel and the average value of the Y value of the original skin color pixel, the standard average value of the U value of the standard skin color pixel, and the U value of the original skin color pixel are respectively calculated. The difference between the average values, the standard average of the V values of the standard skin tone pixels, and the difference between the average values of the V values of the original skin tone pixels.

Step S505: Acquire a Y value, a U value, and a V value of each image pixel of the original image to obtain a feature parameter value of each image pixel of the original image.

The Y value, the U value, and the V value of each image pixel are similar to the calculation methods of the Y value, the U value, and the V value of the original skin color pixel, and can be implemented by referring to the above formula 4, and will not be described herein.

Step S506: Calculate the Y value and the U value of each image pixel according to the average value of the Y value of the original skin color pixel, the average value of the U value, and the average value of the V value, and the Y value, the U value, and the V value of each image pixel. And the correction factor of the V value.

Wherein, the correction values of the Y value, the U value, and the V value of each image pixel in the original image are calculated according to the above formula 2 and formula 3. Taking the Y value as an example, when the Y value of the image pixel is less than or equal to the average value of the Y value of the original skin color pixel, the correction coefficient of the Y value of the image pixel is calculated by Equation 2, where C _{0 in} Formula 2 represents the image pixel. The Y value, C _a represents the average of the Y values of the original skin color pixels. When the Y value of the image pixel is greater than the average value of the Y value of the original skin color pixel, the correction coefficient of the red component of the image pixel is calculated using Equation 3, where n in Equation 3 is 1. The correction coefficients of the U value and the V value can be obtained in a similar manner, and will not be described again.

Step S507: The difference between the standard average value of the Y value, the U value, and the V value of the standard skin color pixel in the preset standard image and the corresponding Y value, the U value, and the average value of the V value of the original skin color pixel in the original image. And a correction coefficient of the Y value, the U value, and the V value of each image pixel, and correct the Y value, the U value, and the V value of each image pixel.

Wherein, the Y value, the U value, and the V value of each image pixel of the pair of original images are corrected according to the above formula. Specifically, taking the value of Y as an example, in Equation 1, C ₁ represents the Y value of the corrected image pixel, C ₀ represents the Y value of the image pixel before correction, and ΔC represents the standard average value of the Y value of the standard skin color pixel. The difference from the average of the Y values of the original skin color pixels, k represents the correction coefficient of the image pixels. The way in which the U and V values of the image pixels are corrected is thus derived. In the above manner, the corrected Y value, U value, and V value of each image pixel can be obtained, and the original image can be corrected.

Step S508: Output the corrected original image.

The outputting the corrected original image specifically includes: calculating, according to the corrected Y value, U value, and V value of each image pixel, pixel values of the three primary color components of the red, green, and blue color of each corrected image pixel, thereby obtaining RGB The original image of the format, the original image of the corrected RGB format is output.

Wherein, the pixel values of the three primary color components of the red, green and blue pixels of each image pixel after correction are calculated according to the following formula:

Wherein Y ₂ , U ₂ , and V ₂ respectively represent Y values, U values, and V values of the corrected image pixels, and R ₂ , G ₂ , and B ₂ respectively represent the three primary colors of red, green, and blue of the corrected image pixels. The pixel value of the component.

In this embodiment, by correcting the Y value, the U value, and the V value of each image pixel of the original image, the image quality of the corrected original image can be further improved, so that the brightness of the corrected original image is closer to the natural situation. The displayed skin tone is closer to the user's natural skin tone.

In still another embodiment of the present invention, only the Y value of the image pixel may be corrected without adjusting the U value and the Y value, and the U value and the V value are unchanged before and after the correction, that is, only the brightness of the original image is adjusted, and the original The chromaticity of the image is not adjusted. In this way, the probability of the original image being overexposed or too dark can be reduced to some extent, and the correction process can be simplified. The main difference between the present embodiment and the embodiment shown in FIG. 5 is that the present embodiment only corrects the Y value of the image pixel. Therefore, the specific calibration process may be performed by referring to the Y value correction process of the embodiment shown in FIG. 5. For the sake of brevity, we will not repeat them here.

In still another embodiment of the present invention, the feature parameter value includes a tone value, a saturation value, and a brightness value in a third color space, wherein the third color space is HSV (Hue, Saturation, Value (Lightness)) The color space. For convenience of description, the tone value is represented by an H value, the saturation value is represented by an S value, and the brightness value is represented by a V value.

The standard value of the H value, the S value, and the V value of the standard skin color pixel of the standard image may be calculated in advance according to the above steps S201 to S204.

Referring to FIG. 6, the image processing method of this embodiment specifically includes the following steps:

Step S601: Acquire an original image.

Step S602: Perform skin color detection on image pixels of the original image to determine original skin color pixels in the original image.

Step S603: acquiring pixel values of three primary color components of red, green, and blue of each original skin color pixel in the original image, according to pixel values of three primary color components of red, green, and blue of each original skin color pixel in the original image. Calculate the H value, S value and V value of each original skin color pixel in the HSV color space, and calculate the average value of the H value of all the original skin color pixels, the average value of the S value, and the average value of the V value to obtain the original skin color pixel. The average of the characteristic parameter values.

The H value, S value, and V value in the HSV color space according to each original skin color pixel are calculated as follows:

Let M = max(R ₁ , G ₁ , B ₁ ), ie M is equal to the largest of R ₁ , G ₁ and B ₁ , N=min(R ₁ , G ₁ , B ₁ ), ie N is equal to R ₁ The smallest of G ₁ and B ₁ , the H value is calculated as follows:

The S value is calculated as follows:

The V value is calculated as follows:

V ₁ =M (eight)

Wherein, H ₁ , S ₁ , and V ₁ respectively represent H values, S values, and V values of the original skin color pixels in the HSV color space, and R ₁ , G ₁ , and B ₁ respectively represent the original skin color pixels on the RGB color space. The pixel values of the primary components of red, green, and blue.

