CN107909542A - Image processing method, device, computer-readable recording medium and electronic equipment - Google Patents

Abstract

The present application relates to an image processing method, device, computer-readable storage medium and electronic equipment. The method includes: obtaining the image to be processed and the corresponding definition; obtaining the corresponding beauty parameter model according to the definition, the beauty parameter model refers to a model used to calculate the beauty parameter; according to the definition Acquiring corresponding target beautification parameters from the beautification parameter model; performing beautification processing on the image to be processed according to the target beautification parameters. The above image processing method, device, computer-readable storage medium and electronic equipment improve the accuracy of image processing.

Description

Image processing method, device, computer-readable storage medium, and electronic device

technical field

The present application relates to the technical field of image processing, in particular to an image processing method, device, computer-readable storage medium and electronic equipment.

Background technique

Whether at work or in life, taking pictures is an essential skill. In order to take a satisfactory photo, it is not only necessary to improve the shooting parameters during the shooting process, but also to improve the photo itself after the shooting is completed. Beauty treatment refers to a method of beautifying photos. After beauty treatment, the characters in the photos will look more in line with human aesthetics.

Contents of the invention

Embodiments of the present application provide an image processing method, device, computer-readable storage medium, and electronic device, which can improve the accuracy of image processing.

An image processing method, the method comprising:

Obtain the image to be processed and the corresponding sharpness;

Obtaining a corresponding beautification parameter model according to the definition, the beautification parameter model refers to a model used to calculate the beautification parameter;

Acquiring corresponding target beauty parameters according to the definition and beauty parameter model;

Perform beautification processing on the image to be processed according to the target beautification parameters.

An image processing device, the device comprising:

The image acquisition module is used to acquire the image to be processed and the corresponding definition.

A model acquisition module, configured to acquire a corresponding beautification parameter model according to the clarity, and the beautification parameter model refers to a model used to calculate the beautification parameter.

A parameter acquisition module, configured to acquire corresponding target beauty parameters according to the definition and beauty parameter models.

A beautification processing module, configured to perform beautification processing on the image to be processed according to the target beautification parameters.

A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

Obtain the image to be processed and the corresponding sharpness;

Obtaining a corresponding beautification parameter model according to the definition, the beautification parameter model refers to a model used to calculate the beautification parameter;

Acquiring corresponding target beauty parameters according to the definition and beauty parameter model;

Perform beautification processing on the image to be processed according to the target beautification parameters.

An electronic device, comprising a memory and a processor, wherein computer-readable instructions are stored in the memory, and when the instructions are executed by the processor, the processor is made to perform the following steps:

Obtain the image to be processed and the corresponding sharpness;

Obtaining a corresponding beautification parameter model according to the definition, the beautification parameter model refers to a model used to calculate the beautification parameter;

Acquiring corresponding target beauty parameters according to the definition and beauty parameter model;

Perform beautification processing on the image to be processed according to the target beautification parameters.

The above image processing method, device, computer-readable storage medium, and electronic device obtain the corresponding beauty parameter model according to the definition of the image to be processed, obtain the corresponding target beauty parameter according to the definition and the beauty parameter model, and obtain the corresponding target beauty parameter according to the acquired The target beautification parameter performs beautification processing on the image to be processed. In this way, different beautification processes can be performed according to images of different resolutions, which improves the accuracy of image processing and optimizes the beautification process.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present application. Those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is an application environment diagram of an image processing method in an embodiment;

Fig. 2 is a flowchart of an image processing method in an embodiment;

Fig. 3 is the flowchart of image processing method in another embodiment;

Fig. 4 is a color histogram generated in an embodiment;

Fig. 5 is a change curve diagram of the beauty coefficient in an embodiment;

Fig. 6 is the flow chart of image processing method in another embodiment;

Fig. 7 is a change curve diagram of the beautification coefficient in another embodiment;

Fig. 8 is a schematic structural diagram of an image processing device in an embodiment;

FIG. 9 is a schematic diagram of an image processing circuit in one embodiment.

Detailed ways

In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, and are not intended to limit the present application.

It can be understood that the terms "first", "second" and the like used in this application may be used to describe various elements herein, but these elements are not limited by these terms. These terms are only used to distinguish one element from another element. For example, a first acquisition module could be termed a second acquisition module, and, similarly, a second acquisition module could be termed a first acquisition module, without departing from the scope of the present application. Both the first acquisition module and the second acquisition module are acquisition modules, but they are not the same acquisition module.

Fig. 1 is an application environment diagram of an image processing method in an embodiment. As shown in FIG. 1 , the application environment includes a user terminal 102 and a server 104 . The user terminal 102 can be used to collect images to be processed, and send the images to be processed to the server 104 . After receiving the image to be processed, the server 104 acquires the image to be processed and the corresponding definition; obtains the corresponding beauty parameter model according to the definition, and the beauty parameter model refers to a model used to calculate the beauty parameter; The parameter model obtains the corresponding target beautification parameters; performs beautification processing on the image to be processed according to the target beautification parameters. Finally, the server 104 returns the image to be processed after the beautification process to the user terminal 102 . It can be understood that the server may also send the acquired target beautification parameters to the user terminal 102, and the user terminal 102 performs beautification processing on the image to be processed according to the target beautification parameters. Among them, the user terminal 102 is an electronic device located at the outermost edge of the computer network, mainly used for inputting user information and outputting processing results, such as a personal computer, a mobile terminal, a personal digital assistant, a wearable electronic device, and the like. The server 104 is a device for responding to service requests and providing computing services, such as one or more computers. It can be understood that, in other embodiments provided in this application, the application environment of the image processing method may only include the user terminal 102, that is, the user terminal 102 is used to collect the image to be processed, and to perform beauty treatment on the image to be processed .

Fig. 2 is a flowchart of an image processing method in an embodiment. As shown in FIG. 2 , the image processing method includes steps 202 to 208 . in:

Step 202, acquire the image to be processed and the corresponding sharpness.

