WO2020207423A1 - 肤质检测方法、肤质等级分类方法及肤质检测装置 - Google Patents

肤质检测方法、肤质等级分类方法及肤质检测装置 Download PDF

Info

Publication number
WO2020207423A1
WO2020207423A1 PCT/CN2020/083894 CN2020083894W WO2020207423A1 WO 2020207423 A1 WO2020207423 A1 WO 2020207423A1 CN 2020083894 W CN2020083894 W CN 2020083894W WO 2020207423 A1 WO2020207423 A1 WO 2020207423A1
Authority
WO
WIPO (PCT)
Prior art keywords
facial
skin
image
face
face image
Prior art date
Application number
PCT/CN2020/083894
Other languages
English (en)
French (fr)
Inventor
蔡细敏
许合欢
王进
Original Assignee
虹软科技股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 虹软科技股份有限公司 filed Critical 虹软科技股份有限公司
Priority to KR1020217037077A priority Critical patent/KR20210149848A/ko
Priority to US17/603,012 priority patent/US20220237811A1/en
Priority to JP2021559127A priority patent/JP7413400B2/ja
Publication of WO2020207423A1 publication Critical patent/WO2020207423A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168—Feature extraction; Face representation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/40—Analysis of texture
    • G06T5/92—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/161—Detection; Localisation; Normalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/161—Detection; Localisation; Normalisation
    • G06V40/162—Detection; Localisation; Normalisation using pixel segmentation or colour matching
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168—Feature extraction; Face representation
    • G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
    • 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

