CN116245759A - Method, device, equipment and storage medium for evaluating uniformity of human face complexion - Google Patents
Method, device, equipment and storage medium for evaluating uniformity of human face complexion Download PDFInfo
- Publication number
- CN116245759A CN116245759A CN202310158756.8A CN202310158756A CN116245759A CN 116245759 A CN116245759 A CN 116245759A CN 202310158756 A CN202310158756 A CN 202310158756A CN 116245759 A CN116245759 A CN 116245759A
- Authority
- CN
- China
- Prior art keywords
- uniformity
- target detection
- detection area
- image
- saturation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 71
- 238000001514 detection method Methods 0.000 claims abstract description 239
- 238000012545 processing Methods 0.000 claims abstract description 31
- 230000000694 effects Effects 0.000 claims description 36
- 238000011156 evaluation Methods 0.000 claims description 30
- 230000002159 abnormal effect Effects 0.000 claims description 27
- 238000004590 computer program Methods 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 9
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 description 20
- 230000008569 process Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 210000004709 eyebrow Anatomy 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000012512 characterization method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 239000011148 porous material Substances 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000009827 uniform distribution Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
- G06T5/92—Dynamic range modification of images or parts thereof based on global image properties
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The application provides a method, a device, equipment and a storage medium for evaluating uniformity of human face skin color, and belongs to the technical field of image processing. The method comprises the following steps: determining a target detection area based on the input initial face image; acquiring a tone saturation brightness image corresponding to an initial face image; determining the uniformity value of the target detection area according to the values of the tone channel and the saturation channel of each pixel point of the target detection area in the tone saturation brightness image; and determining the overall uniformity result of the initial face image based on the uniformity value of the target detection area. The method and the device can improve accuracy and robustness of face skin color uniformity determination.
Description
Technical Field
The present application relates to the field of image processing technologies, and in particular, to a method, an apparatus, a device, and a storage medium for evaluating uniformity of skin color of a face.
Background
The uniformity of the skin color of the human face reflects the variation degree of the tone of the skin, and in order to enable a user to know the health degree of the skin color of the human face in time, the uniformity of the skin color is usually required to be obtained.
In the prior art, the adopted technical means mainly takes pixels in a face image as units, calculates indexes such as mean value, variance and the like of pixel values, calculates the uniformity of the face complexion by using a standardized method based on the indexes, and outputs a specific numerical value of the uniformity of the face complexion to a user as a result for viewing.
The pixel unit is affected by factors such as large pores and color spots in the face skin, so that errors exist in the calculation result, and the robustness is low.
Disclosure of Invention
The invention aims to provide a method, a device, equipment and a storage medium for evaluating the uniformity of human face skin color, which can improve the accuracy and the robustness of the determination of the uniformity of human face skin color.
Embodiments of the present application are implemented as follows:
in one aspect of the embodiments of the present application, a method for evaluating uniformity of skin color of a face is provided, including:
determining a target detection area based on the input initial face image;
acquiring a tone saturation brightness image corresponding to an initial face image;
determining the uniformity value of the target detection area according to the values of the tone channel and the saturation channel of each pixel point of the target detection area in the tone saturation brightness image;
And determining the overall uniformity result of the initial face image based on the uniformity value of the target detection area.
Optionally, determining the uniformity value of the target detection area according to the values of the hue channel and the saturation channel of each pixel point of the target detection area in the hue saturation brightness image includes:
according to the values of the tone channel and the saturation channel of each pixel point in the target detection area, determining a tone mean value and a saturation mean value corresponding to the target detection area;
and determining the uniformity value of the target detection area according to the values of the tone channel and the saturation channel of each pixel point in the target detection area, and the tone average value and the saturation average value corresponding to the target detection area.
Optionally, determining the uniformity value of the target detection area according to the values of the hue channel and the saturation channel of each pixel point in the target detection area, and the hue average value and the saturation average value corresponding to the target detection area includes:
determining the standard deviation of the tone channel according to the numerical value and the tone average value of the tone channel of each pixel point in the target detection area;
determining the standard deviation of the saturation channel according to the numerical value and the saturation mean value of the saturation channel of each pixel point in the target detection area;
And determining the uniformity value of the target detection area according to the standard deviation of the tone channel, the standard deviation of the saturation channel, the preset weight of the tone channel and the preset weight of the saturation channel.
Optionally, determining the uniformity value of the target detection area according to the values of the hue channel and the saturation channel of each pixel point in the target detection area and the hue mean saturation mean value corresponding to the target detection area includes:
determining a plurality of detection sub-areas in the target detection area based on the sliding window;
determining the uniformity value of each detection subarea according to the values of the tone channel and the saturation channel of each pixel point in each detection subarea and the tone average saturation average value corresponding to each detection subarea;
and determining the uniformity value of the target detection area according to the uniformity value of each detection subarea.
Optionally, determining the overall uniformity result of the initial face image based on the uniformity value in the target detection area includes:
and inputting the uniformity value in the target detection area into a target model to obtain an overall uniformity result of the initial face image, wherein the target model is a linear model obtained based on least square fitting.
Optionally, the method further comprises:
carrying out Gaussian blur processing on the initial face image to obtain a blurred image;
performing difference processing based on the initial face image and the blurred image to obtain a residual image;
performing scaling correction processing on the residual image according to the overall uniformity result of the initial face image to obtain a target effect image;
and outputting the target effect image as an evaluation result of the uniformity of the skin color of the human face.
Optionally, scaling correction processing is performed on the residual image according to the overall uniformity result of the initial face image, so as to obtain a target effect image, including:
respectively obtaining a global uniformity score and a local uniformity score according to the overall uniformity result of the initial face image, wherein the global uniformity score is used for representing the overall uniformity of the initial face image, and the local uniformity score is used for representing the uniformity of each demarcation region in the initial face image;
and scaling and correcting the residual image based on the global uniformity fraction and the local uniformity fraction to obtain the target effect image.
Optionally, determining the target detection area based on the input initial face image includes:
determining a plurality of face key points from the initial face image;
Determining a plurality of groups of target key points from the face key points, and establishing a Bezier curve based on each group of target key points;
taking the area surrounded by each Bezier curve as an initial target detection area;
and removing the abnormal region in the initial target detection region to obtain the target detection region.