After calculating the H value, the S value, and the V value of each original skin color pixel, the total amount of the original skin color pixels is counted, and the sum of the H values of all the original skin color pixels, the sum of the S values, and the sum of the V values are calculated, and then respectively calculated. The ratio of the sum of the H values to the total amount of the original skin color pixels, the ratio of the sum of the S values to the total amount of the original skin color pixels, and the ratio of the sum of the V values to the total amount of the original skin color pixels, thereby respectively obtaining the H of the original skin color pixels. The average of the values, the average of the S values, and the average of the V values.

Step S604: calculating a difference between a standard average value of the H value, the S value, and the V value of the standard skin color pixel in the preset standard image and an average value of the corresponding H value, the S value, and the V value of the original skin color pixel in the original image. .

Specifically, the difference between the standard average of the H value of the standard skin color pixel and the average value of the H value of the original skin color pixel, the standard average value of the S value of the standard skin color pixel, and the S value of the original skin color pixel are respectively calculated. The difference between the average values, the standard average of the V values of the standard skin tone pixels, and the difference between the average values of the V values of the original skin tone pixels.

Step S605: Acquire an H value, an S value, and a V value of each image pixel of the original image to obtain a feature parameter value of each image pixel of the original image.

The H value, the S value, and the V value of each image pixel are similar to the calculation methods of the H value, the S value, and the V value of the original skin color pixel, and can be calculated by referring to the above formulas 6, 7, and 8. One by one.

Step S606: Calculate the H value and the S value of each image pixel according to the average value of the H value of the original skin color pixel, the average value of the S value, and the average value of the V value, and the H value, the S value, and the V value of each image pixel. And the correction factor of the V value.

Wherein, the correction coefficients of the H value, the S value, and the V value of each image pixel in the original image are calculated according to the above formula 2 and formula 3. Taking the value of H as an example, when the H value of the image pixel is less than or equal to the average value of the H value of the original skin color pixel, the correction coefficient of the H value of the image pixel is calculated by Equation 2, where C _{0 in} Formula 2 represents the image pixel. The value of H, C _a represents the average of the H values of the original skin tone pixels. When the H value of the image pixel is greater than the average value of the H value of the original skin color pixel, the correction coefficient of the red component of the image pixel is calculated by Equation 3, where n in Equation 3 is 1. The correction coefficients of the S value and the V value can be obtained in a similar manner, and will not be described again.

Step S607: The difference between the standard average value of the H value, the S value, and the V value of the standard skin color pixel in the preset standard image and the average value of the corresponding H value, the S value, and the V value of the original skin color pixel in the original image. And a correction coefficient of the H value, the S value, and the V value of each image pixel, and correct the H value, the S value, and the V value of each image pixel.

Here, the H value, the S value, and the V value of each image pixel of the pair of original images are corrected according to the above formula. Specifically, taking the value of H as an example, in Equation 1, C ₁ represents the H value of the corrected image pixel, C ₀ represents the H value of the image pixel before the correction, and ΔC represents the standard average value of the H value of the standard skin color pixel. The difference from the average of the H values of the original skin color pixels, k represents the correction coefficient of the image pixels. The way to correct the S and V values of the image pixels is thus derived. In the above manner, the corrected H value, S value, and V value of each image pixel can be obtained, and the original image can be corrected.

Step S608: Output the corrected original image.

The outputting the corrected original image specifically includes: calculating, according to the corrected H value, S value, and V value of each image pixel, pixel values of the three primary color components of the red, green, and blue color of each corrected image pixel, thereby obtaining RGB The original image of the format, the original image of the corrected RGB format is output.

The pixel values of the red, green and blue primary color components of each image pixel after correction are calculated as follows:

If S ₂ =0, then R ₂ = G ₂ = B ₂ = V ₂ (9)

If S ₂ ≠ 0, it is calculated according to the following conditions:

Let iH be an integer between 0 and (H ₂ *6), including 0;

f=H ₂ -iH;

a=V ₂ *(1-S ₂ )

b=V ₂ *(1-S ₂ *f)

c=V ₂ *(1-S ₂ *(1-f))

The pixel values of the red, green and blue primary colors of each image pixel after correction are:

Wherein, H ₂ , S ₂ , and V ₂ respectively represent the H value, the S value, and the V value of the corrected image pixel, and R ₂ , G ₂ , and B ₂ respectively represent the three primary colors of red, green, and blue of the corrected image pixel. The pixel value of the component.

In this embodiment, by correcting the H value, the S value, and the V value of each image pixel of the original image, the image quality of the corrected original image can be further improved, so that the brightness of the corrected original image is closer to the natural condition. The displayed skin tone is closer to the user's natural skin tone.

Referring to FIG. 7, in an embodiment of the image processing apparatus of the present invention, the image processing apparatus may be a terminal device such as a mobile phone, a tablet computer or a personal computer. The image processing apparatus includes a first acquisition module 701, a first skin color detection module 702, a first calculation module 703, a correction module 704, and an output module 705.

The first obtaining module 701 is configured to acquire an original image. The original image is the image to be processed. The obtained original image may be various. For example, acquiring the original image may include: acquiring a picture acquired by a camera or other image capturing device in real time to obtain an original image; or acquiring a picture selected by the user to obtain an original image; Or, receiving a picture sent by the terminal device to obtain an original image, and the like.

The first skin color detecting module 702 is configured to perform skin color detection on the image pixels of the original image acquired by the first acquiring module 701 to determine original skin color pixels in the original image. The skin color pixel is the human skin pixel.

Among them, there are various ways of face detection, such as a method based on traditional knowledge, a method based on geometric features, etc., and a combination of one or several methods may be used for face detection. Wherein, when the original image is obtained according to the user's selection, the face region may be actively marked by the user. Specifically, when the user selects the original image, the face in the original image may be marked, thereby The area determines the face area in the original image. There are many ways to detect skin color pixels. For example, skin color detection can be performed by establishing a skin color statistical model, such as based on Y (brightness) Cg (difference between green component and brightness) Cr (difference between red component and brightness) and YCgCb (blue Difference between color component and brightness) Skin color detection in color space.

By performing skin color detection on the face area, the amount of calculation can be reduced, the cost of skin color detection can be reduced, and the noise point can be reduced.