Definitions

  • This application relates to computer vision processing technology, and specifically to a skin quality detection method, a skin quality classification method and a skin quality detection device.
  • bioelectrical impedance analysis evaluates skin quality by measuring skin electrical impedance characteristics.
  • This method has fewer skin characteristics and can only be measured. Skin moisture and fat are secreted, and because the testing equipment needs to be in direct contact with the skin, contact impedance has a greater impact on the testing results, and the testing accuracy is poor.
  • this method requires additional configuration of professional hardware devices, and most of these devices require professionals to operate, and users need to go to a specific testing place to complete, application groups and application scenarios are subject to certain restrictions, and the cost of these hardware devices is relatively high. , It is not convenient for actual promotion and application.
  • the embodiments of the present application provide a skin quality detection method, a skin quality classification method and a skin quality detection device, so as to at least solve the problems in the prior art that the detected skin quality features are few, the detection accuracy is poor, the hardware cost is high, and the volume is large. .
  • a skin quality detection method which includes: acquiring a face image; acquiring a face skin color area and a face feature point in the face image; The facial skin color area and the facial feature points acquire the facial skin quality features of the facial image.
  • acquiring the human face skin color area and the human face feature points in the human face image includes: acquiring the human face skin color area using a skin color detection algorithm.
  • the method further includes: cutting out the facial features of the facial image according to the facial feature points to obtain the facial skin color region.
  • the method further includes using a morphological algorithm to process the skin color area of the human face to enlarge the removed facial features area.
  • the facial skin quality feature includes at least one of the following: skin tone, spots, pores, wrinkles, dark circles, smoothness.
  • acquiring the facial skin quality feature of the facial image according to the facial skin tone region and the facial feature points includes: obtaining a detail map of the facial skin tone region by using a high-contrast algorithm.
  • the method further includes: acquiring face direction information according to the face feature points.
  • acquiring the facial skin characteristics of the facial image according to the facial skin color area and the facial feature points further includes: The first preset algorithm obtains the detection result of the spots and/or pores in the detailed image; distinguishes the spots and/or pores according to the shape feature.
  • the method includes obtaining the skin smoothness through a gray level co-occurrence matrix algorithm according to the detection result of the spots and/or pores.
  • acquiring the facial skin quality features of the facial image according to the facial skin tone area and the facial feature points further includes: using a second preset The algorithm obtains the detection result of the wrinkle in the detailed image; determines the type of the wrinkle according to the face direction information and the facial feature points.
  • acquiring the facial skin quality features of the facial image according to the facial skin color area and the facial feature points includes: according to the facial features For the eye feature points in the feature points, draw two upper and lower Bezier curves in the face image, locate the position of the dark circle, and determine the average brightness of the position of the dark circle and the surrounding area The difference of the detection result of the dark circles in the face image is obtained.
  • the skin quality detection method further includes: performing enhancement processing on the face image to obtain an enhanced face image.
  • performing enhancement processing on the face image to obtain an enhanced face image includes: acquiring brightness information of the face image; enhancing the contrast of a low gray value area in the face image to obtain the enhancement Face image.
  • acquiring the brightness information of the face image includes: converting the face image into a YUV format image through a color conversion algorithm, and extracting the Y channel image in the YUV format image to obtain the brightness of the face image information.
  • a skin quality classification method comprising: acquiring the facial skin quality feature by using any one of the above skin quality detection methods; Quality features, using machine learning methods to classify facial skin quality into different levels.
  • the method includes: setting corresponding beauty parameters according to different levels of the facial skin quality.
  • the method is applied to an electronic device with a video call or camera function.
  • a skin quality detection device including: an image acquisition module configured to acquire a face image; and an acquisition module configured to acquire facial skin color in the face image Regions and facial feature points; a skin quality detection module configured to detect the facial skin quality features of the face image according to the facial skin color region and the facial feature points.
  • the acquiring module includes a skin color acquiring module configured to acquire the face skin color area in the face image by using a skin color detection algorithm.
  • the skin color acquisition module is further configured to cut out the facial features of the face image according to the facial feature points to obtain the skin color area of the face.
  • the skin color acquiring module is further configured to process the skin color area of the human face by using a morphological algorithm to enlarge the removed facial features area.
  • the facial skin quality features include at least one of the following: skin tone, spots, pores, wrinkles, dark circles, smoothness.
  • the skin quality detection module is further configured to use a high-contrast algorithm to obtain a detailed map of the skin color area of the face.
  • the skin quality detection module is further configured to obtain face direction information according to the face feature points.
  • the skin quality detection module includes a spot pore detection module, and the spot pore detection module is configured to obtain a detection result of the spots and/or pores in the detailed image by using a first preset algorithm; Distinguish the spots and/or pores.
  • the skin quality detection module includes a skin smoothness detection module configured to obtain the skin smoothness through a gray-level co-occurrence matrix algorithm according to the detection result of the spots and/or pores.
  • the skin quality detection module includes a wrinkle detection module configured to use a second preset algorithm to obtain the detection result of the wrinkle in the detailed image; determine the wrinkle detection result according to the face direction information and the face feature points. State the type of wrinkles.
  • the skin quality detection module includes a dark circle detection module configured to draw two upper and lower Bezier curves in the face image based on the eye feature points in the face feature points, and locate Go to the position of the dark circle, and obtain the detection result of the dark circle in the face image by judging the difference between the brightness average value of the position of the dark circle and the surrounding area.
  • a dark circle detection module configured to draw two upper and lower Bezier curves in the face image based on the eye feature points in the face feature points, and locate Go to the position of the dark circle, and obtain the detection result of the dark circle in the face image by judging the difference between the brightness average value of the position of the dark circle and the surrounding area.
  • the device further includes an image enhancement module configured to perform enhancement processing on the face image to obtain an enhanced face image.
  • an image enhancement module configured to perform enhancement processing on the face image to obtain an enhanced face image.
  • the image enhancement module includes: a brightness information obtaining unit configured to obtain brightness information of the face image; a contrast enhancement unit configured to enhance the contrast of a low gray value region in the face image to obtain The enhanced face image.
  • the brightness information acquiring unit is further configured to convert the face image into a YUV format image through a color conversion algorithm, and extract the Y channel image in the YUV format image to obtain brightness information of the face image.
  • an electronic device including: a processor; and a memory configured to store executable instructions of the processor; wherein the processor is configured to execute The executable instructions are used to execute any one of the aforementioned skin quality detection methods.
  • a storage medium includes a stored program, wherein, when the program is running, the device where the storage medium is located is controlled to execute any one of the aforementioned skin types. Detection method.
  • the facial image is obtained; the skin skin area and the feature points of the face in the face image are obtained; the skin quality of the face image is obtained according to the skin color area and the feature points of the face Features: It can detect facial skin characteristics based on face images, and can detect more skin characteristics as needed without increasing hardware cost and volume, and improve detection accuracy. Furthermore, the problems in the prior art that the detected skin quality features are few, the detection accuracy is poor, the hardware cost is high, and the volume is large.
  • Fig. 1 is a flowchart of an optional skin quality detection method according to one of the embodiments of the present application
  • FIG. 2 is a flowchart of an optional skin quality classification method according to one of the embodiments of the present application.
  • FIG. 3 is a structural block diagram of an optional skin quality detection device according to one of the embodiments of the present application.
  • Fig. 4 is a structural block diagram of an optional electronic device according to one of the embodiments of the present application.
  • the embodiments of the present application can be applied to a computer system/server, which can operate with many other general-purpose or special-purpose computing system environments or configurations.
  • Examples of well-known computing systems, environments and/or configurations suitable for use with computer systems/servers include, but are not limited to: personal computer systems, handheld or laptop devices, microprocessor-based systems, programmable consumer electronics, Small computer systems, large computer systems, and distributed cloud computing technology environments including any of the above systems, etc.
  • the computer system/server may be described in the general context of computer system executable instructions (such as program modules, etc.) executed by the computer system.
  • program modules can include routines, programs, target programs, components, logic, and data structures, etc., which perform specific tasks or implement specific abstract data types.
  • the computer system/server can be implemented in a distributed cloud computing environment, and tasks are performed by remote processing equipment linked through a communication network.
  • program modules may be located on a storage medium of a local or remote computing system including a storage device.
  • a skin quality detection method is provided.
  • FIG. 1 it is a flowchart of an optional skin quality detection method according to an embodiment of the present application. As shown in Figure 1, the method includes the following steps:
  • S14 Acquire the skin quality feature of the face image according to the skin color area and the feature points of the face.
  • the face image is obtained; the face skin color area and the face feature points in the face image are obtained; the face of the face image is obtained according to the face skin color area and the face feature points
  • Skin characteristics It can detect facial skin characteristics based on face images, and can detect more skin characteristics as needed without increasing hardware cost and volume, and improve detection accuracy.
  • Step S10 obtaining a face image
  • the face image may be acquired through an image acquisition module, where the image acquisition module may be an independent camera device or a camera device integrated on an electronic device, for example, an independent RGB camera , Or the camera that comes with electronic devices such as in-vehicle electronic equipment (including but not limited to central control screens, navigators, etc.), mobile phones, tablet computers, desktop computers, and skin quality testers.
  • image acquisition module may be an independent camera device or a camera device integrated on an electronic device, for example, an independent RGB camera , Or the camera that comes with electronic devices such as in-vehicle electronic equipment (including but not limited to central control screens, navigators, etc.), mobile phones, tablet computers, desktop computers, and skin quality testers.
  • Step S12 Obtain the skin color area and the face feature points in the face image
  • the skin color area of the face in the face image can be obtained through a skin color detection algorithm.
  • the face image can be converted to the YCbCr color space through the skin color detection algorithm, and then the CrCb value of each pixel is substituted into the ellipse statistical model obtained from the skin pixels. If the CrCb coordinates are in the ellipse statistical model, it is judged as skin color , Thereby obtaining the skin tone area of the face.
  • the facial feature points can be obtained through the SDM algorithm.
  • the facial features in the face image may be cut out according to the facial feature points to obtain the face skin color area.
  • Step S14 Acquire the skin quality feature of the face image according to the skin color area and the feature points of the face.
  • the facial skin characteristics include at least one of the following: skin color, spots, pores, wrinkles, dark circles, and smoothness.
  • acquiring the facial skin characteristics of the facial image according to the facial skin color area and the facial feature points includes:
  • the high-contrast algorithm obtains the detailed image of the skin color area of the face;
  • the first preset algorithm for example, the local adaptive threshold algorithm
  • the detection result includes at least one of the following: Location, quantity, area; spots and/or pores are distinguished according to the shape characteristics, where the shape characteristics of the pores are: the area is generally small and similar to a circle; and the shape characteristics of the spots are: the area is larger.
  • a morphological algorithm for example, a corrosion expansion algorithm
  • the connected domain algorithm is used to eliminate the error points with abnormal shapes or large areas in the initial detection results.
  • the above-mentioned steps can be implemented in the skin color area of the face where the facial features are removed, so as to reduce the false detection rate of spots and/or pores.
  • the skin smoothness may be obtained through the gray-level co-occurrence matrix algorithm according to the detection result of the spots and/or pores.