Optionally, before removing the abnormal region in the initial target detection region to obtain the target detection region, the method further includes:
converting the initial target detection area into an image in tone saturation brightness format;
and taking the area of the average value of the brightness channels in the brightness format image of the tone saturation as an abnormal area.
In another aspect of the embodiments of the present application, an apparatus for evaluating uniformity of skin color of a face is provided, including: the device comprises a determining module, a converting module, a mean module and a result module;
the determining module is used for determining a target detection area based on the input initial face image;
the conversion module is used for acquiring a tone saturation brightness image corresponding to the initial face image;
the average module is used for determining the uniformity value of the target detection area according to the values of the tone channel and the saturation channel of each pixel point of the target detection area in the tone saturation brightness image;
And the result module is used for determining the overall uniformity result of the initial face image based on the uniformity value of the target detection area.
Optionally, the average module is specifically configured to determine a hue average value and a saturation average value corresponding to the target detection area according to values of hue channels and saturation channels of each pixel point in the target detection area; and determining the uniformity value of the target detection area according to the values of the tone channel and the saturation channel of each pixel point in the target detection area, and the tone average value and the saturation average value corresponding to the target detection area.
Optionally, the average module is specifically configured to determine a standard deviation of the tone channel according to a numerical value and a tone average value of the tone channel of each pixel point in the target detection area; determining the standard deviation of the saturation channel according to the numerical value and the saturation mean value of the saturation channel of each pixel point in the target detection area; and determining the uniformity value of the target detection area according to the standard deviation of the tone channel, the standard deviation of the saturation channel, the preset weight of the tone channel and the preset weight of the saturation channel.
Optionally, the mean module is specifically configured to determine a plurality of detection sub-areas in the target detection area based on the sliding window; determining the uniformity value of each detection subarea according to the values of the tone channel and the saturation channel of each pixel point in each detection subarea and the tone average saturation average value corresponding to each detection subarea; and determining the uniformity value of the target detection area according to the uniformity value of each detection subarea.
Optionally, the result module is specifically configured to input a uniformity value in the target detection area into a target model, so as to obtain an overall uniformity result of the initial face image, where the target model is a linear model obtained based on least square fitting.
Optionally, the result module is further configured to perform gaussian blur processing on the initial face image to obtain a blurred image; performing difference processing based on the initial face image and the blurred image to obtain a residual image; performing scaling correction processing on the residual image according to the overall uniformity result of the initial face image to obtain a target effect image; and outputting the target effect image as an evaluation result of the uniformity of the skin color of the human face.
Optionally, the result module is specifically configured to obtain a global uniformity score and a local uniformity score according to a global uniformity result of the initial face image, where the global uniformity score is used to represent global uniformity of the initial face image, and the local uniformity score is used to represent uniformity of each delimited area in the initial face image; and scaling and correcting the residual image based on the global uniformity fraction and the local uniformity fraction to obtain the target effect image.
Optionally, the determining module is specifically configured to determine a plurality of face key points from the initial face image; determining a plurality of groups of target key points from the face key points, and establishing a Bezier curve based on each group of target key points; taking the area surrounded by each Bezier curve as an initial target detection area; and removing the abnormal region in the initial target detection region to obtain the target detection region.
Optionally, the determining module is further configured to convert the initial target detection area into an image in a tone saturation brightness format; and taking the area of the average value of the brightness channels in the brightness format image of the tone saturation as an abnormal area.
In another aspect of the embodiments of the present application, there is provided a computer device comprising: the device comprises a memory and a processor, wherein the memory stores a computer program which can be run on the processor, and when the processor executes the computer program, the processor realizes the steps of the evaluation method of the uniformity of the human face skin color.
In another aspect of the embodiments of the present application, a computer readable storage medium is provided, on which a computer program is stored, which when executed by a processor, implements the steps of a method for evaluating the uniformity of skin tone of a human face.
The beneficial effects of the embodiment of the application include:
in the method, the device, the equipment and the storage medium for evaluating the uniformity of the complexion of the human face, the uniformity value of the target detection area can be determined by determining the target detection area based on the input initial human face image and acquiring the hue saturation brightness image corresponding to the initial human face image, further according to the hue channel and the saturation channel value of each pixel point of the target detection area in the hue saturation brightness image, after the uniformity value of the target detection area is obtained, the overall uniformity result of the initial human face image can be determined based on the uniformity value of the target detection area, and therefore the evaluation of the uniformity of the complexion of the human face is realized. According to the values of the tone channel and the saturation channel in the tone saturation brightness image, the uniformity of the skin color of the human face can be obtained more accurately and intuitively, the overall uniformity result of the initial human face image is determined based on the uniformity value of the target detection area, the calculation error can be reduced, and the accuracy and the robustness of the detection result are improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered limiting the scope, and that other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flow chart of a method for evaluating uniformity of skin color of a face according to an embodiment of the present application;
fig. 2 is another flow chart of an evaluation method for uniformity of skin color of a face according to an embodiment of the present application;
fig. 3 is another flow chart of an evaluation method for uniformity of skin color of a face according to an embodiment of the present application;
fig. 4 is another flow chart of an evaluation method for uniformity of skin color of a face according to an embodiment of the present application;
fig. 5 is another flow chart of an evaluation method for uniformity of skin color of a face according to an embodiment of the present application;
fig. 6 is another flow chart of an evaluation method for uniformity of skin color of a face according to an embodiment of the present application;
fig. 7 is another flow chart of an evaluation method for uniformity of skin color of a face according to an embodiment of the present application;
FIG. 8 is a schematic diagram of a specific location of a target detection area according to an embodiment of the present disclosure;
fig. 9 is another flow chart of an evaluation method for uniformity of skin color of a face according to an embodiment of the present application;
fig. 10 is a schematic structural diagram of an evaluation device for uniformity of skin color of a face according to an embodiment of the present application;
fig. 11 is a schematic structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present application, as provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
In the description of the present application, it should be noted that the terms "first," "second," "third," and the like are used merely to distinguish between descriptions and are not to be construed as indicating or implying relative importance.