In one embodiment, the image to be processed refers to an image that requires beauty treatment. The image to be processed may be collected by the mobile terminal. A camera that can be used for shooting is installed on the mobile terminal, and the user can initiate a photographing instruction through the mobile terminal. After detecting the photographing instruction, the mobile terminal collects and shoots an image through the camera. The mobile terminal will store the collected images to form an image collection. It can be understood that the image to be processed may also be acquired through other means, which is not limited here. For example, the image to be processed may also be downloaded from a webpage, or imported from an external storage device, and so on. Acquiring the image to be processed may specifically include: receiving a beautification instruction input by the user, and obtaining the image to be processed according to the beautification instruction, wherein the beautification instruction includes an image identifier. The image identification refers to a unique identification for distinguishing different images to be processed, and the image to be processed is obtained according to the image identification. For example, the image identifier may be one or more of image name, image code, image storage address, and the like. Specifically, after the mobile terminal acquires the image to be processed, it may perform beautification processing locally on the mobile terminal, or may send the image to be processed to the server for beautification processing. Clarity refers to the clarity of the texture and boundary in the image. The higher the clarity, the more clearly the detailed texture information in the image can be seen.

In step 204, a corresponding beauty parameter model is obtained according to the definition, and the beauty parameter model refers to a model used to calculate the beauty parameter.

Specifically, the beautification parameter model refers to a model used to calculate the beautification parameters. Generally speaking, the beautification parameter model can be expressed as a function model, and the function model can represent the functional relationship between input variables and output results. It can be understood that this functional relationship can be linear or non-linear. Input the input variable into the function model, and the output result corresponding to the output can be obtained. For example, the function model can be expressed as Y=X+1, where Y is the output result and X is the input variable. Then when the input variable X is 1, the corresponding output result Y is 2. The corresponding relationship between the sharpness and the beauty parameter model is established in advance, and the corresponding beauty parameter model can be obtained according to the sharpness. When the definition is different, the acquired beauty parameter models may be different, so that different beauty parameter models are used to calculate target beauty parameters according to different resolutions.

Step 206, obtain corresponding target beauty parameters according to the definition and beauty parameter models.

Generally speaking, sharpness can indicate the clarity of detail information in the image. If the sharpness of the image is too low, it means that a lot of detail information is lost in the image. However, when the image is beautified, the detailed information of the image is often affected to a certain extent. For example, when dermabrasion processing is performed on an image, the texture of the skin in the image will become smoother, and at the same time, the texture of the hair strands will become unclear, or the facial features will become blurred. If the definition of the image itself is not high, when the image is beautified, it is more likely to cause the image to lose more detail information, the image distortion is more serious, and the image loses its beauty. Therefore, in order to beautify the image more accurately, the corresponding relationship between the image definition and the target beauty parameter can be established in advance, and the corresponding target beauty parameter can be obtained according to the definition.

In the embodiments provided in this application, the target beautification parameters refer to parameters for performing beautification processing on the image to be processed. Beautification refers to a method of beautifying images. For example, the beautification process may be whitening or skin smoothing on the portrait in the image, or it may refer to performing cosmetic, face-lifting, and body-slimming treatments on the portrait. A beauty parameter model can represent the functional relationship between sharpness and beauty parameters. After obtaining the sharpness and beauty parameter model, the sharpness is used as the input variable of the beauty parameter model, and the calculation is performed through the beauty parameter model, and the output result obtained is the target beauty parameter.

Step 208, perform beautification processing on the image to be processed according to the target beautification parameters.

After the target beautification parameters are acquired, the image to be processed is beautified according to the target beautification parameters. It can be understood that the beautification process can be performed on the entire image, or only on a certain area in the image. For example, the whitening process can be performed on the entire image to increase the brightness of the entire image, or it can be performed only on the skin area, and the face thinning process can only be performed on the face area. It can be understood that the image to be processed is composed of several pixels, each pixel may be composed of multiple color channels, and each color channel represents a color component. For example, an image can be composed of three channels of RGB (Red Green Blue, red, green, blue), or it can be composed of three channels of HSV (Hue Saturation Value, hue, saturation, lightness), or it can be composed of CMY (CyanMagenta Yellow , magenta, cyan, yellow) composed of three channels. Then, when performing beautification processing on the image to be processed, beautification processing may be performed on each color channel of the image to be processed, and the processing degree of each color channel may be different. Specifically, obtain the definition corresponding to each color channel of the image to be processed; obtain the corresponding beauty parameter model according to the definition of each color channel; obtain the corresponding target beauty parameters of each color channel according to the definition and the beauty parameter model ; Perform beauty treatment on each color channel of the image to be processed according to the target beauty parameters.

In the image processing method provided in the above embodiments, the corresponding beauty parameter model is obtained according to the definition of the image to be processed, the corresponding target beauty parameter is obtained according to the definition and the beauty parameter model, and the image to be processed is obtained according to the acquired target beauty parameter model. Perform beauty treatment. In this way, different beautification processes can be performed according to images of different resolutions, which improves the accuracy of image processing and optimizes the beautification process.

Fig. 3 is a flowchart of an image processing method in another embodiment. As shown in FIG. 3 , the image processing method includes step 302 to step 310 . in:

Step 302, acquire the target area in the image to be processed, and calculate the sharpness corresponding to the target area.

If the beautification process is performed on the server, each mobile terminal may send the image to be processed to the server, and after receiving the image to be processed, the server performs beautification processing on the image to be processed among the images to be processed. When the mobile terminal sends the image to be processed, it sends the corresponding terminal ID at the same time. After the server completes the processing, it searches for the corresponding mobile terminal according to the terminal ID, and sends the image to be processed after the processing is completed to the mobile terminal. Wherein, the terminal identifier refers to a unique identifier of the user terminal. For example, the terminal identifier may be at least one of an IP (Internet Protocol, protocol for interconnecting networks) address, a MAC (Media Access Control, media access control) address, and the like.

The target area is generally an area that the user pays more attention to, and specifically may refer to an area that needs to be beautified in the image to be processed. For example, the target area may refer to a face area, a portrait area, a skin area, a lip area, a hair area, etc., which are not limited herein. It can be understood that the image to be processed is composed of several pixels, and the target area is composed of several pixels in the image to be processed. The area area corresponding to the target area refers to the size of the area occupied by the target area, which can be expressed as the total number of pixels contained in the target area, or as the area ratio of the target area to the corresponding image to be processed. The target area generally appears as an independent connected area in the image, and the connected area refers to a closed area. For example, each face in the image corresponds to an independent connected region. When there are multiple faces in the image, there will be multiple connected regions, and each face corresponds to an independent connected region.