  • the gray-level co-occurrence matrix algorithm is used to calculate the energy, entropy, contrast, and inverse difference moments of 0 degrees, 45 degrees, 90 degrees and 135 degrees, and then use these 16 parameters to obtain skin smoothness features.
  • acquiring the facial skin quality features of the face image according to the facial skin color area and the facial feature points includes: using a high-contrast algorithm to obtain The detail map of the skin color area of the face; the face direction information is obtained according to the face feature points; the second preset algorithm is adopted to obtain the wrinkle detection result in the detail map; the type of wrinkle is determined according to the face direction information and the face feature points.
  • the detection result includes at least one of the following: location, quantity, area.
  • the morphological algorithm and the connected domain algorithm are used to remove some objects that are obviously not wrinkle features.
  • the adaptive threshold algorithm obtains the wrinkle position, and determines the horizontal wrinkles as the eye wrinkles according to the facial feature points and the face direction information; uses the Canny edge extraction algorithm to obtain the wrinkle position, and determines it according to the face feature points and the face direction information
  • Horizontal wrinkles are forehead wrinkles, and diagonal wrinkles are nasolabial wrinkles.
  • acquiring the facial skin characteristics of the facial image according to the facial skin color area and the facial feature points includes: For the eye feature points in the face feature points, draw the upper and lower Bezier curves in the face image to locate the dark circle detection area; calculate the difference between the average brightness of the dark circle detection area and the surrounding area to determine the severity of the dark circle .
  • step S10 after obtaining the face image in step S10, it may further include step S11: performing enhancement processing on the face image to obtain an enhanced face image.
  • performing enhancement processing on the face image to obtain the face image includes: obtaining brightness information of the face image through a third predetermined algorithm; and enhancing the low gray value area in the face image Contrast, get enhanced face image.
  • the details of the face image in dark light can be enhanced.
  • the facial skin characteristics can be acquired on the enhanced facial image to obtain more accurate detection results.
  • the method of expanding the low gray value region in the face image includes but not limited to logarithmic transformation, histogram equalization, exponential transformation, and so on.
  • acquiring the brightness information of the face image through the third predetermined algorithm may be converting the face image into a YUV format image through a color conversion algorithm, and extracting the Y channel image in the YUV format image to obtain the brightness information of the face image.
  • the face frame detection may also be included, and the image is scaled according to the size of the detected face frame to obtain the size matching The requested face image.
  • the detected skin quality features can include spots, pores, skin smoothness, wrinkles, dark circles, etc., and the detection accuracy can be improved without increasing hardware cost and volume.
  • the skin quality can also be evaluated in combination with the skin color area of the face.
  • the spot area and the skin color area of the face can be used.
  • the area ratio is used as an indicator to evaluate the severity of spots.
  • a skin quality classification method is also provided.
  • FIG. 2 it is a flowchart of an optional skin quality classification method according to one embodiment of the present application. As shown in Figure 2, the method includes the following steps:
  • the machine learning method may be a support vector machine or a perceptron.
  • the machine learning method can obtain a preferred classification function through training samples, where the facial skin characteristics of each sample include at least one of the following: skin color, spots, pores, wrinkles, dark circles, and smoothness.
  • the skin quality detection method described above is used to obtain the facial skin quality features; according to the facial skin quality features, the machine learning method is used to classify the facial skin quality into different levels; The facial skin quality features are detected, and the facial skin quality can also be divided into different levels, which is convenient for providing care suggestions for different users, recommending suitable skin care products or achieving intelligent beauty.
  • corresponding beauty parameters can be set according to different levels of human facial skin quality to realize smart beauty. For example, if a person’s skin has more pores, wrinkles, and spots, and the skin is rough, the skin quality of the face is defined as poor, and the corresponding beauty parameters can be set accordingly. For example, set the beauty parameters in The microdermabrasion strength setting is stronger. On the contrary, if a person’s skin has fewer pores, wrinkles, and spots, and the skin is smooth, and the skin quality of the face is defined as good, the corresponding beauty parameters can be set accordingly.
  • the microdermabrasion intensity is set a little lower; so as to ensure that everyone's beauty parameters are optimal, to achieve natural beauty effects and achieve intelligent beauty.
  • the smart beauty technology realized by setting the corresponding beauty parameters according to the different levels of facial skin quality can be carried on electronic devices with functions such as video calls or taking pictures, such as in-vehicle electronic devices (including but not limited to central control screens, Navigator, etc.), mobile phones, digital cameras, tablet computers, desktop computers, skin quality testers, etc.
  • in-vehicle electronic devices including but not limited to central control screens, Navigator, etc.
  • mobile phones digital cameras
  • tablet computers desktop computers
  • skin quality testers etc.
  • the smart beautification technology can be used to enhance the user experience through the above-mentioned in-vehicle electronic devices with video call or taking pictures.
  • a skin quality detection device is also provided.
  • FIG. 3 it is a structural block diagram of an optional skin quality detection device according to one of the embodiments of the present application. As shown in Figure 3, the skin quality detection device 3 includes:
  • the image acquisition module 30 is configured to acquire a face image
  • the image acquisition module may be an independent camera device or a camera device integrated on an electronic device, etc., such as an independent RGB camera, or a mobile phone, tablet computer, desktop computer, or skin type. Cameras that come with electronic devices such as detectors.
  • the obtaining module 32 is configured to obtain facial skin color areas and facial feature points in the face image
  • the obtaining module 34 includes a skin color obtaining module and a feature point obtaining module.
  • the skin color acquiring module is configured to acquire the skin color area of the face in the face image through the skin color detection algorithm.
  • the face image can be converted to the YCbCr color space through the skin color detection algorithm, and then the CrCb value of each pixel is substituted into the ellipse statistical model obtained from the skin pixels. If the CrCb coordinates are in the ellipse statistical model, it is judged as skin color , Thereby obtaining the skin tone area of the face.
  • the facial feature points can be obtained through the SDM algorithm.
  • the skin color acquisition module in order to obtain a more accurate face skin color area, is configured to cut out the facial features in the face image according to the facial feature points to obtain the face skin color area.
  • it is also configured to use a morphological algorithm (for example, the corrosion expansion algorithm) to process the skin color area of the face to enlarge the removed facial features, so as to avoid the residual facial features that are not completely removed due to the image shift problem.
  • a morphological algorithm for example, the corrosion expansion algorithm
  • the skin quality detection module 34 is configured to detect the skin quality features of the face image according to the skin color area and the face feature points.
  • the facial skin characteristics include at least one of the following: skin color, spots, pores, wrinkles, dark circles, and smoothness.
  • the skin quality detection module 34 includes a spot pore detection module 340 configured to use a high-contrast algorithm to obtain a detailed map of the skin color area of the face;
  • a preset algorithm (for example, a local adaptive threshold algorithm) obtains the detection result of spots and/or pores in the detail map, where the detection result includes at least one of the following: position, number, area; distinguish spots and/or pores according to shape characteristics, Among them, the shape of pores is characterized by: the area is generally small and similar to a circle; the shape of spots is characterized by: the area is larger.
  • a morphological algorithm for example, the corrosion expansion algorithm
  • the connected domain algorithm eliminates the error points with abnormal shapes or large areas in the initial detection results.
  • the above-mentioned steps can be implemented in the skin color area of the face where the facial features are removed, so as to reduce the false detection rate of spots and/or pores.
  • the skin quality detection module 34 includes a skin smoothness detection module 342, configured to obtain the detection result of the spots and/or pores, according to the detection result of the spots and/or pores,
  • the skin smoothness is obtained through the gray-level co-occurrence matrix algorithm.
  • the gray-level co-occurrence matrix algorithm is used to calculate the energy, entropy, contrast, and inverse difference moments of 0 degrees, 45 degrees, 90 degrees and 135 degrees, and then use these 16 parameters to obtain skin smoothness features.
  • the skin quality detection module 34 includes a wrinkle detection module 344.
  • the wrinkle detection module 344 is configured to use a high-contrast algorithm to obtain a detailed image of a face image; obtain face orientation information according to facial feature points; use a second preset algorithm to obtain a wrinkle detection result in the detail image, where the detection result includes the following At least one: location, number, area; distinguish the types of wrinkles according to the face direction information and the face feature points.
  • the morphological algorithm and the connected domain algorithm are used to remove some objects that are obviously not wrinkle features.
  • the adaptive threshold algorithm obtains the wrinkle position, and marks horizontal wrinkles as eye wrinkles according to the face direction information and facial feature points;
  • the Canny edge extraction algorithm is used to obtain the wrinkle position, and according to the face direction information and face feature points Mark horizontal wrinkles as forehead wrinkles, and diagonal wrinkles as nasolabial wrinkles.
  • the skin quality detection module 34 includes a dark circle detection module 346.
  • the dark circle detection module 346 is configured to draw the upper and lower Bezier curves in the face image according to the eye feature points in the face feature points to locate the dark circle detection area; calculate the average brightness of the dark circle detection area and the surrounding area The difference determines the severity of dark circles.
  • the skin quality detection device 3 further includes an image enhancement module 31 configured to process the face image to obtain the face image.
  • the image enhancement module 31 includes a brightness information acquisition unit 310 and a contrast enhancement unit 312.
  • the brightness information obtaining unit 310 is configured to obtain the brightness information of the face image through a third predetermined algorithm
  • the contrast enhancement unit 312 is configured to increase the contrast of the low gray value region in the face image to obtain an enhanced face image.
  • the details of the face image in dark light can be enhanced.
  • the facial skin characteristics can be acquired on the enhanced facial image to obtain more accurate detection results.
  • the method of expanding the low gray value region in the face image includes but not limited to logarithmic transformation, histogram equalization, exponential transformation, and so on.
  • acquiring the brightness information of the face image through the third predetermined algorithm may be converting the face image into a YUV format image through a color conversion algorithm, and extracting the Y channel image in the YUV format image to obtain the brightness information of the face image.
  • the image may also be scaled to obtain a face image with a size that meets the requirements.
  • the aforementioned modules namely, the image acquisition module 30, are configured to acquire facial images; the acquisition module 34 is configured to acquire facial skin color areas and facial feature points in the facial images; skin quality
  • the detection module 36 is configured to detect the facial skin quality features of the facial image according to the facial skin color area and the facial feature points; it can detect a variety of facial skin quality features based on the facial image, and can be used without increasing hardware cost and volume In the case of, detect more skin characteristics as needed to improve the detection accuracy.
  • an electronic device is further provided.
  • the electronic device 40 includes a processor 400; and a memory 402 configured to store executable instructions of the processor 400; wherein The processor 400 is configured to execute any one of the skin quality detection methods described above by executing the executable instructions.
  • a storage medium wherein the storage medium includes a stored program, wherein the device where the storage medium is located is controlled to execute any one of the foregoing when the program is running.
  • the skin quality detection method described in item is also provided.
  • the disclosed technical content can be implemented in other ways.
  • the device embodiments described above are merely illustrative.
  • the division of the units may be a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components may be combined or may be Integrate into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, units or modules, and may be in electrical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • each unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of this application essentially or the part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , Including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present application.
  • the aforementioned storage media include: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program code .
  • Face skin quality features By acquiring the face image; acquiring the skin color area and face feature points in the face image; acquiring the skin texture features of the face image according to the face skin color area and the face feature points; detecting people based on the face image Face skin quality features, and can detect more skin quality features as needed without increasing hardware cost and volume, and improve detection accuracy. Furthermore, the problems in the prior art that the detected skin quality features are few, the detection accuracy is poor, the hardware cost is high, and the volume is large.