The following specifically explains a specific implementation procedure of the method for evaluating the uniformity of the complexion of the face provided in the embodiment of the present application.
Fig. 1 is a flow chart of an evaluation method of skin color uniformity of a human face according to an embodiment of the present application, please refer to fig. 1, which includes:
s110: a target detection area is determined based on the input initial face image.
Alternatively, the execution subject of the method may be any computer device, such as: the mobile phone, the computer, the tablet computer, the professional electronic device, etc. are not particularly limited herein, and may be an application software program running in the computer device.
The initial face image may be an image obtained by a user through real-time photographing, or may also be an image obtained in advance, specifically may be a face image, and may include an integral part of a face.
Alternatively, the target detection area may be a part of the initial face image, specifically may be an area where a part selected based on actual requirements can highlight the face feature, and is not limited herein.
It should be noted that, due to the light or shielding in the whole face skin, there is a certain difference between the color in the image and the actual face skin color, so the target detection area with small difference can be found out through the above process, thereby ensuring the accuracy of the subsequent calculation.
S120: and acquiring a hue saturation brightness image corresponding to the initial face image.
Optionally, after the initial face image is obtained, the format of the initial face image may be converted into an HSV (Hue-Saturation-brightness) format, and after the conversion, a corresponding HSV image, that is, a Hue-Saturation-brightness image may be obtained.
In the actual execution process, the steps S110 and S120 are not limited in actual sequence, and may be executed sequentially or synchronously, and fig. 1 illustrates a sequential execution example, but is not limited in the embodiment of the present application.
Alternatively, the initial face image input may include an RGB image, that is, a conventional red, green, and blue image, or may be the HSV image directly, for example: if the image is an RGB image, the RGB image can be subjected to format conversion to obtain an HSV image; in the case of an HSV image, the image may be used directly.
It should be noted that, the input initial face image is converted into HSV space, so as to separate the V channel, i.e. the brightness channel, so as to avoid the influence of illumination variation on uniformity calculation, and make the result more robust. In the scheme, the skin color uniformity is obtained by numerical calculation of an H channel and an S channel: the values of the H channels characterize the change in hue, e.g. the change in red-yellow, the values of the S channels characterize the change in saturation, e.g. the light red to the dark red, the emphasis planes of the two channels being different.
S130: and determining the uniformity value of the target detection area according to the values of the hue channel and the saturation channel of each pixel point of the target detection area in the hue saturation brightness image.
Optionally, after the HSV image is obtained, values of a hue channel and a saturation channel of each pixel point in the target detection area corresponding to the HSV image may be determined, where the hue channel is an H channel, and the saturation channel is an S channel, and values of the H channel and the S channel of each pixel point in the target detection area may be determined.
And further, the uniformity value of the target detection area can be determined based on the values of the H channel and the S channel of each pixel point in the target detection area, where the uniformity value of the target detection area can be specifically a parameter representing the distribution condition of colors on the H channel and the S channel in the target detection area, and can be expressed as "hs_value".
S140: and determining the overall uniformity result of the initial face image based on the uniformity value of the target detection area.
Optionally, after obtaining the uniformity value of the target detection area, the overall uniformity result of the initial face image may be determined based on the uniformity value.
It should be noted that the overall uniformity result may be a scoring result specifically, which is used to represent the scoring result of the uniformity of the skin color of the face corresponding to the input initial face image, and may be a result calculated based on a linear model, which is not limited herein.
According to the method for evaluating the uniformity of the human face complexion, the target detection area can be determined based on the input initial human face image, the tone saturation brightness image corresponding to the initial human face image is obtained, the uniformity value of the target detection area can be determined according to the tone channel and the saturation channel value of each pixel point of the target detection area in the tone saturation brightness image, and after the uniformity value of the target detection area is obtained, the overall uniformity result of the initial human face image can be determined based on the uniformity value of the target detection area, so that the evaluation of the uniformity of the human face complexion is achieved. According to the values of the tone channel and the saturation channel in the tone saturation brightness image, the uniformity of the skin color of the human face can be obtained more accurately and intuitively, the overall uniformity result of the initial human face image is determined based on the uniformity value of the target detection area, the calculation error can be reduced, and the accuracy and the robustness of the detection result are improved.
The implementation process of determining the uniformity value in the method for evaluating the uniformity of the complexion of the human face provided in the embodiment of the present application is specifically explained below.
Fig. 2 is another flow chart of the method for evaluating the uniformity of the skin color of the face provided in the embodiment of the present application, please refer to fig. 2, which determines the uniformity value of the target detection area according to the values of the hue channel and the saturation channel of each pixel point of the target detection area in the hue saturation brightness image, including:
s210: and determining a tone average value and a saturation average value corresponding to the target detection area according to the values of the tone channel and the saturation channel of each pixel point in the target detection area.
Alternatively, the tone average value corresponding to the target detection area may be calculated according to the values of the tone channels of the pixel points in the target detection area
Specifically, the average value of the h values of all the pixel points may be obtained.
Correspondingly, the saturation mean value corresponding to the target detection area can be calculated according to the value of the saturation channel of each pixel point in the target detection area
Specifically, the average value of s values of all the pixel points may be used.
S220: and determining the uniformity value of the target detection area according to the values of the tone channel and the saturation channel of each pixel point in the target detection area, and the tone average value and the saturation average value corresponding to the target detection area.
Optionally, after calculating the hue average value and the saturation average value corresponding to the target detection area in the above manner, the uniformity value of the target detection area may be calculated according to the values of the hue channel and the saturation channel of each pixel point in the target detection area and the two average values.
Specifically, the uniformity value of the target detection area can be obtained by calculating in a standard deviation mode.
It should be noted that, a plurality of calculation modes may be configured in advance, and a best-effect mode may be selected from the plurality of calculation modes to calculate the uniformity value of the target detection area.
The following specifically explains a specific implementation process of determining a uniformity value of a target detection area in the method for evaluating uniformity of skin color of a human face provided in the embodiment of the present application.