The first calculation module 703 is configured to calculate an average value of the feature parameter values of the original skin color pixels. The feature parameter value refers to the value of the parameter of the pixel, and may be, for example, a grayscale value, a chrominance value, or the like of the pixel. The feature parameter values of the original skin color pixels may be one or more, and when there are multiple feature parameter values, the average value of each feature parameter value is calculated separately. Specifically, the first calculating module 703 acquires the feature parameter values of each original skin color pixel, and counts the number of the original skin color pixels, and then calculates the sum of the same feature parameter values of all the original skin color pixels, and calculates the sum and the original skin color pixels. The ratio of the number of the numbers, resulting in an average of the same characteristic parameter values for all of the original skin color pixels.

The correction module 704 is configured to correct the feature parameter values of each image pixel of the original image according to the difference between the preset standard average value and the calculated average value, and the standard average value is according to the preset standard image. The characteristic parameter values of the standard skin color pixels are calculated. The output module 705 is configured to output the corrected original image.

The output module 705 is configured to output and display the corrected original image as a real-time image when the first acquisition module 701 acquires a real-time image acquired by a camera or other image acquisition device to obtain an original image. When the first acquisition module 701 acquires a picture selected by the user to acquire an original image, the output module 705 is configured to output and save the corrected original image into the album.

Referring to FIG. 8, in another embodiment of the image processing apparatus of the present invention, the image processing apparatus further includes a first face detection module 801, a second acquisition module 802, a second skin color detection module 803, a second calculation module 804, and a second Face detection module 805.

In this embodiment, the image pixels of the face region in the original image are selected for skin color detection. Specifically, the first face detection module 801 is configured to perform face detection on the original image acquired by the first acquisition module 701 to determine a face region of the original image before the first skin color detection module 702 performs skin color detection. After determining the face region, the first skin color detecting module 702 is specifically configured to perform skin color detection on image pixels in the face region of the original image to determine original skin color pixels in the original image. Wherein, when no human face is detected in the original image, the first skin color detecting module 702 performs skin color detection on the image pixels of the entire original image to determine the original skin color pixel of the original image.

The second obtaining module 802 is configured to acquire at least two standard images. The standard image may be selected by the user, the user may select an image that is closer to the skin color of the user as the standard image, and the second obtaining module 707 obtains the standard image according to the user's selection.

The second face detection module 805 is configured to perform face detection on each standard image to determine a face area of the standard image. When the face region is determined, the second skin color detecting module 803 is configured to perform skin color detection on the image pixels in the face region of each standard image to determine a standard skin color pixel of the standard image. When no face is detected in the standard image, the second skin color detecting module 803 is configured to perform skin color detection on the entire standard image to determine a standard skin color pixel.

The second calculation module 804 is configured to calculate an average value of the feature parameter values of the standard skin color pixels of each standard image, and average the average values of the feature parameter values of the standard skin color pixels of all the standard images to obtain a standard average value.

For the average value of each feature parameter value of the standard skin color pixel of each standard image, the calculation process may be as follows: the second calculation module 804 acquires the feature parameter value of the standard skin color pixel of the standard image, and then counts the number of standard skin color pixels. And calculating the sum of the same feature parameter values of all standard skin color pixels, and then calculating the ratio of the sum to the number of standard skin color pixels, thereby obtaining an average of the same feature parameter values of all standard skin color pixels of each standard image.

The calculation process of the standard average value may be as follows: the second calculation module 804 counts the total number of standard images, and calculates the sum of the average values of the same feature parameter values of the standard skin color pixels of all standard images, and then calculates the sum and the standard image. The ratio of the totals, resulting in a standard average of the same characteristic parameter values for standard skin tone pixels.

The first face detection module 801 and the second face detection module 805 may be the same module or different modules, and the second acquisition module 802 and the first acquisition module 701 may be the same module or different modules. The second skin color detecting module 803 and the first skin color detecting module 702 may be the same module or different modules, and the second calculating module 804 and the first calculating module 703 may be the same module or different modules.

In this embodiment, the correction module 704 includes a first calculation unit 7041, an acquisition unit 7042, a second calculation unit 7043, and a correction unit 7044.

The first calculating unit 7041 is configured to calculate a difference between a preset standard average value and an average value of the feature parameter values of the original skin color pixel, where the difference specifically refers to a standard average value minus a characteristic parameter value of the original skin color pixel. The difference between the averages.

The obtaining unit 7042 is configured to acquire feature parameter values of each image pixel of the original image.

The second calculating unit 7043 is configured to calculate a correction coefficient of each image pixel according to an average value of the feature parameter values of the original skin color pixel and a feature parameter value of each image pixel. Specifically, the second calculating unit 7043 is configured to calculate an average value of the feature parameter value of the image pixel and the feature parameter value of the original skin color pixel when the feature parameter value of the image pixel is less than or equal to the average value of the feature parameter values of the original skin color pixel. Ratio of the image, thereby obtaining a correction coefficient of the image pixel; and for calculating the first between the preset constant and the characteristic parameter value of the image pixel when the feature parameter value of the image pixel is greater than the average value of the feature parameter value of the original skin color pixel a difference value, and calculating a second difference between the preset constant and an average value of the feature parameter values of the original skin color pixel, and a ratio of the first difference value to the second difference value, thereby obtaining a correction coefficient of the image pixel.

The correcting unit 7044 is configured to correct the feature parameter values of the image pixels of the original image according to the difference between the preset standard average value and the average value of the feature parameter values of the original skin color pixels, and the correction coefficient of the image pixels.

C ₁ =C ₀ +ΔC*k (1)

k=C ₀ /C _a , where C ₀ ≤C _a (b)

k = (nC ₀ ) / (nC _a ), where C ₀ > C _a , n is a preset constant (3).

In a specific embodiment of the present invention, the feature parameter value includes pixel values of three primary color components of red, green, and blue in the first color space, wherein the first color space is red, green, and blue RGB (Red (Red), Green (Green) ), Blue (blue)) color space. In this embodiment, the pixel value of each primary color component of the pixel ranges from 0 to 1.

In this embodiment, the first calculating module 703 is specifically configured to obtain pixel values of three primary color components of the original skin color pixels in the original image, and calculate red, green, and blue colors of all the original skin color pixels in the original image. The average of the pixel values of the primary color components, thereby obtaining an average of the characteristic parameter values of the original skin color pixels.