Abstract

本申请公开了一种肤质检测方法、肤质等级分类方法及肤质检测装置。一种肤质检测方法通过获取人脸图像；获取人脸图像中的人脸肤色区域和人脸特征点；根据人脸肤色区域和人脸特征点获取人脸图像的人脸肤质特征；由此可以基于人脸图像检测人脸肤质特征，并且可以在不增加硬件成本和体积的情况下，根据需要检测更多的肤质特征，提高检测精度。进而解决现有技术中检测的肤质特征少，检测精度差且硬件成本高、体积大的问题。

Description

肤质检测方法、肤质等级分类方法及肤质检测装置
本申请要求于2019年04月12日提交中国专利局、优先权号为201910292616.3、发明名称为“肤质检测方法、肤质等级分类方法及肤质检测装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机视觉处理技术，具体而言，涉及一种肤质检测方法、肤质等级分类方法及肤质检测装置。
背景技术
随着社会发展和生活水平的提高，人们对个人形象越来越重视，尤其是对肤质的关注度日益增强，一般而言，通过普通的人为观察可以辨别一些常见的问题性皮肤，但是想要对肤质进行更准确地检测，则需要借助科学手段实现。当前应用于皮肤肤质检测的主要技术手段有生物电阻抗分析法，生物电阻抗分析法通过测量皮肤电阻抗特性来评价皮肤肤质，但这种方法检测的肤质特征较少，仅能测量皮肤水分和脂肪分泌，并且因为需要检测仪器直接接触皮肤，接触阻抗对检测结果影响较大，检测精度差。另外，这种方式需要额外配置专业的硬件设备，而这些设备大多需要专业人士才能操作，并需要用户到特定检测场所完成，应用群体和应用场景都受到一定限制，同时这些硬件设备的成本较高，不便于实际推广和应用。
发明内容
本申请实施例提供了一种肤质检测方法、肤质等级分类方法及肤质检测装置，以至少解决现有技术中检测的肤质特征少，检测精度差且硬件成本高、体积大的问题。
根据本申请其中一实施例的一个方面，提供了一种肤质检测方法，该方法包括：获取人脸图像；获取所述人脸图像中的人脸肤色区域和人脸特征点；根据所述人脸肤色区域和所述人脸特征点获取所述人脸图像的人脸肤质特征。
可选地，获取所述人脸图像中的人脸肤色区域和人脸特征点包括：采用肤色检测算法获取所述人脸肤色区域。
可选地，所述方法还包括：根据所述人脸特征点抠除所述人脸图像的五官区域，得到所述人脸肤色区域。
可选地，所述方法还包括利用形态学算法处理所述人脸肤色区域，以扩大被抠除的所述五官区域。
可选地，所述人脸肤质特征包括以下至少之一：肤色、斑点、毛孔、皱纹、黑眼圈、光滑度。
可选地，根据所述人脸肤色区域和所述人脸特征点获取所述人脸图像的人脸肤质特征包括：采用高反差算法得到所述人脸肤色区域的细节图。
可选地，所述方法还包括：根据所述人脸特征点获取人脸方向信息。
可选地，当所述人脸肤质特征包括斑点和/或毛孔时，根据所述人脸 肤色区域和所述人脸特征点获取所述人脸图像的人脸肤质特征还包括：采用第一预设算法获得所述细节图中所述斑点和/或毛孔的检测结果；根据形状特征区分所述斑点和/或毛孔。
可选地，当所述人脸肤质特征包括皮肤光滑度时，所述方法包括根据所述斑点和/或毛孔的检测结果，通过灰度共生矩阵算法获取所述皮肤光滑度。
可选地，当所述人脸肤质特征包括皱纹时，根据所述人脸肤色区域和所述人脸特征点获取所述人脸图像的人脸肤质特征还包括：采用第二预设算法获取所述细节图中所述皱纹的检测结果；根据人脸方向信息和所述人脸特征点确定所述皱纹的种类。
可选地，当所述人脸肤质特征包括黑眼圈时，根据所述人脸肤色区域和所述人脸特征点获取所述人脸图像的人脸肤质特征包括：根据所述人脸特征点中的眼部特征点，在所述人脸图像中绘制出上下两条贝塞尔曲线，定位到所述黑眼圈的位置，并通过判别所述黑眼圈的位置的亮度均值与周围区域的差异性，获取所述人脸图像中所述黑眼圈的检测结果。
可选地，所述肤质检测方法还包括：对所述人脸图像进行增强处理，得到增强人脸图像。
可选地，对所述人脸图像进行增强处理，得到增强人脸图像包括：获取所述人脸图像的亮度信息；增强所述人脸图像中低灰度值区域的对比度，得到所述增强人脸图像。
可选地，获取所述人脸图像的亮度信息包括：通过颜色转换算法将所述人脸图像转换为YUV格式图像，并提取所述YUV格式图像中的Y通道图像以获得人脸图像的亮度信息。
根据本申请其中一实施例的一个方面，提供了一种肤质等级分类方法，所述方法包括：采用上述任一项肤质检测方法获取所述人脸肤质特征；根据所述人脸肤质特征，采用机器学习方法将人脸肤质分为不同等级。
可选地，所述方法包括：根据所述人脸肤质的不同等级设置对应的美颜参数。
可选地，所述方法应用于具有视频通话或拍照功能的电子设备上。
根据本申请其中一实施例的另一个方面，提供了一种肤质检测装置，包括：图像采集模块，配置为获取人脸图像；获取模块，配置为获取所述人脸图像中的人脸肤色区域和人脸特征点；肤质检测模块，配置为根据所述人脸肤色区域和所述人脸特征点检测人脸图像的人脸肤质特征。
可选地，所述获取模块包括肤色获取模块，配置为通过肤色检测算法获取所述人脸图像中的所述人脸肤色区域。
可选地，所述肤色获取模块还配置为根据所述人脸特征点抠除所述人脸图像的五官区域，得到所述人脸肤色区域。
可选地，所述肤色获取模块还配置为利用形态学算法处理所述人脸肤色区域，以扩大被抠除的所述五官区域。
可选地，所述人脸肤质特征包括以下至少之一：肤色、斑点、毛孔、 皱纹、黑眼圈、光滑度。
可选地，所述肤质检测模块还配置为采用高反差算法得到所述人脸肤色区域的细节图。
可选地，所述肤质检测模块还配置为根据所述人脸特征点获取人脸方向信息。
可选地，所述肤质检测模块包括斑点毛孔检测模块，所述斑点毛孔检测模块配置为采用第一预设算法获得所述细节图中所述斑点和/或毛孔的检测结果；根据形状特征区分所述斑点和/或毛孔。
可选地，所述肤质检测模块包括皮肤光滑度检测模块，配置为根据所述斑点和/或毛孔的检测结果，通过灰度共生矩阵算法获取所述皮肤光滑度。
可选地，所述肤质检测模块包括皱纹检测模块，配置为采用第二预设算法获取所述细节图中所述皱纹的检测结果；根据人脸方向信息和所述人脸特征点确定所述皱纹的种类。
可选地，所述肤质检测模块包括黑眼圈检测模块，配置为根据所述人脸特征点中的眼部特征点，在所述人脸图像中绘制出上下两条贝塞尔曲线，定位到所述黑眼圈的位置，并通过判别所述黑眼圈的位置的亮度均值与周围区域的差异性，获取所述人脸图像中所述黑眼圈的检测结果。
可选地，所述装置还包括图像增强模块，配置为对所述人脸图像进行增强处理，得到增强人脸图像。
可选地，所述图像增强模块包括：亮度信息获取单元，配置为获取所述人脸图像的亮度信息；对比度增强单元，配置为增强所述人脸图像中低灰度值区域的对比度，得到所述增强人脸图像。
可选地，亮度信息获取单元还配置为通过颜色转换算法将所述人脸图像转换为YUV格式图像，并提取所述YUV格式图像中的Y通道图像以获得人脸图像的亮度信息。
根据本申请其中一实施例的另一个方面，提供了一种电子设备，包括：处理器；以及存储器，配置为存储所述处理器的可执行指令；其中，所述处理器配置为经由执行所述可执行指令来执行上述任意一项肤质检测方法。
根据本申请其中一实施例的另一个方面，提供了一种存储介质，所述存储介质包括存储的程序，其中，在所述程序运行时控制所述存储介质所在设备执行上述任意一项肤质检测方法。
在本申请其中一实施例中，通过获取人脸图像；获取人脸图像中的人脸肤色区域和人脸特征点；根据人脸肤色区域和人脸特征点获取人脸图像的人脸肤质特征；可以基于人脸图像检测人脸肤质特征，并且可以在不增加硬件成本和体积的情况下，根据需要检测更多的肤质特征，提高检测精度。进而解决现有技术中检测的肤质特征少，检测精度差且硬件成本高、体积大的问题。
附图说明
此处所说明的附图用来提供对本申请的进一步理解，构成本申请的一部分，本申请的示意性实施例及其说明用于解释本申请，并不构成对本申请的不当限定。在附图中：
图1根据本申请其中一实施例的一种可选的肤质检测方法的流程图；
图2是根据本申请其中一实施例的一种可选的肤质等级分类方法的流程图；
图3是根据本申请其中一实施例的一种可选的肤质检测装置的结构框图；
图4是根据本申请其中一实施例一种可选的电子设备的结构框图。
具体实施方式