Fig. 3 is another flow chart of an evaluation method for skin color uniformity of a face provided in an embodiment of the present application, referring to fig. 3, determining a uniformity value of a target detection area according to values of hue channels and saturation channels of each pixel point in the target detection area and a hue mean value and a saturation mean value corresponding to the target detection area, where the determining includes:
s310: and determining the standard deviation of the tone channel according to the numerical value and the tone average value of the tone channel of each pixel point in the target detection area.
Alternatively, the specific calculation formula is as follows:
wherein h_value is the standard deviation of the tone channel, n is the total number of pixel points, i is the ith pixel point, h i The h value of the ith pixel point.
S320: and determining the standard deviation of the saturation channel according to the numerical value and the saturation mean value of the saturation channel of each pixel point in the target detection area.
Alternatively, the specific calculation formula is as follows:
wherein s_value is the standard deviation of the saturation channel, s i The s value of the ith pixel point.
The steps of S310 and S320 may be performed sequentially or simultaneously, and are illustrated in fig. 3 by way of example, and are not limited thereto.
S330: and determining the uniformity value of the target detection area according to the standard deviation of the tone channel, the standard deviation of the saturation channel, the preset weight of the tone channel and the preset weight of the saturation channel.
Optionally, the uniformity value of the target detection area can be determined by obtaining the standard deviation of the tone channel and the standard deviation of the saturation channel, and the specific formula is as follows:
where α is a preset weight of the hue channel, and (1- α) is a preset weight of the saturation channel.
It should be noted that, the above formula is one possible calculation mode selected in advance through a calculation mode, and in the actual use process, other similar calculation formulas may be adopted, for example:
Other possible scenarios (1):
hs_value＝h_value+s_value；
namely
Other possible scenarios (2):
other possible scenarios (3):
wherein, the liquid crystal display device comprises a liquid crystal display device,
a i ＝s i *cos(h i /360*2*π)；
b i ＝h i *cos(s i /360*2*π)；
in order to judge the rationality of the above-mentioned various different calculation modes, a batch of face data can be photographed and collected under a controllable environment, and the face skin color uniformity score (for example, 0 score indicates the most non-uniformity and 100 score indicates the most uniformity) is marked by professional personnel such as doctors as gold standard, and is reasonably divided into a training set, a verification set and a test set (for example, 8:1:1) according to a proportion, and the relevance coefficient is determined, specifically as follows:
wherein X and Y may refer to the values of h and s described above. And judging the correlation between the hs_value calculated in the schemes and the gold standard, so as to screen an optimal scheme, wherein the scheme screened in the mode is the scheme adopted in the S330.
In the method for evaluating the uniformity of the skin color of the human face, which is adopted in the embodiment of the application, the uniformity value of the target detection area can be determined through the standard deviation of the tone channel, the standard deviation of the saturation channel, the preset weight of the tone channel and the preset weight of the saturation channel, and can be selected through the various feasible schemes, for example, the scheme with the highest correlation can be selected for use, so that the accuracy and the correlation of the uniformity value can be improved.
Another specific implementation process of determining the uniformity value of the target detection area in the method for evaluating the uniformity of the skin color of the human face provided in the embodiment of the present application is specifically explained below.
Fig. 4 is another flow chart of an evaluation method for skin color uniformity of a face provided in an embodiment of the present application, referring to fig. 4, determining a uniformity value of a target detection area according to values of hue channels and saturation channels of each pixel point in the target detection area and a hue mean saturation mean value corresponding to the target detection area, including:
s410: a plurality of detection sub-regions in the target detection region is determined based on the sliding window.
Alternatively, in the actual implementation process, the target detection area may be divided into a plurality of detection sub-areas by means of a sliding window, for example: if the sliding window is square, a plurality of square detection sub-areas can be divided, and each sub-area can be overlapped or not overlapped, and the corresponding division can be performed according to the actual requirement, and the method is not particularly limited.
S420: and determining the uniformity value of each detection subarea according to the values of the tone channel and the saturation channel of each pixel point in each detection subarea and the tone average saturation average value corresponding to each detection subarea.
Alternatively, the uniformity value may be determined for each detection sub-area after the area division, and the specific determining step may determine the uniformity value for each detection sub-area according to the specific calculation formulas of steps S310 to S330.
S430: and determining the uniformity value of the target detection area according to the uniformity value of each detection subarea.
Optionally, after obtaining the uniformity value in each detection area, a statistical manner may be adopted to obtain the uniformity value of the overall target detection area, for example: the method of partition characterization, or the method of calculating the mean value, etc. may be adopted, and are not particularly limited herein.
For example: if the data of the actual requirement pay attention to the local details, a partition characterization mode can be adopted to obtain the uniformity value of each detection subarea in the target detection area; correspondingly, if the data of the actual requirement pay attention to the overall distribution, the uniformity value of the whole target detection area can be determined by calculating the mean value.
In the method for evaluating the uniformity of the skin color of the face, which is adopted in the embodiment of the application, a plurality of detection subareas in a target detection area can be determined based on the sliding window, further, the uniformity value of each detection subarea is determined according to the values of the tone channel and the saturation channel of each pixel point in each detection subarea and the tone average saturation mean value corresponding to each detection subarea, and the uniformity value of the target detection area is determined according to the uniformity value of each detection subarea. The obtained uniformity value result can be more relevant and accurate through a sliding window mode.
Optionally, determining the overall uniformity result of the initial face image based on the uniformity value in the target detection area includes: and inputting the uniformity value in the target detection area into a target model to obtain an overall uniformity result of the initial face image, wherein the target model is a linear model obtained based on least square fitting.
It should be noted that the target model may be a preset linear model formula, for example:
y＝θ 1 +θ 2 x；
wherein θ 1 And theta 2 All are undetermined coefficients, a great amount of training can be carried out by collecting training samples as x, an equivalent amount of y is obtained by a manual labeling mode, a least square method is adopted to fit the model, and finally the following model is obtained:
score global ＝θ 1 +θ 2 *hs_value；
θ in the formula 1 And theta 2 Score, which is a fixed constant value obtained by the training global The overall uniformity result of the initial face image is obtained.