The first calculating unit 7041 is configured to calculate a standard average value of the pixel values of the red, green, and blue primary color components of the standard skin color pixel in the preset standard image and an average value of the pixel values of the corresponding primary color components of the original skin color pixel in the original image. The difference between the two.

The obtaining unit 7042 is configured to obtain pixel values of three primary color components of red, green, and blue of each image pixel, to obtain feature parameter values of each image pixel.

The second calculating unit 7043 is configured to calculate the red color of each image pixel according to the average value of the pixel values of the three primary color components of the red, green, and blue color of the original skin color pixel, and the pixel values of the three primary color components of the red, green, and blue color of each image pixel. Correction coefficients for pixel values of the three primary color components of green, blue, and blue.

The correcting unit 7044 is configured to pre-equalize the average value of the pixel values of the corresponding primary color components of the original skin color pixels in the original image according to a standard average value of the pixel values of the red, green, and blue primary color components of the standard skin color pixel in the preset standard image. The difference value, and the correction coefficient of the pixel values of the three primary color components of the red, green, and blue components of each image pixel, correct the pixel values of the three primary color components of the red, green, and blue color of each image pixel.

The correcting unit 7044 corrects the pixel values of the three primary color components of the red, green, and blue color of each image pixel of the original image according to the above formula. Specifically, taking the red component as an example, in Equation 1, C ₁ represents the pixel value of the red component of the corrected image pixel, C ₀ represents the pixel value of the red component of the image pixel before correction, and ΔC represents the standard skin color pixel. The difference between the standard average of the pixel values of the red component and the average of the pixel values of the red component of the original skin tone pixel, k represents the correction factor for the image pixel. The manner in which the pixel values of the green and blue components of the image pixel are corrected is thus derived. In the above manner, the pixel values of the respective primary color components corrected for each image pixel can be obtained, and the original image can be corrected.

In this embodiment, the first calculating module 703 is configured to obtain pixel values of three primary color components of the original color skin pixels in the first color space in the original image, and according to the red color of each original skin color pixel in the original image. And the pixel values of the three primary color components of green, blue, calculate the luminance value, the first chrominance value, and the second chrominance value of each original skin color pixel in the second color space, and calculate an average value of the luminance values of all the original skin color pixels. The average of the first chrominance values and the average of the second chrominance values, thereby obtaining an average of the characteristic parameter values of the original skin color pixels.

Wherein, the Y value, the U value, and the V value of each original skin color pixel in the YUV color space can be calculated according to the above formula 4.

The first calculating unit 7041 is configured to calculate a brightness value of the standard skin color pixel, a first chromaticity value, a standard average value of the second chromaticity value, and a corresponding brightness value, a first chromaticity value, and a second chromaticity value of the original skin color pixel. The difference between the averages.

The obtaining unit 7042 is configured to acquire a luminance value, a first chrominance value, and a second chrominance value of each image pixel of the original image, thereby obtaining feature parameter values of each image pixel of the original image.

The second calculating unit 7043 is configured to calculate an average value of the brightness values of the original skin color pixels, an average value of the first color value, and an average value of the second color value, and a brightness value, a first color value, and a color value of each image pixel. The second chrominance value calculates a luminance coefficient of each image pixel, a first chrominance value, and a correction coefficient of the second chrominance value.

The correcting unit 7044 is configured to: according to the brightness value of the standard skin color pixel in the preset standard image, the first chromaticity value, the standard average value of the second chromaticity value, and the corresponding brightness value of the original skin color pixel in the original image, the first chromaticity value a difference between the average values of the second chrominance values, and a luminance value of each image pixel, a first chrominance value, and a correction coefficient of the second chrominance value, a luminance value for each image pixel, and a first chrominance The value and the second chrominance value are corrected.

The correcting unit 7044 corrects the Y value, the U value, and the V value of each image pixel of the pair of original images according to the above formula. Specifically, taking the value of Y as an example, in Equation 1, C ₁ represents the Y value of the corrected image pixel, C ₀ represents the Y value of the image pixel before correction, and ΔC represents the standard average value of the Y value of the standard skin color pixel. The difference from the average of the Y values of the original skin color pixels, k represents the correction coefficient of the image pixels. The way in which the U and V values of the image pixels are corrected is thus derived. In the above manner, the corrected Y value, U value, and V value of each image pixel can be obtained, and the original image can be corrected.

The output module 705 is specifically configured to calculate pixel values of the three primary color components of the red, green, and blue color of each image pixel according to the corrected Y value, U value, and V value of each image pixel, thereby obtaining the original RGB format. Image and output the original image in corrected RGB format. The pixel values of the three primary color components of the red, green, and blue colors of the corrected image pixels may be calculated according to the above formula 5.

In still another embodiment of the present invention, only the Y value of the image pixel may be corrected. The main difference from the above embodiment is that the present embodiment only corrects the Y value of the image pixel, and does not correct the U value and the V value. The U value and the V value are unchanged before and after the correction, and the specific correction manner is the same as the above embodiment. Similarly, I will not repeat them here.

In still another embodiment of the present invention, the feature parameter value includes a tone value, a saturation value, and a brightness value in a third color space, wherein the third color space is HSV (Hue, Saturation, Value ( Lightness)) The color space, for ease of description, the hue value is represented by the H value, the saturation value is represented by the S value, and the brightness value is represented by the V value.

In this embodiment, the first calculating module 703 is configured to obtain pixel values of three primary color components of the original color skin pixels in the first color space in the original image, and according to the red color of each original skin color pixel in the original image. The pixel values of the three primary color components of green, blue, and blue, calculate the hue value, the saturation value, and the brightness value of each original skin color pixel in the third color space, and calculate the average value and the saturation value of the hue values of all the original skin color pixels. The average value and the average value of the brightness values are averaged to obtain the characteristic parameter values of the original skin color pixels.

Wherein, the H value, the S value, and the V value of each original skin color pixel in the HSV color space may be calculated according to the above formulas 6, 7, and 8.

The first calculating unit 7041 is configured to calculate a tone value, a saturation value, a standard average value of the brightness value of the standard skin color pixel in the preset standard image, and an average value of the corresponding tone value, the saturation value, and the brightness value of the original skin color pixel in the original image. The difference between the values.