Obtaining the target area may include: detecting a face area in the image to be processed, and acquiring a corresponding target area according to the face area. Specifically, the target area may refer to a face area, portrait area, skin area, lip area, hair area, facial features area, etc. in the image to be processed, which is not limited in this embodiment. The face area refers to the area where the face is located in the image to be processed. The face area of the image to be processed can be obtained through a face detection algorithm. The face detection algorithm can include a detection method based on geometric features, an eigenface detection method, a linear discriminant analysis method, a detection method based on a hidden Markov model, etc. Herein No limit. The portrait area refers to the area where the entire portrait in the image to be processed is located. The method for obtaining the portrait region may specifically include: obtaining depth information of the image to be processed; detecting the face region in the image to be processed, and obtaining the portrait region in the image to be processed according to the face region and depth information. It can be understood that when an image is captured by the image capture device, a depth map corresponding to the image can be acquired at the same time, and pixels in the depth map correspond to pixels in the image. The pixels in the depth map represent the depth information of the corresponding pixels in the image, and the depth information is the depth information from the object corresponding to the pixels to the image acquisition device. It is generally believed that the portrait and the face are on the same vertical plane, and the depth information from the portrait to the image acquisition device is within the same range as the depth information from the face to the image acquisition device. Therefore, after obtaining the face area, the depth information corresponding to the face area can be obtained from the depth map, and then the depth information corresponding to the portrait area can be obtained according to the depth information corresponding to the face area, and then according to the depth information corresponding to the portrait area, that is The portrait area in the image to be processed can be obtained.

Specifically, the skin area refers to the area where the skin is located. The skin area can be divided into a face skin area and a portrait skin area. The face skin area refers to the area where the face skin is located, and the portrait skin area includes the area where the face and torso skin are located. The method for obtaining the corresponding human face skin area according to the human face area may specifically include: generating a color histogram according to the color information corresponding to the human face area; obtaining the peak value and the corresponding color interval in the color histogram; dividing the skin color interval according to the color interval , taking the area corresponding to the skin color interval in the face area as the face skin area. Then, the area corresponding to the skin color interval in the portrait area is taken as the portrait skin area. The color histogram is used to describe the proportion of different colors in the face area, and the color information refers to related parameters used to represent the color of the image. For example, in the HSV color space, the color information may include information such as H (Hue, hue), S (Saturation, saturation) and V (Value, lightness) of the color in the image. To obtain the color information corresponding to the face area, the color information can be divided into multiple small color intervals, and the number of pixels falling into each color interval in the face area is calculated respectively, so as to obtain the color histogram. Wherein, the color histogram may be an RGB color histogram, an HSV color histogram, or a YUV color histogram, etc., but is not limited thereto. In the HSV color space, the components can include H (Hue, hue), S (Saturation, saturation) and V (Value, lightness). Calculated counterclockwise, red is 0°, green is 120°, and blue is 240°; S indicates the degree of color close to spectral color, the larger the proportion of spectral color, the higher the degree of color close to spectral color, the color The higher the saturation, the higher the saturation, the color is generally deep and bright; V indicates the degree of brightness of the color. For the color of the light source, the brightness value is related to the brightness of the illuminant; for the object color, this value is related to the transmittance of the object or Relevant to reflectance, V usually ranges from 0% (black) to 100% (white). Specifically, the method for generating the HSV color histogram may include: converting the face area from the RGB color space to the HSV color space; respectively quantizing the three components of H, S, and V in the HSV, and converting the quantized H, The three components of S and V synthesize a one-dimensional feature vector; according to the value of each pixel in the face area in the HSV color space, determine the quantization level of the pixel in the three components of H, S, and V; according to the quantization of the pixel The level calculates the corresponding eigenvector, and counts the number of pixels distributed on each quantization level according to the eigenvector; generates a color histogram according to the statistical result.

Wherein, the value of the feature vector can be between 0 and 255, a total of 256 values, that is, the HSV color space can be divided into 256 color intervals, and each color interval corresponds to a value of the feature vector. For example, the H component can be quantized to 16 levels, the S component and the V component can be quantized to 4 levels respectively, and the synthesized one-dimensional feature vector can be shown in the following formula:

L＝H*Q _S *Q _V +S*Q _V +V;

L represents the one-dimensional feature vector synthesized by the quantized H, S, and V components; Q _S represents the quantization level of the S component, and Q _V represents the quantization level of the V component. The peak refers to the maximum value of the wave amplitude in a section formed by the color histogram, which can be determined by calculating the first-order difference of each point in the color histogram, and the peak value is the maximum value on the peak. After the peak value in the color histogram is obtained, a quantized color interval corresponding to the peak value is obtained, and the color interval may be a value of an eigenvector corresponding to the peak value in the HSV color space. Figure 4 is a color histogram generated in one embodiment. As shown in FIG. 4 , the vertical axis of the color histogram represents the distribution of pixels, that is, the number of pixels corresponding to the color interval. The horizontal axis represents the feature vector of the HSV color space, that is, multiple color intervals divided by the HSV color space. It can be seen that the color histogram in FIG. 4 includes peak 402, the peak value corresponding to peak 402 is 850, and the color interval corresponding to this peak value is 150, that is, there are 850 pixels in the statistical image with eigenvector values of 150.

The skin color interval of the face area is divided according to the color interval corresponding to the peak value of the color histogram, the range value of the skin color interval can be preset, and the skin color interval is calculated according to the color interval corresponding to the peak value and the preset range value. The area corresponding to the skin color interval in the face area is taken as the face skin area. Optionally, the computer device can multiply the color interval corresponding to the peak value by a preset range value, wherein the preset range value can include an upper limit value and a lower limit value, and the color interval corresponding to the peak value can be divided by the upper limit value and the upper limit value respectively. Multiply the value and the lower limit value to get the skin color range. For example, the computer device can pre-set the skin color range as 80%-120%. If the color range corresponding to the peak value of the color histogram is 150, the skin color range can be calculated as 120-180. The computer device can obtain the feature vector of each pixel in the face area in the HSV color space, and judge whether the feature vector falls into the skin color interval, and if it falls, the corresponding pixel can be defined as the pixel of the face skin area. For example, if the calculated skin color range is 120-180, then the computer device can define the pixel points in the face area with feature vectors between 120-180 in the HSV color space as pixels in the face skin area.

Step 304, determine the parameter interval where the sharpness is located, and obtain the beautification parameter model corresponding to the parameter interval.