The following specifically explains a specific implementation process of the evaluation in the evaluation method of the uniformity of the complexion of the face provided in the embodiment of the present application.
Fig. 5 is another flow chart of an evaluation method for skin color uniformity of a face according to an embodiment of the present application, referring to fig. 5, the method further includes:
s510: and carrying out Gaussian blur processing on the initial face image to obtain a blurred image.
Alternatively, the specific formula is as follows:
I blur ＝G(I o )；
wherein I is o For initial face image, I blur To blur the image, G (I o ) Refers to a processing function that performs gaussian blur processing on an initial face image.
S520: and performing difference processing based on the initial face image and the blurred image to obtain a residual image.
Alternatively, the specific calculation formula is as follows:
I res ＝|I o -I blur |；
wherein I is res Is a residual image.
S530: and performing scaling correction processing on the residual image according to the overall uniformity result of the initial face image to obtain a target effect image.
Optionally, after the residual image is obtained, scaling correction processing can be performed on the residual image based on the overall uniformity result of the initial face image, scaling adjustment, pixel value conversion and the like of the image are performed, and finally a target effect image is obtained, wherein the target effect image can be an image for displaying the uniformity degree of the face complexion, and the uniform distribution condition of the face complexion can be determined more intuitively through the image.
S540: and outputting the target effect image as an evaluation result of the uniformity of the skin color of the human face.
Optionally, after the target effect image is obtained, the target effect image may be used as an evaluation result of the uniformity of the skin color of the face.
For comparison, the initial input image and the target effect image may be output together, and the specific output mode may be selected based on actual requirements, which is not limited herein.
In addition, in order to facilitate better display, a proper gray value can be allocated to the target effect image, and a color effect image is obtained through a color mapping method, so that an image of the distribution condition of the uniformity of the complexion of the face is output more directly and truly.
And when the target effect image is output, the overall uniformity result of the initial face image can be output, and the result and the image are combined, so that a user can more accurately obtain which region of the face skin color has the problem of lower uniformity, wherein the lower the uniformity is, the lower the score corresponding to the overall uniformity result is, the user can determine the uniform distribution degree of the face skin color in the initial input image through the score, and the condition of uneven skin color in which region is can be determined by combining the target result image.
According to the evaluation method for the uniformity of the skin color of the face, provided by the embodiment of the application, gaussian blur processing can be performed on an initial face image to obtain a blurred image; performing difference processing based on the initial face image and the blurred image to obtain a residual image; performing scaling correction processing on the residual image according to the overall uniformity result of the initial face image to obtain a target effect image; and outputting the target effect image as an evaluation result of the uniformity of the skin color of the human face. The distribution condition of the uniformity of the human face skin color in the initial input image can be more intuitively and clearly known by a user through the mode of outputting the target effect image.
Another specific implementation procedure of the evaluation in the method for evaluating the uniformity of the complexion of a human face provided in the embodiment of the present application is specifically explained below.
Fig. 6 is another flow chart of the evaluation method of the uniformity of the skin color of the face provided in the embodiment of the present application, please refer to fig. 6, in which scaling correction processing is performed on the residual image according to the overall uniformity result of the initial face image, to obtain a target effect image, which includes:
s610: and respectively obtaining a global uniformity score and a local uniformity score according to the overall uniformity result of the initial face image.
The global uniformity score is used for representing the overall uniformity of the initial face image, and the local uniformity score is used for representing the uniformity of each delimited area in the initial face image.
In order to highlight the area with uneven skin color in the target effect image, the area with even skin color is desalted, and a rectangular area can be obtained by dividing by taking each pixel point as the center, wherein the uniformity score of the rectangular area is the local uniformity score.
Alternatively, the overall uniformity result of the initial face image may represent a global uniformity score, i.e., score as described above global In addition, in a similar manner, each of the delineated regions may be scored to obtain a local uniformity score, as follows:
score local ＝θ 1 +θ 2 *hs_value local ；
Wherein score local Namely the local uniformity fraction, hs_value local The uniformity in the rectangular region is the result.
By the method, the global uniformity fraction and the local uniformity fraction can be determined respectively.
S620: and scaling and correcting the residual image based on the global uniformity fraction and the local uniformity fraction to obtain the target effect image.
Optionally, after the global uniformity score and the local uniformity score are obtained, the specific processing procedure is as follows:
wherein I is n For target effect image, I n (i, j) is the pixel value of the pixel point of the ith row and the jth column in the target effect image; i res (i, j) is the pixel value of the pixel point of the ith row and jth column in the residual image.
Through the method, the pixel value of each pixel point in the target effect image can be determined, and then the target effect image can be obtained.
The following specifically explains a specific implementation procedure for determining the target detection area provided in the embodiment of the present application.
Fig. 7 is another flow chart of an evaluation method for skin color uniformity of a face provided in an embodiment of the present application, please refer to fig. 7, which includes:
s710: a plurality of face keypoints are determined from the initial face image.
Optionally, the face part in the initial face image may be determined by face recognition technology, so as to determine a plurality of face key points on the face, where the face key points may cover the key positions such as the eyebrow, the nose bridge recess, the canthus, and the eyebrow.
S720: and determining a plurality of groups of target key points from the face key points, and establishing a Bezier curve based on each group of target key points.
Optionally, after the key points of the face are determined, the key points can be divided in a Bezier curve mode, and the specific process is as follows:
for each Bezier curve, four points P0, P1, P2, P3 are found, which define a cubic Fang Beici curve in planar space. The curve starts at P0, goes to P1, and goes from P2 to P3. Typically, P1 or P2 will not pass, these two points providing mainly directional information. The spacing between P0 and P1 determines the "specific length" of the curve in the direction of P2 before turning to P3, and the specific calculation formula is as follows:
B(t)＝P 0 (1-t) 3 +3P 1 t(1-t) 2 +3P 2 t 2 (1-t)+P 3 t 3 ，t∈[0，1]；
wherein B (t) is the parameter form of the curve, P 0 -P 3 I.e. the parameter values of the four points, t is a coefficient which can be preset, in particular between 0 and 1.