为了使本技术领域的人员更好地理解本申请方案，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分的实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本申请保护的范围。
需要说明的是，本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆 盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
本申请实施例可以应用于计算机系统/服务器，其可与众多其它通用或者专用计算系统环境或配置一起操作。适于与计算机系统/服务器一起使用的众所周知的计算系统、环境和/或配置的例子包括但不限于：个人计算机系统、手持或膝上设备、基于微处理器的系统、可编程消费电子产品、小型计算机系统、大型计算机系统和包括上述任何系统的分布式云计算技术环境，等等。
计算机系统/服务器可以在由计算机系统执行的计算机系统可执行指令(诸如程序模块等)的一般语境下描述。通常，程序模块可以包括例程、程序、目标程序、组件、逻辑以及数据结构等等，它们执行特定的任务或者实现特定的抽象数据类型。计算机系统/服务器可以在分布式云计算环境中实施，由通过通信网络链接的远程处理设备执行任务。在分布式云计算环境中，程序模块可以位于包括存储设备的本地或者远程计算系统存储介质上。
下面通过详细的实施例来说明本申请。
根据本申请的一个方面，提供了一种肤质检测方法。参考图1，是根据本申请其中一实施例的一种可选的肤质检测方法的流程图。如图1所示，该方法包括以下步骤：
S10：获取人脸图像；
S12：获取人脸图像中的人脸肤色区域和人脸特征点；
S14：根据人脸肤色区域和人脸特征点获取人脸图像的人脸肤质特征。
在本申请实施例中，通过上述步骤，即获取人脸图像；获取人脸图像中的人脸肤色区域和人脸特征点；根据人脸肤色区域和人脸特征点获取人脸图像的人脸肤质特征；可以基于人脸图像检测人脸肤质特征，并且可以在不增加硬件成本和体积的情况下，根据需要检测更多的肤质特征，提高检测精度。
下面对上述各步骤进行详细说明。
步骤S10，获取人脸图像；
可选的，在本申请其中一实施例中，可以通过图像采集模块获取人脸图像，其中，图像采集模块可以为独立的摄像装置或集成在电子设备上的摄像装置等，例如独立的RGB摄像头，或者车载电子设备(包括但不限于中控屏、导航器等)、手机、平板电脑、台式电脑、肤质检测仪等电子设备上自带的摄像头。
步骤S12：获取人脸图像中的人脸肤色区域和人脸特征点；
可选的，在本申请其中一实施例中，可以通过肤色检测算法获取人脸图像中的人脸肤色区域。例如，可以通过肤色检测算法将人脸图像转换到YCbCr颜色空间，然后将每个像素点的CrCb值代入根据皮肤像素点获得的椭圆统计模型，如果CrCb坐标在椭圆统计模型中，即判断为肤色，由此获得人脸肤色区域。人脸特征点可以通过SDM算法获得。
可选的，在本申请其中一实施例中，为了获得更准确的人脸肤色区域，可以根据人脸特征点抠除人脸图像中的五官区域，得到人脸肤色区域。另外，还可以利用形态学算法(例如，腐蚀膨胀算法)处理人脸肤色区域，以扩大被抠除的五官区域，从而避免因图像偏移所导致的五官区域未完全被抠除而残留的问题。
步骤S14：根据人脸肤色区域和人脸特征点获取人脸图像的人脸肤质特征。
可选的，在本申请其中一实施例中，人脸肤质特征包括以下至少之一：肤色、斑点、毛孔、皱纹、黑眼圈、光滑度。
可选的，在本申请其中一实施例中，当人脸肤质特征包括斑点和/或毛孔时，根据人脸肤色区域和人脸特征点获取人脸图像的人脸肤质特征包括：采用高反差算法得到人脸肤色区域的细节图；采用第一预设算法(例如，局部自适应阈值算法)获得细节图中斑点和/或毛孔的检测结果，其中，检测结果包括以下至少之一：位置、数量、面积；根据形状特征区分斑点和/或毛孔，其中，毛孔的形状特征为：面积普遍偏小并且类似圆形；而斑点的形状特征为：面积较大。优选的，可以在采用局部自适应阈值算法获得细节图中斑点和/或毛孔的检测结果后，采用形态学算法(例如，腐蚀膨胀算法)剔除斑点和/或毛孔的孤立点，排除噪声影响；采用连通域算法排除初始检测结果中形状异常或面积过大的错误点。优选的，上述步骤可以在抠除五官区域的人脸肤色区域中实现，以减少斑点和/或毛孔的误检率。
可选的，在本申请其中一实施例中，在获取斑点和/或毛孔的检测结果后，还可以根据斑点和/或毛孔的检测结果，通过灰度共生矩阵算法获取皮肤光滑度。例如，通过灰度共生矩阵算法计算0度，45度，90度以及135度的能量、熵、对比度和逆差分矩等参数，然后通过这16个参数去获取皮肤光滑度特征。
可选的，在本申请其中一实施例中，当人脸肤质特征包括皱纹时，根据人脸肤色区域和人脸特征点获取人脸图像的人脸肤质特征包括：采用高反差算法得到人脸肤色区域的细节图；根据人脸特征点获取人脸方向信息；采取第二预设算法获取细节图中皱纹的检测结果；根据人脸方向信息和人脸特征点确定皱纹的种类。其中，检测结果包括以下至少之一：位置、数量、面积。优选的，在采取第二预设算法获取细节图中皱纹的检测结果后，采用形态学算法和连通域算法剔除一些明显不属于皱纹特征的对象。
可选的，在本申请其中一实施例中，采取第二预设算法获取皱纹的细节图中皱纹的检测结果；根据人脸方向信息和人脸特征点区分皱纹的种类可以包括：采取局部自适应阈值算法获取皱纹位置，并根据人脸特征点和人脸方向信息确定偏水平方向的皱纹为眼周纹；采用Canny边缘提取算法获取皱纹位置，并根据人脸特征点和人脸方向信息确定偏水平方向的皱纹为抬头纹，偏对角线方向的皱纹为法令纹。
可选的，在本申请其中一实施例中，当人脸肤质特征包括黑眼圈时，根据人脸肤色区域和所述人脸特征点获取人脸图像的人脸肤质特征包括：根据人脸特征点中的眼部特征点，在人脸图像中绘制上下两条贝塞尔曲线， 定位黑眼圈检测区域；计算黑眼圈检测区域的亮度均值与周围区域的差异，确定黑眼圈的严重程度。
可选的，在本申请其中一实施例中，在步骤S10获取人脸图像之后，还可以包括步骤S11：对人脸图像进行增强处理，得到增强人脸图像。
可选的，在本申请其中一实施例中，对人脸图像进行增强处理，得到人脸图像包括：通过第三预定算法获取人脸图像的亮度信息；增强人脸图像中低灰度值区域的对比度，得到增强人脸图像。由此，可以增强暗光下的人脸图像的细节。优选地，人脸肤质特征可以在增强人脸图像上获取，以获得更精确的检测结果。其中，对人脸图像中的低灰度值区域进行扩展的方法包括但不限于对数变换、直方图均衡化、指数变换等。
具体地，通过第三预定算法获取人脸图像的亮度信息可以是通过颜色转换算法将人脸图像转换为YUV格式图像，并提取YUV格式图像中的Y通道图像以获得人脸图像的亮度信息。
当然，本领域技术人员可知，在对人脸图像进行增强处理，得到人脸图像之前或之后，还可以包括检测人脸框，根据检测到的人脸框大小对图像进行缩放，以获得大小符合要求的人脸图像。
通过上述步骤，可以实现人脸肤质检测，检测的肤质特征可以包括斑点、毛孔、皮肤光滑度、皱纹、黑眼圈等，并且可以在不增加硬件成本和体积的情况下，提高检测精度。
可选的，在本申请其中一实施例中，在检测到各类肤质特征后，还可 以结合人脸肤色区域对肤质进行评价，例如，对于斑点，可以使用斑点面积与人脸肤色区域的面积比作为一个指标，评价斑点的严重程度。
根据本申请其中一实施例的另一方面，还提供了一种肤质等级分类方法。参考图2，是根据本申请其中一实施例的一种可选的肤质等级分类方法的流程图。如图2所示，该方法包括以下步骤：
S20：采用上述肤质检测方法获取人脸肤质特征；
S22：根据人脸肤质特征，采用机器学习方法将人脸肤质分为不同等级。
可选的，在本申请其中一实施例中，机器学习方法可以是支持向量机或感知机。具体地，机器学习方法可以通过训练样本，得到优选的分类函数，其中每个样本的人脸肤质特征包括以下至少之一：肤色、斑点、毛孔、皱纹、黑眼圈、光滑度。
在本申请其中一实施例中，通过上述步骤，即用上述肤质检测方法获取人脸肤质特征；根据人脸肤质特征，采用机器学习方法将人脸肤质分为不同等级；除了可以检测人脸肤质特征，还可以将人脸肤质分为不同等级，方便为不同用户提供护理建议、推荐适合的护肤品或实现智能美颜。