Obtain the sharpness corresponding to the target area, and obtain the target beautification parameters according to the sharpness of the target area. Generally speaking, the target area is the area that the user pays more attention to, and the beautification parameters obtained according to the definition of the target area are more accurate. If the definition of the target area is relatively low, it is easy to cause loss of detail information after beautification processing, reducing the beauty of the image. When performing beautification processing, you can consider reducing the degree of beautification processing accordingly, so as to better retain some detail information without causing image distortion. Specifically, the sharpness is divided into different parameter intervals, and when the sharpness is in different parameter intervals, different degrees of beautification processing are performed correspondingly. For example, the value of the sharpness can be 0 to 1, and the larger the value, the clearer the image. The sharpness is divided into different intervals, and each interval corresponds to a beauty parameter model. Assume that the sharpness is divided into four intervals of 0-0.2, 0.2-0.6, 0.6-0.8, and 0.8-1, corresponding to model 1, model 2, model 3, and model 4 respectively. Then when the sharpness of the target area is 0.5, the obtained beauty parameter model is model 2.

Step 306, acquire the basic parameters of beautification.

Step 308, calculate the beautification coefficient according to the definition and the beautification parameter model, and obtain the target beautification parameter according to the basic beautification parameters and the beautification coefficient.

Specifically, there may be multiple target areas in the image to be processed. For example, if there are multiple faces in the image to be processed, the area where each face is located is regarded as an independent target area, and multiple faces correspond to multiple target areas. . The basic beautification parameter refers to a reference value for beautification processing. The basic beautification parameter may be a fixed parameter value preset by the user, or may be corresponding to a target area. The corresponding basic beautification parameters are obtained according to the target area, and different target areas may correspond to the same basic beautification parameter, or may correspond to different basic beautification parameters. Calculate the beautification coefficient according to the definition and beautification parameter model, and then obtain the target beautification parameters according to the basic beautification parameters and beautification coefficient. Wherein, the beautification coefficient refers to the weight for obtaining the beautification parameter.

For example, suppose the basic beautification parameter is Param, the beautification coefficient is Factor, and the calculated definition of the target area is Clarity. The clarity Clarity is divided into three parameter areas, namely Clarity<stdClarity _min , stdClarity _min <Clarity<stdClarity _max , Clarity>stdClarity _max . The formula for calculating the beauty coefficient is as follows:

The beautification coefficient can be calculated according to the above formula, and then the target beautification parameter adjustParam can be calculated according to the basic beautification parameters and beautification coefficient, as follows:

adjustParam=Param*Factor ^T

Fig. 5 is a graph showing the variation of the beautification coefficient in an embodiment. As shown in Figure 5, the beauty coefficient changes in a stepwise manner, which is divided into three stages. The first stage: when Clarity<stdClarity _min , the beauty coefficient remains unchanged; the second stage: when stdClarity _min <stdClarity min When Clarity<stdClarity _max , the beautification factor and the sharpness increase linearly; the third stage: when Clarity>stdClarity _max , the beautification factor remains unchanged.

Step 310, perform beautification processing on the image to be processed according to the target beautification parameters.

Specifically, the beautification process is performed on the target area in the image to be processed according to the target beautification parameters. After the target area is obtained, an area identification can be established for each target area, and then the relationship between the area identification, position coordinates and target beauty parameters can be established, and the corresponding target area can be obtained according to the position coordinates, and then the corresponding target beauty parameters can be obtained according to the corresponding target area. Beautify the target area. For example, it is detected that the to-be-processed image "pic.jpg" contains three face areas, including face 1, face 2, and face 3, and the corresponding face identifiers are face1, face2, and face3 respectively, and the corresponding target beautification parameters They are level 1 whitening, level 2 whitening and level 1 acne removal.

In the image processing method provided by the above-mentioned embodiments, firstly, the target area in the image to be processed is obtained, and a beautification parameter model is obtained according to the definition corresponding to the target area in the image to be processed. Then obtain the corresponding target beauty parameters according to the definition and beauty parameter model, and then perform beauty processing on the image to be processed according to the acquired target beauty parameters. In this way, different beautification processes can be performed according to different resolutions of the target areas in the image, which improves the accuracy of image processing and optimizes the beautification process.

Fig. 6 is a flowchart of an image processing method in yet another embodiment. As shown in FIG. 6 , the image processing method includes step 602 to step 612 . in:

Step 602, acquire the image to be processed and the corresponding sharpness.

There may be one or more target areas in the image to be processed. For example, there may be one human face or multiple human faces in the image to be processed, and the area where the human face is located is used as the target area. It can be understood that there may be no target area in the image to be processed. The target area in the image to be processed can be obtained by region marking, or directly detected in the image to be processed. The area mark refers to a mark used to indicate the range of the target area in the image to be processed. For example, the target area is marked with a red rectangular frame in the image to be processed, and the area in the red rectangular frame is considered as the target area. Separately extract the target area in the image to be processed, and establish a corresponding relationship with the image identification and position coordinates of the image to be processed. If there are multiple target areas in an image to be processed, each target area is individually associated with the image identification Establish a corresponding relationship with the position coordinates. Wherein, the position coordinate refers to the coordinate indicating the position of the target region in the image to be processed. For example, the position coordinates may be the coordinates of the position of the center of the target area in the image to be processed, or the coordinates of the position of the upper left corner in the image to be processed. After the target area is processed, the corresponding image to be processed is found through the image identification, and then the specific position of the target area in the image to be processed is found through the position coordinates, and the target area is restored.

Step 604: Acquire a corresponding beauty parameter model according to the definition, where the beauty parameter model refers to a model used to calculate the beauty parameter.

In an embodiment, the algorithm for calculating sharpness may include a spatial domain gradient algorithm, a frequency domain analysis method, and the like. Common spatial domain gradient algorithms include Brenner algorithm, Tenengrad algorithm, SMD algorithm and other algorithms. The frequency domain analysis method can calculate the sharpness by counting the high-frequency components in the frequency domain. The higher the high-frequency components, the clearer the image. Taking the Tenengrad algorithm as an example, the Sobel gradient operator is used to calculate the gradient values in the horizontal and vertical directions respectively, and the image definition based on Tenengrad is defined as follows:

Among them, T is a given edge detection threshold, Gx and Gy are the convolution of the Sobel horizontal and vertical edge detection operators at the pixel point (x, y) respectively, and the edge can be detected by the following Sobel gradient operator template:

Step 606: Obtain the character attribute features corresponding to the face area in the image to be processed, and obtain the corresponding basic beautification parameters according to the character attribute features, where the basic beautification parameters include the first basic beautification parameter and the second basic beautification parameter.