Illustratively, a Bezier curve from the eyebrow (A1) to above the left eye corner (A3) may be defined as:
P 3 ＝A 1 *0.2+A 3 *0.8；
P 2 ＝A 2 *0.2+A 4 *0.8；
P 1 ＝A 1 *0.5+A 4 *0.5；
P 0 ＝A 1 ；
Wherein, each point in A1-A4 can be a target key point of the human face, and the target key points can be at any position of the human face, for example: a1 can be positioned at the eyebrow, A3 can be positioned at the left eye corner, A2 can be positioned at the right side of a connecting line between A1 and A3, A4 can be positioned at the left side of the connecting line between A1 and A3, and corresponding coefficients can be determined based on specific distances, so that the formula is obtained, and the Bezier curve can be determined.
S730: the area surrounded by each Bezier curve is taken as an initial target detection area.
Alternatively, a plurality of bezier curves may be obtained in the above manner, and then an area surrounded by each bezier curve may be used as the initial target detection area.
S740: and removing the abnormal region in the initial target detection region to obtain the target detection region.
Alternatively, in the initial target detection area, there may be abnormal areas that occur due to the influence of factors such as pores in the skin of the human face being too large or stains, and these areas may be removed, thereby obtaining the target detection area.
In the method for evaluating the uniformity of the skin color of the face, a plurality of face key points can be determined from an initial face image, a plurality of groups of target key points are determined from the face key points, a Bezier curve is established based on each group of target key points, the area surrounded by the Bezier curves is used as an initial target detection area, and abnormal areas in the initial target detection area are removed to obtain a target detection area. The Bezier curve can be more accurately determined by a face key point mode, an initial target detection area can be more accurately divided, and the accuracy of a subsequent calculation result can be improved by removing an abnormal area.
The specific division of the target detection area provided in the embodiments of the present application is specifically explained below.
Fig. 8 is a schematic diagram of a specific position of a target detection area provided in an embodiment of the present application, please refer to fig. 8, wherein a portion on the left side of fig. 8 is a division situation of one of bezier curves, and a portion on the right side is a division situation of the whole face.
For the left-hand portion, please explain in connection with the embodiment of FIG. 7, each Bezier curve may include P 0 -P 3 Four points, wherein P 0 And P 3 Indicating start and end point, P 1 And P 2 Indicating the direction of curvature of the curve.
The right side portion is a face area surrounded by a plurality of bezier curves of the left side portion, and specifically may be areas on both sides of the face and near the nose, that is, areas surrounded by a broken line in fig. 8.
It should be noted that the area shown in fig. 8 is only an example, and the corresponding area change may be performed based on the actual requirement in the actual implementation process, which is not limited herein.
The following specifically explains a specific implementation procedure of determining an abnormal region in a target detection region provided in the embodiment of the present application.
Fig. 9 is another flow chart of the method for evaluating the uniformity of skin color of a face provided in the embodiment of the present application, please refer to fig. 9, and before removing an abnormal region in an initial target detection region to obtain a target detection region, the method further includes:
S910: the initial target detection area is converted into an image in tone saturation brightness format.
It should be noted that, step S910 may be performed synchronously when step S120 is performed, or may be performed separately, which is not limited herein, and any HSV format image conversion may be implemented.
Alternatively, the initial target detection area may be converted into an image of HSV.
S920: and taking the area of the average value of the brightness channels in the brightness format image of the tone saturation as an abnormal area.
Optionally, after obtaining the HSV format image, the abnormal region may be a region of the average value of the V channel, that is, the luminance channel, in the abnormal region, which specifically includes the following steps:
mean of V-channels in an initial target detection area, which may be obtained by counting a large number of face pictures (e.g., 2000 face pictures) v And standard deviation std v Can be larger than mean v +3*std v And less than mean v -3*std v The region where the pixel point position corresponding to the value of (a) is located is an abnormal region.
It should be noted that, the values of the H and S channels in the abnormal region have no reference meaning, and are usually greatly different from the original skin color due to light or shielding. Through the steps, the abnormal values of the skin color can be filtered, and the subsequent stable calculation of the uniformity result is facilitated.
In the method for evaluating the uniformity of the skin color of the face provided by the embodiment of the application, the initial target detection area can be converted into the image in the tone saturation brightness format, and then the area, in which the average value of the brightness channels in the image in the tone saturation brightness format is in the abnormal section, is used as the abnormal area. The pixel points with the influence of the abnormal value can be obtained more accurately through the brightness channel, and then the area where the pixel points with the abnormal value are located can be used as the abnormal area and removed, so that the accuracy of the target detection area can be improved.
The following describes a device, equipment, a storage medium, etc. corresponding to the method for evaluating the uniformity of the skin color of the face, and the specific implementation process and technical effects of the method are referred to above, and are not described in detail below.
Fig. 10 is a schematic structural diagram of an apparatus for evaluating skin color uniformity of a human face according to an embodiment of the present application, please refer to fig. 10, including: a determination module 110, a conversion module 120, a mean module 130, and a result module 140;
a determining module 110, configured to determine a target detection area based on the input initial face image;
The conversion module 120 is configured to obtain a tone saturation brightness image corresponding to the initial face image;
the average module 130 is configured to determine a uniformity value of the target detection area according to values of a hue channel and a saturation channel of each pixel point of the target detection area in the hue-saturation luminance image;
and a result module 140, configured to determine an overall uniformity result of the initial face image based on the uniformity value of the target detection area.
Optionally, the average module 130 is specifically configured to determine, according to the values of the hue channel and the saturation channel of each pixel point in the target detection area, a hue average value and a saturation average value corresponding to the target detection area; and determining the uniformity value of the target detection area according to the values of the tone channel and the saturation channel of each pixel point in the target detection area, and the tone average value and the saturation average value corresponding to the target detection area.
Optionally, the average module 130 is specifically configured to determine a standard deviation of the tone channel according to the value of the tone channel and the tone average of each pixel point in the target detection area; determining the standard deviation of the saturation channel according to the numerical value and the saturation mean value of the saturation channel of each pixel point in the target detection area; and determining the uniformity value of the target detection area according to the standard deviation of the tone channel, the standard deviation of the saturation channel, the preset weight of the tone channel and the preset weight of the saturation channel.