对于智能美颜方面，在本申请其中一实施例的一个应用场景中，可以根据人脸肤质的不同等级设置对应的美颜参数实现智能美颜。例如，如果一个人的皮肤毛孔、皱纹以及斑点较多，皮肤比较粗糙，将该人脸肤质的等级定义为较差，则据此可设置对应的美颜参数，例如，将美颜参数中的 磨皮强度设置强一点。相反，如果一个人的皮肤毛孔、皱纹以及斑点较少，皮肤光滑，将该人脸肤质的等级定义为较好，则据此可设置对应的美颜参数，例如，将美颜参数中的磨皮强度设置低一点；从而保证每个人的美颜参数都是优选的，达到自然的美颜效果，实现智能美颜。通过上述根据人脸肤质的不同等级设置对应的美颜参数实现的智能美颜技术能够搭载于具有视频通话或拍照等功能的电子设备上，例如车载电子设备(包括但不限于中控屏、导航器等)、手机、数码相机、平板电脑、台式电脑、肤质检测仪等。在一种应用环境中，例如，在具有辅助驾驶功能的汽车中，特别是具有自动驾驶功能的汽车中，由于人们开车时不用进行方向盘操控等动作，因此在空闲时刻可能会进行视频聊天、开视频会议、拍照等，通过在上述具有视频通话或拍照等功能车载电子设备上搭载智能美颜技术，能够提升用户的体验感。
根据本申请其中一实施例的另一方面，还提供了一种肤质检测装置。参考图3，是根据本申请其中一实施例的一种可选的肤质检测装置的结构框图。如图3所示，肤质检测装置3包括：
图像采集模块30，配置为获取人脸图像；
可选的，在本申请其中一实施例中，图像采集模块可以为独立的摄像装置或集成在电子设备上的摄像装置等，例如独立的RGB摄像头，或者手机、平板电脑、台式电脑、肤质检测仪等电子设备上自带的摄像头。
获取模块32，配置为获取人脸图像中的人脸肤色区域和人脸特征点；
可选的，在本申请其中一实施例中，获取模块34包括肤色获取模块 和特征点获取模块。其中，肤色获取模块配置为通过肤色检测算法获取人脸图像中的人脸肤色区域。例如，可以通过肤色检测算法将人脸图像转换到YCbCr颜色空间，然后将每个像素点的CrCb值代入根据皮肤像素点获得的椭圆统计模型，如果CrCb坐标在椭圆统计模型中，即判断为肤色，由此获得人脸肤色区域。人脸特征点可以通过SDM算法获得。
可选的，在本申请其中一实施例中，为了获得更准确的人脸肤色区域，肤色获取模块配置为根据人脸特征点抠除人脸图像中的五官区域，得到人脸肤色区域。另外，还配置为利用形态学算法(例如，腐蚀膨胀算法)处理人脸肤色区域，以扩大被抠除的五官区域，从而避免因图像偏移所导致的五官区域未完全被抠除而残留的问题。
肤质检测模块34，配置为根据人脸肤色区域和人脸特征点检测人脸图像的人脸肤质特征。
可选的，在本申请其中一实施例中，人脸肤质特征包括以下至少之一：肤色、斑点、毛孔、皱纹、黑眼圈、光滑度。
可选的，在本申请其中一实施例中，肤质检测模块34包括斑点毛孔检测模块340，所述斑点毛孔检测模块340配置为采用高反差算法得到人脸肤色区域的细节图；采用第一预设算法(例如，局部自适应阈值算法)获得细节图中斑点和/毛孔的检测结果，其中，检测结果包括以下至少之一：位置、数量、面积；根据形状特征区分斑点和/或毛孔，其中，毛孔的形状特征为：面积普遍偏小并且类似圆形；而斑点的形状特征为：面积较大。优选的，可以在采用局部自适应阈值算法获得细节图中斑点和/或 毛孔的检测结果后，采用形态学算法(例如，腐蚀膨胀算法)剔除斑点和/毛孔的孤立点，排除噪声影响；采用连通域算法排除初始检测结果中形状异常或面积过大的错误点。优选的，上述步骤可以在抠除五官区域的人脸肤色区域中实现，以减少斑点和/或毛孔的误检率。
可选的，在本申请其中一实施例中，肤质检测模块34包括皮肤光滑度检测模块342，配置为在获取斑点和/或毛孔的检测结果后，根据斑点和/或毛孔的检测结果，通过灰度共生矩阵算法获取皮肤光滑度。例如，通过灰度共生矩阵算法计算0度，45度，90度以及135度的能量、熵、对比度和逆差分矩等参数，然后通过这16个参数去获取皮肤光滑度特征。
可选的，在本申请其中一实施例中，肤质检测模块34包括皱纹检测模块344。皱纹检测模块344配置为采用高反差算法得到人脸图像的细节图；根据人脸特征点获取人脸方向信息；采取第二预设算法获取细节图中皱纹的检测结果，其中，检测结果包括以下至少之一：位置、数量、面积；根据人脸方向信息和人脸特征点区分皱纹的种类。优选的，在采取第二预设算法获取细节图中皱纹的检测结果后，采用形态学算法和连通域算法剔除一些明显不属于皱纹特征的对象。
可选的，在本申请其中一实施例中，采取第二预设算法获取皱纹的细节图中皱纹的检测结果；根据人脸方向信息和人脸特征点区分皱纹的种类可以包括：采取局部自适应阈值算法获取皱纹位置，并根据人脸方向信息和人脸特征点将偏水平方向的皱纹标记为眼周纹；采用Canny边缘提取算法获取皱纹位置，并根据人脸方向信息和人脸特征点将偏水平方向的皱纹 标记为抬头纹，将偏对角线方向的皱纹标记为法令纹。
可选的，在本申请其中一实施例中，肤质检测模块34包括黑眼圈检测模块346。黑眼圈检测模块346配置为根据人脸特征点中的眼部特征点，在人脸图像中绘制上下两条贝塞尔曲线，定位黑眼圈检测区域；计算黑眼圈检测区域的亮度均值与周围区域的差异，确定黑眼圈的严重程度。
可选的，在本申请其中一实施例中，肤质检测装置3还包括图像增强模块31，配置为对人脸图像进行处理，得到人脸图像。
可选的，在本申请其中一实施例中，图像增强模块31包括亮度信息获取单元310和对比度增强单元312。其中，亮度信息获取单元310，配置为通过第三预定算法获取人脸图像的亮度信息；对比度增强单元312，配置为增强人脸图像中低灰度值区域的对比度，得到增强人脸图像。由此，可以增强暗光下的人脸图像的细节。优选地，人脸肤质特征可以在增强人脸图像上获取，以获得更精确的检测结果。其中，对人脸图像中的低灰度值区域进行扩展的方法包括但不限于对数变换、直方图均衡化、指数变换等。
具体地，通过第三预定算法获取人脸图像的亮度信息可以是通过颜色转换算法将人脸图像转换为YUV格式图像，并提取YUV格式图像中的Y通道图像以获得人脸图像的亮度信息。
当然，本领域技术人员可知，在对人脸图像进行增强处理，得到人脸图像之前或之后，还可以包括对图像进行缩放，以获得大小符合要求的人脸图像。
在本申请其中一实施例中，通过上述模块，即图像采集模块30，配置为获取人脸图像；获取模块34，配置为获取人脸图像中的人脸肤色区域和人脸特征点；肤质检测模块36，配置为根据人脸肤色区域和人脸特征点检测人脸图像的人脸肤质特征；可以基于人脸图像检测多种人脸肤质特征，并且可以在不增加硬件成本和体积的情况下，根据需要检测更多的肤质特征，提高检测精度。
根据本申请其中一实施例的另一方面，还提供了一种电子设备，电子设备40包括：处理器400；以及存储器402，配置为存储所述处理器400的可执行指令；其中，所述处理器400被配置为经由执行所述可执行指令来执行上述任意一项所述的肤质检测方法。
根据本申请其中一实施例的另一方面，还提供了一种存储介质，其中，所述存储介质包括存储的程序，其中，在所述程序运行时控制所述存储介质所在设备执行上述任意一项所述的肤质检测方法。
在本申请的上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。
在本申请所提供的几个实施例中，应该理解到，所揭露的技术内容，可通过其它的方式实现。其中，以上所描述的装置实施例仅仅是示意性的，例如所述单元的划分，可以为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，单元或模块的间接耦合或 通信连接，可以是电性或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可为个人计算机、服务器或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、移动硬盘、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述仅是本申请的优选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本申请原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也应视为本申请的保护范围。
工业实用性
通过获取人脸图像；获取人脸图像中的人脸肤色区域和人脸特征点；根据人脸肤色区域和人脸特征点获取人脸图像的人脸肤质特征；可以基于人脸图像检测人脸肤质特征，并且可以在不增加硬件成本和体积的情况下，根据需要检测更多的肤质特征，提高检测精度。进而解决现有技术中检测的肤质特征少，检测精度差且硬件成本高、体积大的问题。