The character attribute feature refers to the feature representing the character attribute of the person in the image, for example, the character attribute feature may refer to one or more of gender characteristics, age characteristics, ethnic characteristics, and the like. Specifically, the face area in the image to be processed is acquired, and then the corresponding character attribute features are identified according to the face area. Furthermore, the face area in the image to be processed is obtained, and the character attribute features corresponding to the face area are obtained through the feature recognition model. Wherein, the feature recognition model refers to a model that recognizes the attribute characteristics of a person, and the feature recognition model is obtained through training a set of face samples. A face sample set refers to an image set composed of several face images. The feature recognition model is obtained by training the face sample set. Generally, the more face images in the face sample set, the more accurate the trained feature recognition model is. . For example, in supervised learning, each face image in the face sample set is marked with a corresponding label to mark the type of face image, and a feature recognition model can be obtained by training the face sample set. The feature recognition model can classify the face area to obtain the corresponding character attribute features. For example, if the face area can be divided into yellow race, black race and white race, then the obtained corresponding character attribute feature is one of yellow race, black race or white race. That is, classification by feature recognition models is based on the same criteria. If you want to obtain the character attribute features of different dimensions in the face area, you can obtain them separately through different feature recognition models.

Specifically, the character attribute features may include race feature parameters, gender feature parameters, age feature parameters, skin color feature parameters, skin quality feature parameters, face shape feature parameters, and makeup feature parameters, which are not limited here. For example, the race characteristic parameters corresponding to the face region are obtained through the race recognition model, the age characteristic parameters corresponding to the face region are obtained according to the age recognition model, and the gender characteristic parameters corresponding to the face region are obtained according to the gender recognition model. Pre-establish the relationship between the character attribute characteristics and the basic beauty parameters, and obtain the corresponding basic beauty parameters according to the character attributes. For example, the character attribute features may include male and female. When the recognized face is male, the corresponding basic parameters of beautification are skin smoothing. When the recognized face is female, the corresponding basic parameters of beautification are whitening. The correspondence between character attributes and basic beauty parameters can be set by the user, or learned by the system through big data.

Step 608: Calculate the beautification coefficient according to the definition and the beautification parameter model, obtain the corresponding first beautification parameter according to the first beautification basic parameter and the beautification coefficient, and use the second beautification basic parameter as the second beautification parameter.

Beautification treatments can include different treatments such as skin smoothing, face slimming, big eyes, freckle removal, dark circles removal, whitening and brightening, sharpening, red lips, bright eyes, etc. Some of the treatments will cause the loss of detailed information and will be subject to clear The effect of sharpness, and some processing will not cause the loss of detail information, so it is not directly related to sharpness. For example, skin resurfacing, face thinning, big eyes, freckle removal, dark circles removal, etc. will change the size of the face, and will change the details of the image. , thin face and other processing may cause the image to lose more detailed information, resulting in serious image distortion; while whitening and brightening, sharpening, red lips, bright eyes and other processing are processing color information, and will not affect the image. Details. The target beautification parameters may include the first beautification parameter and the second beautification parameter. The first beautification parameter will be affected by the sharpness, and there is a corresponding relationship with the sharpness. The second beautification parameter will not be affected by the sharpness. influences. Specifically, the basic beauty parameters include first basic beauty parameters and second basic beauty parameters. The beautification coefficient is calculated according to the definition and the beautification parameter model, the corresponding first beautification parameter is obtained according to the first beautification basic parameter and the beautification coefficient, and the second beautification basic parameter is used as the second beautification parameter.

Furthermore, when the sharpness is lower than a certain value, the target area may not be beautified, or a part of the treatment may be performed. For example, only a small degree of whitening process is applied to the target area, which can ensure that the processed image will not be too distorted. Then a parameter threshold can be set, and the beautification factor can be divided into a first beautification factor and a second beautification factor. When the sharpness is greater than the parameter threshold, the first beautification coefficient is calculated according to the sharpness and the beautification parameter model, and the corresponding first beautification parameter is obtained according to the first beautification basic parameter and the first beautification coefficient, and the second beautification parameter is calculated. The basic color parameter is used as the second beauty parameter. When the clarity is less than the parameter threshold, obtain the first beauty coefficient and the second beauty coefficient, obtain the first beauty parameter according to the first beauty coefficient and the first basic parameter of beauty, and obtain the first beauty parameter according to the second beauty coefficient and the first Second beautification basic parameter Get the second beautification parameter. Wherein, the first beautification parameter and the second beautification parameter are set to a relatively small value, then the first beautification coefficient and the second beautification coefficient obtained finally will be relatively small values.

Step 610, acquiring target beauty parameters according to the first beauty parameters and the second beauty parameters.

In one embodiment, assume that the basic beauty parameter is defaultParam, where the basic beauty parameter is divided into a first basic beauty parameter adjustParam and a second basic beauty parameter unchangedParam. Beautification treatments can include skin resurfacing, face-lifting, big eyes, freckle removal, dark circles removal, whitening and brightening, sharpening, red lips, eye brightening, etc., then:

defaultParam = [adjustParam|unchangedParam];

adjustParam = [softenP, faceSlenderP, eyeLargerP, deblemishP, depouchP];

unchangedParam = [skinBrightenP, sharpP, lipP, eyeBrightenP];

Among them, softenP represents the basic parameters of skin smoothing, faceSlenderP represents the basic parameters of thin face, eyeLargerP represents the basic parameters of large eyes, deblemishP represents the basic parameters of freckle removal, and depouchP represents the basic parameters of removing dark circles. skinBrightenP is the basic parameters for whitening and brightening, sharpP is the basic parameters for sharpening, lipP is the basic parameters for red lips, and eyeBrightenP is the basic parameters for brightening eyes.

Assuming that the beautification factor is Factor, the beautification factor includes a first beautification factor adjustFactor and a second beautification factor unchangedFactor. The first beautification parameter can be obtained according to the first beautification coefficient and the first basic beautification parameter, and the second beautification parameter can be obtained according to the second beautification coefficient and the second basic beautification parameter.

Factor = [adjustFactor|unchangedFactor];

adjustFactor = [softenF, faceSlenderF, eyeLargerF, deblemishF, depouchF];

unchangedFactor = [skinBrightenF, sharpF, lipF, eyeBrightenF];

Among them, softenF represents the skin smoothing coefficient, faceSlenderF represents the face thinning coefficient, eyeLargerF represents the large eye coefficient, deblemishF represents the freckle removal coefficient, and depouchF represents the dark circle removal coefficient. skinBrightenF is a whitening and brightening factor, sharpF is a sharpening factor, lipF is a red lip factor, and eyeBrightenF is a brightening factor.