The obtaining unit 7042 is configured to acquire a tone value, a saturation value, and a brightness value of each image pixel of the original image to obtain a feature parameter value of each image pixel of the original image.

The second calculating unit 7043 is configured to calculate each image pixel according to an average value of the hue value of the original skin color pixel, an average value of the saturation value, and an average value of the brightness value, and a hue value, a saturation value, and a brightness value of each image pixel. The tonal value, the saturation value, and the correction factor for the brightness value.

The correcting unit 7044 is configured to: according to a standard average value of the hue value, the saturation value, and the brightness value of the standard skin color pixel in the preset standard image, and an average value of the corresponding hue value, the saturation value, and the brightness value of the original skin color pixel in the original image. The difference between the image, the tone value of each image pixel, the saturation value, and the correction factor of the brightness value are corrected for the tone value, the saturation value, and the brightness value of each image pixel.

Wherein, the H value, the S value, and the V value of each image pixel of the pair of original images can be corrected according to the above formula. Specifically, taking the value of H as an example, in Equation 1, C ₁ represents the H value of the corrected image pixel, C ₀ represents the H value of the image pixel before the correction, and ΔC represents the standard average value of the H value of the standard skin color pixel. The difference from the average of the H values of the original skin color pixels, k represents the correction coefficient of the image pixels. The way to correct the S and V values of the image pixels is thus derived. In the above manner, the corrected H value, S value, and V value of each image pixel can be obtained, and the original image can be corrected.

The output module 705 is specifically configured to calculate pixel values of the three primary color components of the red, green, and blue color of each image pixel according to the corrected H value, S value, and V value of each image pixel, thereby obtaining the original RGB format. Image and output the original image in corrected RGB format. Wherein, the pixel values of the three primary color components of the red, green and blue of each corrected image pixel can be calculated according to the above formulas IX and IX.

The embodiment of the present invention further provides a terminal. As shown in FIG. 9, the terminal may include a radio frequency (RF) circuit 901, a memory 902 including one or more computer readable storage media, an input unit 903, and a display. unit 904, a sensor 905, an audio circuit 906, wireless fidelity (WiFi, W ireless fidelity) module 907, a processor comprises one or

more processing cores

908, 909 and a power supply and other components. It will be understood by those skilled in the art that the terminal structure shown in FIG. 9 does not constitute a limitation to the terminal, and may include more or less components than those illustrated, or combine some components, or different component arrangements. among them:

The RF circuit 901 includes, but is not limited to, an antenna, at least one amplifier, a tuner, one or more oscillators, a Subscriber Identity Module (SIM), a transceiver, a coupler, and a low noise amplifier (LNA, Low Noise). Amplifier), duplexer, etc. In addition, the RF circuit 901 can also communicate with the network and other devices through wireless communication. The wireless communication may use any communication standard or protocol, including but not limited to Global System of Mobile communication (GSM), General Packet Radio Service (GPRS), and Code Division Multiple Access (CDMA). , Code Division Multiple Access), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), e-mail, Short Messaging Service (SMS), and the like.

The memory 902 can be used to store software programs and modules, such as computer readable instructions. The processor 908 executes various functional applications and data processing by executing software programs and modules stored in the memory 902, for example, the above-described FIGS. 1b, 1c can be performed. The method shown in 2a, 3, 4, 5, and 6 and the apparatus shown in Figs. 7 and 8. The memory 902 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may be stored according to Data created by the use of the terminal (such as audio data, phone book, etc.). Moreover, memory 902 can include high speed random access memory, and can also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device. Accordingly, memory 902 may also include a memory controller to provide access to memory 902 by processor 908 and input unit 903.

Input unit 903 can be used to receive input numeric or character information, as well as to generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function controls. In particular, in one particular embodiment, input unit 903 can include a touch-sensitive surface as well as other input devices. Touch-sensitive surfaces, also known as touch screens or trackpads, collect touch operations on or near the user (such as the user using a finger, stylus, etc., any suitable object or accessory on a touch-sensitive surface or touch-sensitive Operation near the surface), and drive the corresponding connecting device according to a preset program. In some examples, the touch sensitive surface can include both portions of the touch detection device and the touch controller. Wherein, the touch detection device detects the touch orientation of the user, and detects a signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts the touch information into contact coordinates, and sends the touch information. The processor 908 is provided and can receive commands from the processor 908 and execute them. In addition, touch-sensitive surfaces can be implemented in a variety of types, including resistive, capacitive, infrared, and surface acoustic waves. In addition to the touch-sensitive surface, the input unit 903 can also include other input devices. Specifically, other input devices may include, but are not limited to, one or more of a physical keyboard, function keys (such as volume control buttons, switch buttons, etc.), trackballs, mice, joysticks, and the like.

Display unit 904 can be used to display information entered by the user or information provided to the user, as well as various graphical user interfaces of the terminal, which can be composed of graphics, text, icons, video, and any combination thereof. The display unit 904 can include a display panel. In some examples, the display panel can be configured in the form of a Liquid Crystal Display (LCD), an Organic Light-Emitting Diode (OLED), or the like. Further, the touch-sensitive surface can cover the display panel, and when the touch-sensitive surface detects a touch operation thereon or nearby, it is transmitted to the processor 908 to determine the type of the touch event, and then the processor 908 displays the type according to the type of the touch event. A corresponding visual output is provided on the panel. Although in FIG. 9, the touch-sensitive surface and display panel are implemented as two separate components to implement input and input functions, in some embodiments, the touch-sensitive surface can be integrated with the display panel to implement input and output functions.

The terminal may also include at least one type of sensor 905, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display panel according to the brightness of the ambient light, and the proximity sensor may close the display panel and/or the backlight when the terminal moves to the ear. . As a kind of motion sensor, the gravity acceleration sensor can detect the magnitude of acceleration in all directions (usually three axes). When it is stationary, it can detect the magnitude and direction of gravity. It can be used to identify the gesture of the mobile phone (such as horizontal and vertical screen switching, related Game, magnetometer attitude calibration), vibration recognition related functions (such as pedometer, tapping), etc.; as for the terminal can also be configured with gyroscopes, barometers, hygrometers, thermometers, infrared sensors and other sensors, no longer Narration.