Claims (33)

  1. 一种肤质检测方法，所述方法包括：
    获取人脸图像；
    获取所述人脸图像中的人脸肤色区域和人脸特征点；
    根据所述人脸肤色区域和所述人脸特征点获取所述人脸图像的人脸肤质特征。
  2. 根据权利要求1所述的方法，其中，获取所述人脸图像中的人脸肤色区域和人脸特征点包括：
    采用肤色检测算法获取所述人脸肤色区域。
  3. 根据权利要求1所述的方法，其中，所述方法还包括：根据所述人脸特征点抠除所述人脸图像的五官区域，得到所述人脸肤色区域。
  4. 根据权利要求3所述的方法，其中，所述方法还包括：
    利用形态学算法处理所述人脸肤色区域，以扩大被抠除的所述五官区域。
  5. 根据权利要求1所述的方法，其中，所述人脸肤质特征包括以下至少之一：肤色、斑点、毛孔、皱纹、黑眼圈、光滑度。
  6. 根据权利要求1所述的方法，其中，根据所述人脸肤色区域和所述人脸特征点获取所述人脸图像的人脸肤质特征包括：
    采用高反差算法得到所述人脸肤色区域的细节图。
  7. 根据权利要求1所述的方法，其中，所述方法还包括：
    根据所述人脸特征点获取人脸方向信息。
  8. 根据权利要求6所述的方法，其中，当所述人脸肤质特征包括斑点和/或毛孔时，根据所述人脸肤色区域和所述人脸特征点获取所述人脸图像的人脸肤质特征还包括：
    采用第一预设算法获得所述细节图中所述斑点和/或毛孔的检测结果；
    根据形状特征区分所述斑点和/或毛孔。
  9. 根据权利要求8所述的方法，其中，当所述人脸肤质特征包括皮肤光滑度时，所述方法包括：
    根据所述斑点和/或毛孔的检测结果，通过灰度共生矩阵算法获取所述皮肤光滑度。
  10. 根据权利要求6所述的方法，其中，当所述人脸肤质特征包括皱纹时，根据所述人脸肤色区域和所述人脸特征点获取所述人脸图像的人脸肤质特征还包括：
    采用第二预设算法获取所述细节图中所述皱纹的检测结果；
    根据人脸方向信息和所述人脸特征点确定所述皱纹的种类。
  11. 根据权利要求1所述的方法，其中，当所述人脸肤质特征包括黑眼圈时，根据所述人脸肤色区域和所述人脸特征点获取所述人脸图像的人 脸肤质特征包括：
    根据所述人脸特征点中的眼部特征点，在所述人脸图像中绘制出上下两条贝塞尔曲线，定位到所述黑眼圈的位置，并通过判别所述黑眼圈的位置的亮度均值与周围区域的差异性，获取所述人脸图像中所述黑眼圈的检测结果。
  12. 根据权利要求1至11任一项所述的方法，还包括：
    对所述人脸图像进行增强处理，得到增强人脸图像。
  13. 根据权利要求12所述的方法，其中，对所述人脸图像进行增强处理，得到增强人脸图像包括：
    获取所述人脸图像的亮度信息；
    增强所述人脸图像中低灰度值区域的对比度，得到所述增强人脸图像。
  14. 根据权利要求13所述的方法，其中，获取所述人脸图像的亮度信息包括：
    通过颜色转换算法将所述人脸图像转换为YUV格式图像，并提取所述YUV格式图像中的Y通道图像以获得人脸图像的亮度信息。
  15. 一种肤质等级分类方法，所述方法包括：
    采用如权利要求1至14中任一项所述的肤质检测方法获取所述人脸肤质特征；
    根据所述人脸肤质特征，采用机器学习方法将人脸肤质分为不同等级。
  16. 根据权利要求15所述的方法，其中，所述方法包括：
    根据所述人脸肤质的不同等级设置对应的美颜参数，实现智能美颜。
  17. 根据权利要求16所述的方法，其中，所述方法应用于具有视频通话或拍照功能的电子设备上。
  18. 一种肤质检测装置，所述装置包括：
    图像采集模块，配置为获取人脸图像；
    获取模块，配置为获取所述人脸图像中的人脸肤色区域和人脸特征点；
    肤质检测模块，配置为根据所述人脸肤色区域和所述人脸特征点检测人脸图像的人脸肤质特征。
  19. 根据权利要求18所述的装置，其中，所述获取模块包括肤色获取模块，配置为通过肤色检测算法获取所述人脸图像中的所述人脸肤色区域。
  20. 根据权利要求19所述的装置，其中，所述肤色获取模块还配置为根据所述人脸特征点抠除所述人脸图像的五官区域，得到所述人脸肤色区域。
  21. 根据权利要求20所述的装置，其中，所述肤色获取模块还配置为利用 形态学算法处理所述人脸肤色区域，以扩大被抠除的所述五官区域。
  22. 根据权利要求18所述的装置，其中，所述人脸肤质特征包括以下至少之一：肤色、斑点、毛孔、皱纹、黑眼圈、光滑度。
  23. 根据权利要求18所述的装置，其中，所述肤质检测模块还配置为采用高反差算法得到所述人脸肤色区域的细节图。
  24. 根据权利要求18所述的装置，其中，所述肤质检测模块还配置为根据所述人脸特征点获取人脸方向信息。
  25. 根据权利要求23所述的装置，其中，所述肤质检测模块包括斑点毛孔检测模块，所述斑点毛孔检测模块配置为采用第一预设算法获得所述细节图中所述斑点和/或毛孔的检测结果；根据形状特征区分所述斑点和/或毛孔。
  26. 根据权利要求25所述的装置，其中，所述肤质检测模块包括皮肤光滑度检测模块，配置为根据所述斑点和/或毛孔的检测结果，通过灰度共生矩阵算法获取所述皮肤光滑度。
  27. 根据权利要求23所述的装置，其中，所述肤质检测模块包括皱纹检测模块，配置为采用第二预设算法获取所述细节图中所述皱纹的检测结果；根据人脸方向信息和所述人脸特征点确定所述皱纹的种类。
  28. 根据权利要求18所述的装置，其中，所述肤质检测模块包括黑眼圈检测模块，配置为根据所述人脸特征点中的眼部特征点，在所述人脸图像中绘制出上下两条贝塞尔曲线，定位到所述黑眼圈的位置，并通过 判别所述黑眼圈的位置的亮度均值与周围区域的差异性，获取所述人脸图像中所述黑眼圈的检测结果。
  29. 根据权利要求18至28任一项所述的装置，还包括图像增强模块，配置为对所述人脸图像进行增强处理，得到增强人脸图像。
  30. 根据权利要求29所述的装置，所述图像增强模块包括：
    亮度信息获取单元，配置为获取所述人脸图像的亮度信息；
    对比度增强单元，配置为增强所述人脸图像中低灰度值区域的对比度，得到所述增强人脸图像。
  31. 根据权利要求30所述的装置，其中，所述亮度信息获取单元还配置为通过颜色转换算法将所述人脸图像转换为YUV格式图像，并提取所述YUV格式图像中的Y通道图像以获得人脸图像的亮度信息。
  32. 一种电子设备，包括：
    处理器；以及
    存储器，配置为存储所述处理器的可执行指令；
    其中，所述处理器配置为经由执行所述可执行指令来执行权利要求1至14中任意一项所述的肤质检测方法。
  33. 一种存储介质，所述存储介质包括存储的程序，其中，在所述程序运行时控制所述存储介质所在设备执行权利要求1至14中任意一项所述的肤质检测方法。
PCT/CN2020/083894 2019-04-12 2020-04-09 肤质检测方法、肤质等级分类方法及肤质检测装置 WO2020207423A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
KR1020217037077A KR20210149848A (ko) 2019-04-12 2020-04-09 피부질 검출 방법, 피부질 등급 분류 방법, 피부질 검출 장치, 전자 기기 및 저장 매체
US17/603,012 US20220237811A1 (en) 2019-04-12 2020-04-09 Method for Testing Skin Texture, Method for Classifying Skin Texture and Device for Testing Skin Texture
JP2021559127A JP7413400B2 (ja) 2019-04-12 2020-04-09 肌質測定方法、肌質等級分類方法、肌質測定装置、電子機器及び記憶媒体

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910292616.3 2019-04-12
CN201910292616.3A CN111814520A (zh) 2019-04-12 2019-04-12 肤质检测方法、肤质等级分类方法及肤质检测装置

Publications (1)

Publication Number Publication Date
WO2020207423A1 true WO2020207423A1 (zh) 2020-10-15

Family

ID=72750958

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/083894 WO2020207423A1 (zh) 2019-04-12 2020-04-09 肤质检测方法、肤质等级分类方法及肤质检测装置

Country Status (5)

Country Link
US (1) US20220237811A1 (zh)
JP (1) JP7413400B2 (zh)
KR (1) KR20210149848A (zh)
CN (1) CN111814520A (zh)
WO (1) WO2020207423A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112533024A (zh) * 2020-11-26 2021-03-19 北京达佳互联信息技术有限公司 一种人脸视频处理方法、装置及存储介质
CN113411507A (zh) * 2021-05-10 2021-09-17 深圳数联天下智能科技有限公司 测肤图像的获取方法、装置、设备及存储介质
CN113397480A (zh) * 2021-05-10 2021-09-17 深圳数联天下智能科技有限公司 美容仪的控制方法、装置、设备及存储介质

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111507944B (zh) * 2020-03-31 2023-07-04 北京百度网讯科技有限公司 皮肤光滑度的确定方法、装置和电子设备
CN116634925A (zh) * 2020-08-03 2023-08-22 强生消费者公司 选择性施用化妆品组合物以赋予眼下亮白的系统和方法
CN112749708B (zh) * 2021-01-28 2024-04-23 深圳艾摩米智能科技有限公司 面部雀斑检测分析方法、装置、计算机设备及存储介质
CN113128377A (zh) * 2021-04-02 2021-07-16 西安融智芙科技有限责任公司 基于图像处理的黑眼圈识别方法、黑眼圈识别装置及终端
CN113034354B (zh) * 2021-04-20 2021-12-28 北京优彩科技有限公司 一种图像处理方法、装置、电子设备和可读存储介质
WO2022226728A1 (en) * 2021-04-26 2022-11-03 L'oreal Computing device, method and apparatus for detecting skin conditions of human subject
CN113191322A (zh) * 2021-05-24 2021-07-30 口碑(上海)信息技术有限公司 人脸皮肤肤质的检测方法、装置、存储介质及计算机设备
CN113486768A (zh) * 2021-07-01 2021-10-08 成都九章丽欣科技有限公司 一种肤质的图像识别方法
CN113610844A (zh) * 2021-08-31 2021-11-05 深圳市邻友通科技发展有限公司 一种智能护肤方法、装置、设备和存储介质
CN113576421A (zh) * 2021-09-10 2021-11-02 广州帕克西软件开发有限公司 一种智能测肤方法及装置
US20230196835A1 (en) * 2021-12-16 2023-06-22 The Gillette Company Llc Digital imaging systems and methods of analyzing pixel data of an image of a skin area of a user for determining dark eye circles
US20230196816A1 (en) * 2021-12-16 2023-06-22 The Gillett Company LLC Digital imaging systems and methods of analyzing pixel data of an image of a skin area of a user for determining skin hyperpigmentation
CN115937199B (zh) * 2023-01-06 2023-05-23 山东济宁圣地电业集团有限公司 一种用于电力配电柜绝缘层的喷涂质量检测方法
CN116167949B (zh) * 2023-04-25 2023-06-30 天津中智云海软件科技有限公司 一种基于医疗影像大数据的临床辅助决策系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8437516B2 (en) * 2009-04-30 2013-05-07 Novatek Microelectronics Corp. Facial expression recognition apparatus and facial expression recognition method thereof
CN104299011A (zh) * 2014-10-13 2015-01-21 吴亮 一种基于人脸图像识别的肤质与皮肤问题识别检测方法
CN104732214A (zh) * 2015-03-24 2015-06-24 吴亮 一种基于人脸图像识别的定量化肤质检测方法
CN107369133A (zh) * 2016-05-13 2017-11-21 炬芯（珠海）科技有限公司 一种人脸图像美化方法和装置
CN109522839A (zh) * 2018-11-15 2019-03-26 北京达佳互联信息技术有限公司 一种人脸皮肤区域确定方法、装置、终端设备及存储介质