If the acquired clarity of the image to be processed is Clarity, the value range of the best clarity can be analyzed through big data, or the value range can be customized by the user. Assume that the value range for defining the best clarity is [stdClarity _min , stdClarity _max ]. The coefficient adjustFactor is [1,1,1,1,1]. When the sharpness exceeds this value range, a function can be used to obtain the corresponding beautification coefficient. Generally, the smaller the sharpness is, the smaller the first beautification coefficient is, and the larger the sharpness is, the larger the first beautification coefficient is. In order to prevent the beautification process from being too deep or too shallow, you can set a lower limit value minClarity and an upper limit value maxClarity for the sharpness. When the sharpness is greater than the upper limit value or lower than the lower limit value, the first beautification The value of the coefficient remains unchanged. Specifically, the upper limit and the lower limit can be customized, assuming that the lower limit and the upper limit are defined as minClarity=0.25*stdClarity _min and maxClarity=1.75*stdClarity _max respectively, then the first beauty coefficient can be obtained The formula is as follows:

When the clarity Clarity is greater than minClarity, the first beautification coefficient can be obtained according to the above formula, the first beautification parameter can be obtained according to the first beautification coefficient and the first beautification basic parameter, and the second beautification basic parameter can be directly used as The second beautification parameter, that is, the second beautification coefficient is [1, 1, 1, 1]. When the clarity Clarity is less than minClarity, you can choose not to perform beautification processing on the target area, or perform a small degree of beautification processing on the target area. Define the first beautification coefficient and the second beautification coefficient as smaller values, obtain the first beautification parameter according to the obtained first beautification coefficient and the first beautification basic parameter, and obtain the second beautification coefficient and The second beautification basic parameter obtains the second beautification parameter, then the obtained first beautification parameter and the second beautification parameter are also a smaller value. For example, when Clarity is smaller than minClarity, the first beauty factor can be adjustedFactor=[0.5,0,0,0,0], and the second beauty factor can be unchangedFactor=[0,0,0,0]. Obtain the corresponding target beauty parameters according to the obtained beauty coefficient and basic beauty parameters, then the target beauty parameter adjustParam is:

adjustParam=defaultParam*Factor ^T

Fig. 7 is a graph showing the variation of the beautification factor in another embodiment. As shown in Figure 7, when the clarity Clarity is greater than minClarity, the change of the first beautification coefficient increases stepwise, which is divided into four stages. The first stage: when minClarity<Clarity<stdClarity _min , the first beautification coefficient The coefficient and clarity increase linearly; the second stage: when stdClarity _min <Clarity<stdClarity _max , the first beauty coefficient remains unchanged; the third stage: when stdClarity _max <Clarity<maxClarity, the first beauty coefficient and The clarity increases linearly; the fourth stage: when Clarity>maxClarity, the first beauty coefficient remains unchanged.

Step 612, perform beautification processing on the image to be processed according to the target beautification parameters.

The system can include multiple beautifying modules, and each beautifying module can perform a kind of beautifying treatment. For example, the system may include a skin-removal module, a whitening module, an eye-enlargement module, a face-slimming module, and a skin tone adjustment module, which can respectively perform skin-removal processing, whitening processing, eye enlargement processing, face-slimming processing, and skin color adjustment processing on the image to be processed. In one embodiment, each beautification module may be a code function module, and the beautification process on the image is realized through the code function module. Each code function module corresponds to a flag bit, which is used to determine whether to perform corresponding processing. For example, each beauty module corresponds to a flag bit Stat. When the value of the flag bit Stat is 1 or true, it indicates that the beauty treatment corresponding to the beauty module needs to be performed; when the value of the flag bit Stat is When it is 0 or false, it means that the beauty treatment corresponding to the beauty module does not need to be performed.

Specifically, assign a value to the flag bits of each beauty-beautification module according to the target beauty-beautification parameters, obtain the beauty-beauty modules used for beauty-beauty processing according to the flag bits, and input the target beauty-beauty parameters into each acquired beauty-beauty module to be processed The image is beautified. For example, if the target beautification parameters include whitening the face, assign a value of 1 to the flag corresponding to the whitening module, and assign a value of 0 to the flag corresponding to the eye enlargement module if no eye-enlargement processing is required. When performing beautification processing, each beautification module is traversed to determine whether corresponding processing is required according to the flag bit. It is understandable that the beauty treatment performed by each beauty module is independent of each other and does not affect each other. Assuming that the image requires multiple beautification processes, each beautification module can be processed sequentially to obtain the final beautification image.

When there are multiple target areas in the image to be processed, each target area may be traversed, and the target beautification parameters corresponding to each target area may be obtained, and the target areas may be beautified according to the acquired target beautification parameters. If only the target area in the image to be processed is beautified, but the remaining area in the image to be processed is not processed, there may be a significant difference between the target area and the remaining area after processing. . For example, after whitening is performed on the target area, the brightness of the target area is obviously higher than that of the remaining area, which makes the image look unnatural. Then, in the generated beautification image, the boundary of the target area can be transitioned, so that the obtained beautification image looks more natural.

In the image processing method provided by the above embodiments, firstly, the sharpness of the image to be processed is obtained, and a beautification parameter model is obtained according to the sharpness. Then obtain the corresponding target beauty parameters according to the definition and beauty parameter model, and then perform beauty processing on the image to be processed according to the acquired target beauty parameters. In this way, different beautification processes can be performed according to different resolutions of the images, which improves the accuracy of image processing and optimizes the beautification process.

Fig. 8 is a schematic structural diagram of an image processing device in an embodiment. As shown in FIG. 8 , the image processing device 800 includes an image acquisition module 802 , a model acquisition module 804 , a parameter acquisition module 806 and a beautification processing module 808 . in:

The image acquisition module 802 is configured to acquire the image to be processed and the corresponding sharpness.

The model acquiring module 804 is configured to acquire a corresponding beautification parameter model according to the definition, and the beautification parameter model refers to a model used to calculate the beautification parameter.

A parameter acquisition module 806, configured to acquire corresponding target beauty parameters according to the definition and beauty parameter models.