The audio circuit 906, the speaker, and the microphone provide an audio interface between the user and the terminal. The audio circuit 906 can transmit the converted electrical signal of the audio data to the speaker, and convert it into a sound signal output by the speaker; on the other hand, the microphone converts the collected sound signal into an electrical signal, which is received by the audio circuit 906 and then converted. After the audio data is processed by the audio data output processor 908, it is sent to, for example, another terminal via the RF circuit 901, or the audio data is output to the memory 902 for further processing. The audio circuit 906 may also include an earbud jack to provide communication between the peripheral earphone and the terminal.

WiFi is a short-range wireless transmission technology, terminal through WiFi module 907 can help users send and receive email, browse the web and access streaming media and so on, which provides wireless broadband Internet access for users. Although FIG. 9 shows the WiFi module 907, it can be understood that it does not belong to the necessary configuration of the terminal, and may be omitted as needed within the scope of not changing the essence of the invention.

The processor 908 is the control center of the terminal, which connects various portions of the entire handset using various interfaces and lines, by executing or executing software programs and/or modules stored in the memory 902, and invoking data stored in the memory 902, executing Various functions of the terminal and processing data. In some examples, processor 908 can include one or more processing cores; preferably, processor 908 can integrate an application processor and a modem processor, wherein the application processor primarily processes an operating system, a user interface, and an application Etc. The modem processor primarily handles wireless communications. It will be appreciated that the above described modem processor may also not be integrated into the processor 908.

The terminal also includes a power supply 909 (such as a battery) that supplies power to the various components. The power supply 909 can be logically coupled to the processor 908 through a power management system to manage functions such as charging, discharging, and power management through the power management system. The power supply 909 may also include any one or more of a DC or AC power source, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and the like.

Although not shown, the terminal may further include a camera, a Bluetooth module, and the like, and details are not described herein again. Specifically, in this embodiment, the processor 908 in the terminal loads the executable file corresponding to the process of one or more applications into the memory 902 according to the following instructions, and is executed by the processor 908 to be stored in the memory. The application in 902 to implement various functions:

Get the original image. The original image is the image to be processed. The obtained original image may be various. For example, acquiring the original image may include: acquiring a picture acquired by a camera or other image capturing device in real time to obtain an original image; or acquiring a picture selected by the user to obtain an original image or the like.

Feature detection is performed on image pixels of selected regions in the original image to determine feature pixels of the selected region in the original image. In an embodiment of the present application, the feature pixel may include: a skin color pixel, that is, a human skin pixel.

Before the skin color detection is performed, face detection is performed on the original image to determine the face area of the original image. The skin color detection of the image pixels of the original image is specifically: skin color detection is performed on image pixels in the face region of the original image to determine original skin color pixels in the original image. When a human face is not detected in the original image, skin color detection is performed on the image pixels of the entire original image to determine the original skin color pixel of the original image.

The average of the characteristic parameter values of the feature pixels is calculated.

Correcting a characteristic parameter value of each image pixel of the original image according to a difference between the preset standard average value and the characteristic parameter value, and the preset standard average value includes: according to a standard pixel in the preset standard image The characteristic parameter values are calculated.

Wherein, the difference between the preset standard average value and the average value of the feature parameter values of the original skin color pixels is first calculated. The difference specifically refers to the difference between the standard average value minus the average value of the characteristic parameter values of the original skin color pixels. Then, the correction coefficient of each image pixel is calculated according to the above formulas 2 and 3, and the characteristic parameter values of each image pixel of the original image are calculated according to the formula.

The corrected original image is output.

For the specific implementation of the foregoing operations, refer to the foregoing embodiments, and details are not described herein again.

The embodiments of the present application provide a non-transitory computer readable storage medium storing computer readable instructions that cause at least one processor to perform the operations in the methods and/or apparatus described above.

A person skilled in the art may understand that all or part of the various steps of the foregoing embodiments may be performed by a program to instruct related hardware. The program may be stored in a computer readable storage medium, and the storage medium may include: Read Only Memory (ROM), Random Access Memory (RAM), disk or optical disk.

The image processing method, device, and storage medium provided by the embodiments of the present invention are described in detail. The principles and implementation manners of the embodiments of the present invention are described in the specific examples. The description of the above embodiments is only used. To help understand the method and the core idea of the embodiments of the present invention; at the same time, for those skilled in the art, according to the idea of the embodiment of the present invention, there are some changes in the specific implementation manner and the application scope. The content of the present specification should not be construed as limiting the embodiments of the present invention.

Claims

An image processing method is applied to an image processing apparatus, including:

Get the original image;

Performing feature detection on image pixels of selected regions in the original image to determine feature pixels of the selected region in the original image;

Calculating an average value of characteristic parameter values of the feature pixels;

Correcting a feature parameter value of each image pixel of the original image according to a difference between a preset standard average value and an average value of the feature parameter values, the preset standard average value including an image according to a preset standard The value of the characteristic parameter value of the standard feature pixel;

The corrected original image is output.
The image processing method according to claim 1, wherein the performing feature detection on the image pixels of the selected area in the original image to determine the feature pixels of the selected area in the original image comprises:

Performing skin color detection on image pixels of selected regions in the original image to determine original skin color pixels of the selected region in the original image;

The preset standard average value includes: a value calculated according to a characteristic parameter value of a standard skin color pixel of the preset standard image.
The image processing method according to claim 2, wherein before performing skin color detection on the image pixels of the original image to determine an original skin color pixel in the original image, the method further comprises: performing a face on the original image Detecting to determine a face area of the original image;

Performing skin color detection on the image pixels of the selected area in the original image to determine original skin color pixels of the selected area in the original image, including: image pixels in a face area of the original image Skin color detection is performed to determine the original skin color pixels of the face region in the original image.
The image processing method according to claim 2, wherein said characteristic parameter values of respective image pixels of said original image are performed according to a difference between a preset standard average value and an average value of said characteristic parameter values Correction, including:

Calculating a difference between a preset standard average value and an average value of the feature parameter values of the feature pixels;