Family Cites Families (30)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7082211B2 (en) * 2002-05-31 2006-07-25 Eastman Kodak Company Method and system for enhancing portrait images
KR100467610B1 (ko) * 2002-09-06 2005-01-24 삼성전자주식회사 디지털 화질 개선 방법 및 장치
JP2004252511A (ja) 2003-02-18 2004-09-09 Hitachi Ltd 顔向き推定方法
US7903902B2 (en) * 2004-07-26 2011-03-08 Sheraizin Semion M Adaptive image improvement
US7639892B2 (en) * 2004-07-26 2009-12-29 Sheraizin Semion M Adaptive image improvement
TW200629232A (en) * 2005-02-15 2006-08-16 Quanta Comp Inc Dynamic image contrast enhancement device
JP4217698B2 (ja) * 2005-06-20 2009-02-04 キヤノン株式会社 撮像装置及び画像処理方法
FR2907569B1 (fr) 2006-10-24 2009-05-29 Jean Marc Robin Procede et dispositif de simulation virtuelle d'une sequence d'images video.
JP4479709B2 (ja) * 2006-10-27 2010-06-09 セイコーエプソン株式会社 画像表示装置、画像表示方法、画像表示プログラム、及び画像表示プログラムを記録した記録媒体、並びに電子機器
CN106919911A (zh) * 2008-07-30 2017-07-04 快图有限公司 使用脸部检测的自动脸部和皮肤修饰
KR101475684B1 (ko) * 2008-10-17 2014-12-23 삼성전자주식회사 디지털 영상 처리기에서 얼굴 영상 개선 장치 및 방법
JP5538909B2 (ja) 2010-01-05 2014-07-02 キヤノン株式会社 検出装置およびその方法
FR2962025B1 (fr) 2010-07-02 2013-04-12 Chanel Parfums Beaute Procede et dispositif de detection et de quantification de signes cutanes sur une zone de peau
JP5657494B2 (ja) 2011-09-22 2015-01-21 富士フイルム株式会社 シワ検出方法、シワ検出装置およびシワ検出プログラム、並びに、シワ評価方法、シワ評価装置およびシワ評価プログラム
WO2013078404A1 (en) * 2011-11-22 2013-05-30 The Trustees Of Dartmouth College Perceptual rating of digital image retouching
US20140313303A1 (en) * 2013-04-18 2014-10-23 Digimarc Corporation Longitudinal dermoscopic study employing smartphone-based image registration
US9355406B2 (en) * 2013-07-18 2016-05-31 GumGum, Inc. Systems and methods for determining image safety
US20150099947A1 (en) * 2013-10-04 2015-04-09 Access Business Group International Llc Skin youthfulness index, methods and applications thereof
JP6273640B2 (ja) 2013-10-23 2018-02-07 パナソニックＩｐマネジメント株式会社 撮影画像表示装置
US10430985B2 (en) * 2014-03-14 2019-10-01 Magic Leap, Inc. Augmented reality systems and methods utilizing reflections
JP2016009203A (ja) 2014-06-20 2016-01-18 パナソニックＩｐマネジメント株式会社 皺検出装置および皺検出方法
JP6138745B2 (ja) 2014-11-19 2017-05-31 株式会社 資生堂 シミ評価装置、及びシミ評価プログラム
US10055821B2 (en) * 2016-01-30 2018-08-21 John W. Glotzbach Device for and method of enhancing quality of an image
JP6722866B2 (ja) 2016-02-29 2020-07-15 パナソニックＩｐマネジメント株式会社 画像処理装置および画像処理方法
CN106469302B (zh) * 2016-09-07 2019-05-28 成都知识视觉科技有限公司 一种基于人工神经网络的人脸肤质检测方法
CN107392841B (zh) * 2017-06-16 2020-04-24 Oppo广东移动通信有限公司 人脸区域中黑眼圈消除方法、装置及终端
CN107862659B (zh) * 2017-10-31 2020-05-26 Oppo广东移动通信有限公司 图像处理方法、装置、计算机设备及计算机可读存储介质
US10990858B2 (en) * 2018-01-05 2021-04-27 L'oreal Machine-implemented facial health and beauty assistant
US10593023B2 (en) * 2018-02-13 2020-03-17 Adobe Inc. Deep-learning-based automatic skin retouching
CN109086688A (zh) * 2018-07-13 2018-12-25 北京科莱普云技术有限公司 脸部皱纹检测方法、装置、计算机设备和存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8437516B2 (en) * 2009-04-30 2013-05-07 Novatek Microelectronics Corp. Facial expression recognition apparatus and facial expression recognition method thereof
CN104299011A (zh) * 2014-10-13 2015-01-21 吴亮 一种基于人脸图像识别的肤质与皮肤问题识别检测方法
CN104732214A (zh) * 2015-03-24 2015-06-24 吴亮 一种基于人脸图像识别的定量化肤质检测方法
CN107369133A (zh) * 2016-05-13 2017-11-21 炬芯（珠海）科技有限公司 一种人脸图像美化方法和装置
CN109522839A (zh) * 2018-11-15 2019-03-26 北京达佳互联信息技术有限公司 一种人脸皮肤区域确定方法、装置、终端设备及存储介质

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112533024A (zh) * 2020-11-26 2021-03-19 北京达佳互联信息技术有限公司 一种人脸视频处理方法、装置及存储介质
CN113411507A (zh) * 2021-05-10 2021-09-17 深圳数联天下智能科技有限公司 测肤图像的获取方法、装置、设备及存储介质
CN113397480A (zh) * 2021-05-10 2021-09-17 深圳数联天下智能科技有限公司 美容仪的控制方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN111814520A (zh) 2020-10-23
JP2022528128A (ja) 2022-06-08
JP7413400B2 (ja) 2024-01-15
US20220237811A1 (en) 2022-07-28
KR20210149848A (ko) 2021-12-09

Similar Documents

Publication Publication Date Title
WO2020207423A1 (zh) 肤质检测方法、肤质等级分类方法及肤质检测装置
WO2017092431A1 (zh) 基于肤色的人手检测方法及装置
CN111488756B (zh) 基于面部识别的活体检测的方法、电子设备和存储介质
US20190034702A1 (en) Living body detecting method and apparatus, device and storage medium
WO2019100282A1 (zh) 一种人脸肤色识别方法、装置和智能终端
WO2018086543A1 (zh) 活体判别方法、身份认证方法、终端、服务器和存储介质
EP3767520B1 (en) Method, device, equipment and medium for locating center of target object region
US8824747B2 (en) Skin-tone filtering
TW202014984A (zh) 一種圖像處理方法、電子設備及存儲介質
CN111062891A (zh) 图像处理方法、装置、终端及计算机可读存储介质
WO2019184851A1 (zh) 图像处理方法和装置及神经网络模型的训练方法
CN108234770B (zh) 一种辅助化妆系统、辅助化妆方法、辅助化妆装置
GB2598016A (en) Biometric authentication apparatus and biometric authentication method
CN110473176B (zh) 图像处理方法及装置、眼底图像处理方法、电子设备
CN110222647A (zh) 一种基于卷积神经网络的人脸活体检测方法
CN113128373A (zh) 基于图像处理的色斑评分方法、色斑评分装置及终端设备
CN112396016A (zh) 一种基于大数据技术的人脸识别系统
CN112651962A (zh) 一种ai智能诊断系统平台
CN111582278B (zh) 人像分割方法、装置及电子设备
CN110766631A (zh) 人脸图像的修饰方法、装置、电子设备和计算机可读介质
CN111222374A (zh) 测谎数据处理方法、装置、计算机设备和存储介质
US11205064B1 (en) Measuring quality of depth images in real time
WO2008018459A1 (fr) Procédé de traitement d'image, appareil de traitement d'image, programme de traitement d'image, et appareil de prise d'image
US20220392252A1 (en) Facial Skin Detection Method and Apparatus
CN112115833B (zh) 一种图片的活体检测及分类方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20788552

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021559127

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 20217037077

Country of ref document: KR

Kind code of ref document: A

122 Ep: pct application non-entry in european phase

Ref document number: 20788552

Country of ref document: EP

Kind code of ref document: A1