A beautification processing module 808, configured to perform beautification processing on the image to be processed according to the target beautification parameters.

The image processing device provided in the above embodiment obtains the corresponding beauty parameter model according to the definition of the image to be processed, obtains the corresponding target beauty parameter according to the definition and the beauty parameter model, and obtains the image to be processed according to the acquired target beauty parameter model. Perform beauty treatment. In this way, different beautification processes can be performed according to images of different resolutions, which improves the accuracy of image processing and optimizes the beautification process.

In one embodiment, the image acquisition module 802 is further configured to acquire a target area in the image to be processed, and calculate the sharpness corresponding to the target area.

In one embodiment, the image acquiring module 802 is further configured to detect a face area in the image to be processed, and acquire a corresponding target area according to the face area.

In one embodiment, the model obtaining module 804 is further configured to determine the parameter range in which the sharpness is located, and obtain a beauty parameter model corresponding to the parameter range.

In one embodiment, the parameter acquiring module 806 is also used to acquire basic beautification parameters; calculate a beautification coefficient according to the definition and beautification parameter model, and obtain target beautification parameters according to the basic beautification parameters and beautification coefficients.

In one embodiment, the parameter acquiring module 806 is further configured to acquire the character attribute features corresponding to the face area in the image to be processed, and acquire the corresponding basic beautification parameters according to the character attribute features.

In one embodiment, the parameter acquisition module 806 is further configured to calculate the beautification coefficient according to the definition and the beautification parameter model, obtain the corresponding first beautification parameter according to the first beautification basic parameter and the beautification coefficient, and The second basic beautification parameter is used as a second beautification parameter; and a target beautification parameter is obtained according to the first beautification parameter and the second beautification parameter.

The division of each module in the above image processing device is for illustration only. In other embodiments, the image processing device may be divided into different modules as required to complete all or part of the functions of the above image processing device.

The embodiment of the present application also provides a computer-readable storage medium. One or more non-transitory computer-readable storage media containing a computer program that, when executed by one or more processors, causes the processors to:

Obtain the image to be processed and the corresponding sharpness;

Obtaining a corresponding beautification parameter model according to the definition, the beautification parameter model refers to a model used to calculate the beautification parameter;

Acquiring corresponding target beauty parameters according to the definition and beauty parameter model;

Perform beautification processing on the image to be processed according to the target beautification parameters.

In one embodiment, the acquiring the image to be processed and the corresponding sharpness performed by the processor includes:

A target area in the image to be processed is acquired, and a sharpness corresponding to the target area is calculated.

In one embodiment, the acquiring the target area in the image to be processed performed by the processor includes:

Detect the face area in the image to be processed, and obtain the corresponding target area according to the face area.

In one embodiment, the obtaining the corresponding beauty parameter model according to the definition performed by the processor includes:

Determine the parameter interval where the sharpness is located, and obtain a beautification parameter model corresponding to the parameter interval.

In one embodiment, the acquiring corresponding target beauty parameters according to the sharpness and beauty parameter model performed by the processor includes:

Obtain the basic parameters of beauty;

A beautification coefficient is calculated according to the definition and a beautification parameter model, and a target beautification parameter is obtained according to the basic beautification parameters and the beautification coefficient.

In one embodiment, the acquisition of basic beauty parameters performed by the processor includes:

Obtain the character attribute features corresponding to the face area in the image to be processed, and acquire the corresponding basic beautification parameters according to the character attribute features.

In one embodiment, the acquiring corresponding target beauty parameters according to the sharpness and beauty parameter model performed by the processor includes:

Calculate the beautification coefficient according to the definition and the beautification parameter model, obtain the corresponding first beautification parameter according to the first beautification basic parameter and the beautification coefficient, and use the second beautification basic parameter as the second beautification parameter;

Acquiring target beauty parameters according to the first beauty parameters and the second beauty parameters.

The embodiment of the present application also provides an electronic device. The above-mentioned electronic device includes an image processing circuit, and the image processing circuit may be implemented by hardware and/or software components, and may include various processing units defining an ISP (Image Signal Processing, image signal processing) pipeline. FIG. 9 is a schematic diagram of an image processing circuit in one embodiment. As shown in FIG. 9 , for ease of description, only various aspects of the image processing technology related to the embodiment of the present application are shown.

As shown in FIG. 9 , the image processing circuit includes an ISP processor 940 and a control logic 950 . Image data captured by imaging device 910 is first processed by ISP processor 940 , which analyzes the image data to capture image statistics that may be used to determine and/or control one or more parameters of imaging device 910 . Imaging device 910 may include a camera having one or more lenses 912 and an image sensor 914 . Image sensor 914 may include a color filter array (such as a Bayer filter), and image sensor 914 may obtain light intensity and wavelength information captured with each imaging pixel of image sensor 914 and provide a set of raw images that may be processed by ISP processor 940. image data. The sensor 920 (such as a gyroscope) may provide the collected image processing parameters (such as anti-shake parameters) to the ISP processor 940 based on the interface type of the sensor 920 . The interface of the sensor 920 may utilize a SMIA (Standard Mobile Imaging Architecture, Standard Mobile Imaging Architecture) interface, other serial or parallel camera interfaces, or a combination of the above interfaces.

Additionally, image sensor 914 may also send raw image data to sensor 920 , which may provide raw image data to ISP processor 940 based on the sensor 920 interface type, or sensor 920 may store raw image data in image memory 930 .

The ISP processor 940 processes raw image data on a pixel-by-pixel basis in various formats. For example, each image pixel may have a bit depth of 8, 10, 12, or 14 bits, and the ISP processor 940 may perform one or more image processing operations on raw image data, gather statistical information about the image data. Among other things, image processing operations can be performed with the same or different bit depth precision.

ISP processor 940 may also receive image data from image memory 930 . For example, the sensor 920 interface sends raw image data to the image memory 930, and the raw image data in the image memory 930 is provided to the ISP processor 940 for processing. The image memory 930 may be a part of a memory device, a storage device, or an independent dedicated memory in an electronic device, and may include a DMA (Direct Memory Access) feature.