Obtaining feature parameter values of each image pixel of the original image;

Calculating a correction coefficient of each image pixel according to an average value of the feature parameter values of the feature pixel and a feature parameter value of each image pixel;

The feature parameter values of the respective image pixels are corrected based on the difference and the correction coefficient of each image pixel.
The image processing method according to claim 4, wherein the calculating a correction coefficient of each image pixel according to an average value of the feature parameter values of the feature pixels and a feature parameter value of each image pixel comprises:

When the feature parameter value of the image pixel is less than or equal to an average value of the feature parameter values of the feature pixel, the correction coefficient of the image pixel is a feature parameter value of the image pixel and a feature parameter value of the feature pixel. The ratio of the average values;

Calculating a first difference between a preset constant and a feature parameter value of the image pixel, and calculating the preset constant when a feature parameter value of the image pixel is greater than an average value of a feature parameter value of the feature pixel And a second difference between the average value of the feature parameter values of the feature pixels, the correction coefficient of the image pixel being a ratio of the first difference value to the second difference value.
The image processing method according to claim 4, wherein said characteristic parameter value includes pixel values of three primary color components of red, green, and blue;

The calculating an average value of the feature parameter values of the feature pixel includes: acquiring pixel values of three primary color components of red, green, and blue of each feature pixel in the original image; calculating red, green, and blue of all feature pixels in the original image An average value of the pixel values of the three primary color components, and an average value of the characteristic parameter values of the feature pixels is obtained;

Calculating a difference between an average value of the preset standard and an average value of the feature parameter values of the feature pixel, comprising: calculating pixel values of three primary color components of red, green, and blue of the standard skin color pixel in the preset standard image a difference between a standard average value and an average value of pixel values of respective primary color components of the feature pixel;

And acquiring the feature parameter values of each image pixel of the original image, comprising: acquiring pixel values of three primary color components of red, green, and blue of each image pixel, to obtain feature parameter values of each image pixel;

Calculating, according to an average value of the feature parameter values of the feature pixels and a feature parameter value of each image pixel, a correction coefficient of each image pixel, including: pixel values of three primary color components of red, green, and blue according to the feature pixel The average value, and the pixel values of the three primary color components of the red, green, and blue colors of each image pixel, and the correction coefficients of the pixel values of the three primary color components of the red, green, and blue components of each image pixel are calculated;

And correcting, according to the difference value and a correction coefficient of each image pixel, a feature parameter value of each image pixel, including: according to a pixel value of three primary color components of red, green, and blue of a standard skin color pixel in a preset standard image; a difference between a standard average value and an average value of pixel values of respective primary color components of the characteristic pixel, and a correction coefficient of pixel values of three primary color components of red, green, and blue of each image pixel, red for each image pixel The pixel values of the three primary colors of green, blue, and blue are corrected.
The image processing method according to claim 4, wherein the feature parameter value comprises a luminance value, a first chrominance value, and a second chrominance value;

The calculating an average value of the feature parameter values of the feature pixel includes: acquiring pixel values of the three primary color components of the red, green, and blue color of each feature pixel in the original image; according to the red, green, and blue color of each feature pixel in the original image Calculating a pixel value of the three primary color components, calculating a luminance value, a first chrominance value, and a second chrominance value; calculating an average value of the luminance values of all the characteristic pixels, an average value of the first chromaticity values, and a second color The average value of the degree values, and the average value of the characteristic parameter values of the feature pixels is obtained;

Calculating a difference between the preset standard average value and an average value of the feature parameter values of the feature pixel, comprising: calculating a brightness value, a first chromaticity value, and a second color of a standard skin color pixel in the preset standard image a difference between a standard average of the degrees and a corresponding brightness value of the feature pixel, an average of the first chrominance value and the second chrominance value;

And acquiring the feature parameter value of the image pixel of the original image, comprising: acquiring a luminance value, a first chrominance value, and a second chrominance value of each image pixel of the original image to obtain each image of the original image The characteristic parameter value of the pixel;

And calculating, according to an average value of the feature parameter values of the feature pixels and a feature parameter value of each image pixel, a correction coefficient of each image pixel, including: an average value according to the brightness value of the feature pixel, and a first chromaticity value And an average value of the second chromaticity value, and a luminance value, a first chrominance value, and a second chrominance value of each image pixel, and calculating a luminance value, a first chrominance value, and a second color of each image pixel Correction factor of the degree value;

And correcting, according to the difference value and a correction coefficient of each image pixel, a feature parameter value of each image pixel, comprising: according to a brightness value, a first chromaticity value, and a second color of a standard skin color pixel in a preset standard image a difference between a standard average of the degree values and a corresponding brightness value of the feature pixel, an average of the first chrominance value and the second chrominance value, and a brightness value, a first chromaticity value, and The correction coefficient of the second chrominance value corrects the luminance value, the first chrominance value, and the second chrominance value of each image pixel.
The image processing method according to claim 4, wherein said characteristic parameter value comprises a brightness value;

The calculating an average value of the feature parameter values of the feature pixel includes: acquiring pixel values of the three primary color components of the red, green, and blue color of each feature pixel in the original image; according to the red, green, and blue color of each feature pixel in the original image Calculating the luminance value of each characteristic pixel by calculating the pixel value of the three primary color components; calculating an average value of the luminance values of all the characteristic pixels to obtain an average value of the characteristic parameter values of the characteristic pixel;

Calculating a difference between the preset standard average value and an average value of the feature parameter values of the feature pixels, comprising: calculating a standard average value of the brightness values of the standard skin color pixels in the preset standard image and the feature pixels The difference between the average values of the luminance values;

And acquiring the feature parameter value of the image pixel of the original image, comprising: acquiring a brightness value of each image pixel of the original image, and obtaining a feature parameter value of each image pixel of the original image;

Calculating, according to an average value of the feature parameter values of the feature pixels and a feature parameter value of each image pixel, a correction coefficient of each image pixel, including: an average value according to the brightness value of the feature pixel and a brightness of each image pixel a value, a correction coefficient for calculating a luminance value of each image pixel;

And correcting, according to the difference value and the correction coefficient of each image pixel, the feature parameter value of each image pixel, including: a standard average value of the brightness value of the standard skin color pixel in the preset standard image, and the feature pixel The difference between the average values of the luminance values and the correction coefficient of the luminance values of the respective image pixels correct the luminance values of the respective image pixels.
The image processing method according to claim 4, wherein the feature parameter value comprises a tone value, a saturation value, and a brightness value;

The calculating an average value of the feature parameter values of the feature pixel includes: acquiring pixel values of the three primary color components of the red, green, and blue color of each feature pixel in the original image; according to the red, green, and blue color of each feature pixel in the original image Calculating the pixel value of the three primary color components, calculating the hue value, the saturation value, and the brightness value of each feature pixel; calculating the average value of the hue value of all the feature pixels, the average value of the saturation value, and the average value of the brightness value to obtain the feature pixel The average value of the characteristic parameter values;

Calculating a difference between an average value of the preset standard and an average value of the feature parameter values of the feature pixel, comprising: calculating a standard average of the hue value, the saturation value, and the brightness value of the standard skin color pixel in the preset standard image a difference between a value and an average of a corresponding tone value, a saturation value, and a brightness value of the feature pixel;

And acquiring the feature parameter value of the image pixel of the original image, comprising: acquiring a tone value, a saturation value, and a brightness value of each image pixel of the original image to obtain a feature parameter value of each image pixel of the original image. ;

Calculating, according to an average value of the feature parameter values of the feature pixels and a feature parameter value of each image pixel, a correction coefficient of each image pixel, including: an average value of the tone values of the feature pixels, and an average of saturation values a value, an average value of the brightness value, and a tone value, a saturation value, and a brightness value of each image pixel, and a correction coefficient of a tone value, a saturation value, and a brightness value of each image pixel;

And correcting, according to the difference value and a correction coefficient of each image pixel, a feature parameter value of each image pixel, including: a standard average of a tone value, a saturation value, and a brightness value of a standard skin color pixel according to a preset standard image a difference between a value and an average value of a corresponding tone value, a saturation value, and a brightness value of the feature pixel, and a tone value, a saturation value, and a correction coefficient of a brightness value of each image pixel, and a hue for each image pixel The value, saturation value, and brightness value are corrected.
The image processing method according to any one of claims 1 to 9, wherein said image pixel of said original image is based on a difference between a preset standard average value and an average value of said feature parameter values Before the feature parameter values are corrected, it also includes:

Obtaining at least two of the standard images;

Performing feature detection on image pixels of selected regions in each of the standard images to determine standard feature pixels of the standard image;

Calculating an average value of characteristic parameter values of standard feature pixels of each of the standard images;

The average of the characteristic parameter values of the standard feature pixels of all the standard images is averaged to obtain the preset standard average.
The image processing method according to claim 10, wherein before the feature detection is performed on the image pixels of the selected area in each of the standard images to determine the standard feature pixels of the standard image, the method further comprises: Performing face detection on the standard image to determine a face area of the standard image;

Performing feature detection on the image pixels of the selected area in each of the standard images to determine standard feature pixels of the standard image includes: performing skin color detection on image pixels in the face region of the standard image, A standard skin tone pixel in the standard image is determined.
An image processing apparatus, wherein the apparatus includes a processor and a memory, the memory storing computer readable instructions that enable the processor to:

Get the original image;

Performing skin color detection on image pixels of selected regions in the original image to determine feature pixels of the selected region in the original image;

Calculating an average value of characteristic parameter values of the feature pixels;

Correcting a feature parameter value of each image pixel of the original image according to a difference between a preset standard average value and an average value of the feature parameter values, the preset standard average value including an image according to a preset standard The value of the characteristic parameter value of the standard feature pixel;

The corrected original image is output.
The image processing device according to claim 12, wherein said computer readable instructions are to cause said processor to:

Performing skin color detection on image pixels of selected regions in the original image to determine original skin color pixels of the selected region in the original image;

The preset standard average value includes: a value calculated according to a characteristic parameter value of a standard skin color pixel of the preset standard image.
The image processing device according to claim 13, wherein said computer readable instructions are to cause said processor;

Performing face detection on the original image to determine the face region of the original image;

Skin color detection is performed on image pixels in the face region of the original image to determine original skin color pixels of the face region in the original image.
The image processing device according to claim 13, wherein said computer readable instructions are to cause said processor to:

Calculating a difference between a preset standard average value and an average value of the feature parameter values of the feature pixels;

Obtaining feature parameter values of each image pixel of the original image;

Calculating a correction coefficient of each image pixel according to an average value of the feature parameter values of the feature pixel and a feature parameter value of each image pixel;

The feature parameter values of the respective image pixels are corrected based on the difference and the correction coefficient of each image pixel.
The image processing device according to claim 15, wherein said computer readable instructions are to cause said processor to:

When a characteristic parameter value of the image pixel is less than or equal to an average value of the feature parameter values of the feature pixel, calculating a ratio of a feature parameter value of the image pixel to an average value of the feature parameter value of the feature pixel, a correction coefficient of the image pixel;

Calculating a first difference between a preset constant and a feature parameter value of the image pixel, and calculating the preset when a feature parameter value of the image pixel is greater than an average value of a feature parameter value of the feature pixel And a second difference between the constant and the average value of the feature parameter values of the feature pixel, and calculating a ratio of the first difference value to the second difference value to obtain a correction coefficient of the image pixel.
The image processing device according to any one of claims 12 to 16, wherein the computer readable instructions are to cause the processor to:

Obtaining at least two of the standard images;

Performing feature detection on image pixels of selected regions in each of the standard images to determine standard feature pixels of the standard image;

An average value of the feature parameter values of the standard feature pixels of each of the standard images is calculated, and an average value of the feature parameter values of the standard feature pixels of all the standard images is averaged to obtain a standard average value.
The image processing device according to claim 17, wherein said computer readable instructions are to cause said processor;

Performing face detection on each of the standard images to determine a face region of the standard image;

Skin color detection is performed on image pixels within the face region of the standard image to determine standard skin color pixels in the standard image.
A non-transitory computer readable storage medium storing computer readable instructions, which may cause at least one processor to perform the method of any one of claims 1 to 11.