Optionally, the mean module 130 is specifically configured to determine a plurality of detection sub-areas in the target detection area based on the sliding window; determining the uniformity value of each detection subarea according to the values of the tone channel and the saturation channel of each pixel point in each detection subarea and the tone average saturation average value corresponding to each detection subarea; and determining the uniformity value of the target detection area according to the uniformity value of each detection subarea.
Optionally, the result module 140 is specifically configured to input the uniformity value in the target detection area into a target model, so as to obtain an overall uniformity result of the initial face image, where the target model is a linear model obtained based on least square fitting.
Optionally, the result module 140 is further configured to perform gaussian blur processing on the initial face image to obtain a blurred image; performing difference processing based on the initial face image and the blurred image to obtain a residual image; performing scaling correction processing on the residual image according to the overall uniformity result of the initial face image to obtain a target effect image; and outputting the target effect image as an evaluation result of the uniformity of the skin color of the human face.
Optionally, the result module 140 is specifically configured to obtain a global uniformity score and a local uniformity score according to a global uniformity result of the initial face image, where the global uniformity score is used to represent global uniformity of the initial face image, and the local uniformity score is used to represent uniformity of each delimited area in the initial face image; and scaling and correcting the residual image based on the global uniformity fraction and the local uniformity fraction to obtain the target effect image.
Optionally, the determining module 110 is specifically configured to determine a plurality of face key points from the initial face image; determining a plurality of groups of target key points from the face key points, and establishing a Bezier curve based on each group of target key points; taking the area surrounded by each Bezier curve as an initial target detection area; and removing the abnormal region in the initial target detection region to obtain the target detection region.
Optionally, the determining module 110 is further configured to convert the initial target detection area into an image in a tone saturation brightness format; and taking the area of the average value of the brightness channels in the brightness format image of the tone saturation as an abnormal area.
In the device for evaluating the uniformity of the skin color of the human face, the uniformity value of the target detection area can be determined by determining the target detection area based on the input initial human face image and acquiring the tone saturation brightness image corresponding to the initial human face image, and further according to the tone channel and the saturation channel value of each pixel point of the target detection area in the tone saturation brightness image, after the uniformity value of the target detection area is obtained, the overall uniformity result of the initial human face image can be determined based on the uniformity value of the target detection area, so that the evaluation of the uniformity of the skin color of the human face is realized. According to the values of the tone channel and the saturation channel in the tone saturation brightness image, the uniformity of the skin color of the human face can be obtained more accurately and intuitively, the overall uniformity result of the initial human face image is determined based on the uniformity value of the target detection area, the calculation error can be reduced, and the accuracy and the robustness of the detection result are improved.
The foregoing apparatus is used for executing the method provided in the foregoing embodiment, and its implementation principle and technical effects are similar, and are not described herein again.
The above modules may be one or more integrated circuits configured to implement the above methods, for example: one or more application specific integrated circuits (Application Specific Integrated Circuit, abbreviated as ASICs), or one or more microprocessors, or one or more field programmable gate arrays (Field Programmable Gate Array, abbreviated as FPGAs), etc. For another example, when a module above is implemented in the form of a processing element scheduler code, the processing element may be a general-purpose processor, such as a central processing unit (Central Processing Unit, CPU) or other processor that may invoke the program code. For another example, the modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).
Fig. 11 is a schematic structural diagram of a computer device provided in an embodiment of the present application, referring to fig. 11, the computer device includes: the device comprises a memory 210 and a processor 220, wherein a computer program capable of running on the processor 220 is stored in the memory 210, and the processor 220 realizes the steps of the evaluation method of the uniformity of the human face skin color when executing the computer program.
Optionally, the computer device is a mobile phone, a computer, a tablet computer or a professional electronic device.
In another aspect of the embodiments of the present application, there is further provided a computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the method for evaluating the uniformity of skin color of a human face.
In the several embodiments provided by the present invention, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of elements is merely a logical functional division, and there may be additional divisions of actual implementation, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in hardware plus software functional units.
The integrated units implemented in the form of software functional units described above may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium, and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (english: processor) to perform part of the steps of the methods of the embodiments of the invention. And the aforementioned storage medium includes: u disk, mobile hard disk, read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic disk or optical disk, etc.
The foregoing is merely a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the present application, and the changes or substitutions are covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the same, but rather, various modifications and variations may be made by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.
Claims (12)
1. The method for evaluating the uniformity of the skin color of the human face is characterized by comprising the following steps of:
determining a target detection area based on the input initial face image;
acquiring a tone saturation brightness image corresponding to the initial face image;
determining a uniformity value of the target detection area according to the values of the tone channel and the saturation channel of each pixel point of the target detection area in the tone saturation brightness image;
and determining the overall uniformity result of the initial face image based on the uniformity value of the target detection area.
2. The method for evaluating the uniformity of skin color of a human face according to claim 1, wherein determining the uniformity value of the target detection area according to the values of the hue channel and the saturation channel of each pixel point of the target detection area in the hue saturation brightness image comprises:
Determining a tone average value and a saturation average value corresponding to the target detection area according to the values of the tone channel and the saturation channel of each pixel point in the target detection area;
and determining the uniformity value of the target detection area according to the values of the tone channel and the saturation channel of each pixel point in the target detection area and the tone average value and the saturation average value corresponding to the target detection area.
3. The method for evaluating the uniformity of human face skin color according to claim 2, wherein determining the uniformity value of the target detection area according to the values of the hue channel and the saturation channel of each pixel point in the target detection area and the hue average value and the saturation average value corresponding to the target detection area comprises:
determining the standard deviation of the tone channel according to the value of the tone channel of each pixel point in the target detection area and the tone average value;
determining the standard deviation of the saturation channel according to the value of the saturation channel of each pixel point in the target detection area and the saturation mean value;
and determining the uniformity value of the target detection area according to the standard deviation of the tone channel, the standard deviation of the saturation channel, the preset weight of the tone channel and the preset weight of the saturation channel.