When receiving raw image data from the image sensor 914 interface or from the sensor 920 interface or from the image memory 930, the ISP processor 940 may perform one or more image processing operations, such as temporal filtering. The processed image data may be sent to image memory 930 for additional processing before being displayed. The ISP processor 940 may also receive processed data from the image memory 930 on which to perform image data processing in the raw domain and in the RGB and YCbCr color spaces. The processed image data may be output to the display 980 for viewing by the user and/or further processed by a graphics engine or a GPU (Graphics Processing Unit, graphics processor). In addition, the output of the ISP processor 940 can also be sent to the image memory 930 , and the display 980 can read image data from the image memory 930 . In one embodiment, image memory 930 may be configured to implement one or more frame buffers. Also, the output of the ISP processor 940 may be sent to an encoder/decoder 970 for encoding/decoding image data. The encoded image data may be saved and decompressed prior to display on the display 980 device.

The steps for the ISP processor 940 to process the image data include: performing VFE (Video Front End, video front end) processing and CPP (Camera Post Processing, camera post processing) processing on the image data. VFE processing of image data may include correction of contrast or brightness of image data, modification of digitally recorded light state data, compensation processing of image data (such as white balance, automatic gain control, gamma correction, etc.), filter processing, etc. CPP processing of image data may include scaling the image, providing preview frames and recording frames for each path. Among them, CPP can use different codecs to process preview frames and record frames. The image data processed by the ISP processor 940 can be sent to the beautifying module 960 so as to perform beautifying processing on the image before being displayed. The beautification module 960 can perform beautification processing on image data, including: whitening, freckle removal, skin smoothing, face thinning, acne removal, eye enlargement, etc. Wherein, the beautifying module 960 may be a CPU (Central Processing Unit, central processing unit), a GPU, or a co-processor in the mobile terminal. The data processed by the beauty module 960 can be sent to the encoder/decoder 970 for encoding/decoding image data. The encoded image data may be saved and decompressed prior to display on the display 980 device. Wherein, the beautification module 960 may also be located between the encoder/decoder 970 and the display 980, that is, the beautification module performs beautification processing on the imaged image. The aforementioned encoder/decoder 970 may be a CPU, a GPU, or a coprocessor in a mobile terminal.

The statistics determined by the ISP processor 940 may be sent to the control logic 950 unit. For example, statistics may include image sensor 914 statistics such as auto exposure, auto white balance, auto focus, flicker detection, black level compensation, lens 912 shading correction, etc. Control logic 950 may include a processor and/or a microcontroller that executes one or more routines (e.g., firmware) that determine control parameters of imaging device 910 and ISP processing based on received statistical data. The control parameters of the device 940. For example, control parameters of imaging device 910 may include sensor 920 control parameters (eg, gain, integration time for exposure control), camera flash control parameters, lens 912 control parameters (eg, focal length for focus or zoom), or combinations of these parameters. ISP control parameters may include gain levels and color correction matrices for automatic white balance and color adjustment (eg, during RGB processing), as well as lens 912 shading correction parameters.

The image processing method provided in the above embodiment can be implemented by using the image processing technology in FIG. 9 .

A computer program product containing instructions, when running on a computer, causes the computer to execute the image processing method provided by the above embodiments.

Any reference to memory, storage, database, or other medium as used herein may include non-volatile and/or volatile memory. Suitable nonvolatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM), which acts as external cache memory. By way of illustration and not limitation, RAM is available in many forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Chain Synchlink DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The above-mentioned embodiments only express several implementation modes of the present application, and the description thereof is relatively specific and detailed, but should not be construed as limiting the patent scope of the present application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the scope of protection of the patent application should be based on the appended claims.

Claims (10)

1. A kind of 1. image processing method, it is characterised in that the described method includes：

   Obtain pending image and corresponding clarity；

   Corresponding U.S. face parameter model is obtained according to the clarity, U.S.'s face parameter model refers to be used to calculate U.S. face parameter Model；

   Corresponding target U.S. face parameter is obtained according to the clarity and U.S. face parameter model；

   U.S. face processing carries out the pending image according to target U.S. face parameter.
2. 2. image processing method according to claim 1, it is characterised in that the pending image and corresponding clear of obtaining Clear degree includes：

   The target area in pending image is obtained, and calculates the corresponding clarity in the target area.
3. 3. image processing method according to claim 2, it is characterised in that the target area obtained in pending image Domain includes：

   The human face region in pending image is detected, and corresponding target area is obtained according to the human face region.
4. 4. image processing method according to claim 1, it is characterised in that described corresponding according to clarity acquisition U.S. face parameter model includes：

   Determine the parameter section residing for the clarity, obtain the corresponding U.S. face parameter model in the parameter section.
5. 5. image processing method according to any one of claims 1 to 4, it is characterised in that described according to the clarity Obtaining corresponding target U.S. face parameter with U.S. face parameter model includes：

   Obtain U.S. face underlying parameter；

   U.S. face coefficient is calculated according to the clarity and U.S. face parameter model, and according to the U.S. face underlying parameter and U.S. face coefficient Obtain target U.S. face parameter.
6. 6. image processing method according to claim 5, it is characterised in that U.S.'s face underlying parameter that obtains includes：

   The corresponding character attribute feature of human face region in the pending image is obtained, and is obtained according to the character attribute feature Corresponding U.S.'s face underlying parameter.
7. 7. image processing method according to claim 5, it is characterised in that U.S.'s face underlying parameter includes the first U.S. face Underlying parameter and the second U.S. face underlying parameter；

   It is described to obtain corresponding target U.S. face parameter according to the clarity and U.S. face parameter model and include：

   U.S. face coefficient is calculated according to the clarity and U.S. face parameter model, according to the described first U.S. face underlying parameter and U.S. face system Number obtains the corresponding first U.S. face parameter, and using the described second U.S. face underlying parameter as the second U.S. face parameter；

   According to the described first U.S. face parameter and second U.S. face parameter acquiring target U.S. face parameter.
8. 8. a kind of image processing apparatus, it is characterised in that described device includes：

   Image collection module, for obtaining pending image and corresponding clarity；

   Model acquisition module, for obtaining corresponding U.S. face parameter model according to the clarity, U.S.'s face parameter model is Refer to the model for being used for calculating U.S. face parameter；

   Parameter acquisition module, for obtaining corresponding target U.S. face parameter according to the clarity and U.S. face parameter model；

   U.S. face processing module, for carrying out U.S. face processing to the pending image according to target U.S. face parameter.
9. 9. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the computer program quilt The image processing method as any one of claim 1 to 7 is realized when processor performs.
10. 10. a kind of electronic equipment, including memory and processor, computer-readable instruction is stored in the memory, it is described When instruction is performed by the processor so that the processor performs the image procossing as any one of claim 1 to 7 Method.