4. The method for evaluating the uniformity of skin color of a human face according to claim 2, wherein the determining the uniformity value of the target detection area according to the values of the hue channel and the saturation channel of each pixel point in the target detection area and the average value of the saturation of the hue mean corresponding to the target detection area comprises:
determining a plurality of detection sub-areas in the target detection area based on a sliding window;
determining the uniformity value of each detection subarea according to the values of the tone channel and the saturation channel of each pixel point in each detection subarea and the average value of tone average saturation corresponding to each detection subarea;
and determining the uniformity value of the target detection area according to the uniformity value of each detection subarea.
5. The method of claim 1, wherein the determining the overall uniformity result of the initial face image based on the uniformity value in the target detection area comprises:
and inputting the uniformity value in the target detection area into a target model to obtain an overall uniformity result of the initial face image, wherein the target model is a linear model obtained based on least square fitting.
6. The method for evaluating uniformity of skin color of a human face according to claim 1, further comprising:
carrying out Gaussian blur processing on the initial face image to obtain a blurred image;
performing difference processing based on the initial face image and the blurred image to obtain a residual image;
performing scaling correction processing on the residual image according to the overall uniformity result of the initial face image to obtain a target effect image;
and outputting the target effect image as an evaluation result of the uniformity of the skin color of the human face.
7. The method for evaluating uniformity of skin color of a human face according to claim 6, wherein scaling and correcting the residual image according to the overall uniformity result of the initial human face image to obtain the target effect image comprises:
respectively obtaining a global uniformity score and a local uniformity score according to the overall uniformity result of the initial face image, wherein the global uniformity score is used for representing the overall uniformity of the initial face image, and the local uniformity score is used for representing the uniformity of each delimited area in the initial face image;
and scaling and correcting the residual image based on the global uniformity fraction and the local uniformity fraction to obtain the target effect image.
8. The method for evaluating uniformity of skin color of a human face according to claim 1, wherein said determining a target detection area based on an input initial human face image comprises:
determining a plurality of face key points from the initial face image;
determining a plurality of groups of target key points from the face key points, and establishing a Bezier curve based on each group of target key points;
taking the area surrounded by the Bezier curves as an initial target detection area;
and removing the abnormal region in the initial target detection region to obtain the target detection region.
9. The method for evaluating uniformity of skin color of a human face according to claim 8, wherein before removing abnormal regions in the initial target detection region to obtain the target detection region, the method further comprises:
converting the initial target detection area into an image in a tone saturation brightness format;
and taking the area of the average value of the brightness channels in the brightness format image with the tone saturation as an abnormal area.
10. An apparatus for evaluating uniformity of skin color of a human face, comprising: the device comprises a determining module, a converting module, a mean module and a result module;
The determining module is used for determining a target detection area based on the input initial face image;
the conversion module is used for acquiring a tone saturation brightness image corresponding to the initial face image;
the average module is used for determining the uniformity value of the target detection area according to the values of the tone channel and the saturation channel of each pixel point of the target detection area in the tone saturation brightness image;
and the result module is used for determining the overall uniformity result of the initial face image based on the uniformity value of the target detection area.
11. A computer device, comprising: memory, a processor, in which a computer program is stored which is executable on the processor, when executing the computer program, realizing the steps of the method of any one of claims 1 to 9.
12. A computer-readable storage medium, characterized in that the storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the method according to any of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310158756.8A CN116245759A (en) | 2023-02-23 | 2023-02-23 | Method, device, equipment and storage medium for evaluating uniformity of human face complexion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310158756.8A CN116245759A (en) | 2023-02-23 | 2023-02-23 | Method, device, equipment and storage medium for evaluating uniformity of human face complexion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116245759A true CN116245759A (en) | 2023-06-09 |
Family
ID=86630895
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310158756.8A Pending CN116245759A (en) | 2023-02-23 | 2023-02-23 | Method, device, equipment and storage medium for evaluating uniformity of human face complexion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116245759A (en) |
-
2023
- 2023-02-23 CN CN202310158756.8A patent/CN116245759A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101140533B1 (en) | Method and system for recommending a product based upon skin color estimated from an image | |
| US7720279B2 (en) | Specifying flesh area on image | |
| CN111814520A (en) | Skin type detection method, skin type grade classification method, and skin type detection device | |
| CN112767392B (en) | Image definition determining method, device, equipment and storage medium | |
| JP2004165932A (en) | Device and method for estimating light source, image pickup device, and image processing method | |
| CN110796612B (en) | Image enhancement method and system | |
| CN112614085A (en) | Object detection method and device and terminal equipment | |
| CN108510477B (en) | Method and device for positioning color block of test paper | |
| US10620417B2 (en) | Method for generating a reflection-reduced contrast image and corresponding device | |
| CN107368806A (en) | Image correction method, device, computer-readable recording medium and computer equipment | |
| JP5880184B2 (en) | Color measurement program and color measurement method | |
| CN113642358B (en) | Skin color detection method, device, terminal and storage medium | |
| JP7234086B2 (en) | Biometric information acquisition device and program | |
| CN110458232B (en) | Method and equipment for determining image style similarity | |
| CN115278186B (en) | Controllable uniform projection method, device, equipment and medium based on Internet of things | |
| CN116245759A (en) | Method, device, equipment and storage medium for evaluating uniformity of human face complexion | |
| Hu et al. | Jpeg ringing artifact visibility evaluation | |
| CN112700396A (en) | Illumination evaluation method and device for face picture, computing equipment and storage medium | |
| CN112801997B (en) | Image enhancement quality evaluation method, device, electronic equipment and storage medium | |
| CN115457055A (en) | Illuminance meter value identification method, electronic device, and storage medium | |
| CN114881889A (en) | Video image noise evaluation method and device | |
| JP2007219899A (en) | Personal identification device, personal identification method, and personal identification program | |
| CN113221893A (en) | Instrument image processing method and device, computer equipment and storage medium | |
| CN113706400A (en) | Image correction method, image correction device, microscope image correction method, and electronic apparatus | |
| CN112132000A (en) | Living body detection method and device, computer readable medium and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |