WO2020108322A1 - 检测和评价黑眼圈的方法和装置 - Google Patents

检测和评价黑眼圈的方法和装置 Download PDF

Info

Publication number
WO2020108322A1
WO2020108322A1 PCT/CN2019/118635 CN2019118635W WO2020108322A1 WO 2020108322 A1 WO2020108322 A1 WO 2020108322A1 CN 2019118635 W CN2019118635 W CN 2019118635W WO 2020108322 A1 WO2020108322 A1 WO 2020108322A1
Authority
WO
WIPO (PCT)
Prior art keywords
color
dark
area
interest
type
Prior art date
Application number
PCT/CN2019/118635
Other languages
English (en)
French (fr)
Inventor
姚烨
董辰
丁欣
胡宏伟
郜文美
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to US17/297,684 priority Critical patent/US11779264B2/en
Priority to EP19891460.8A priority patent/EP3879437B1/en
Publication of WO2020108322A1 publication Critical patent/WO2020108322A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/44—Detecting, measuring or recording for evaluating the integumentary system, e.g. skin, hair or nails
    • A61B5/441—Skin evaluation, e.g. for skin disorder diagnosis
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/1032—Determining colour for diagnostic purposes
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/44—Detecting, measuring or recording for evaluating the integumentary system, e.g. skin, hair or nails
    • A61B5/441—Skin evaluation, e.g. for skin disorder diagnosis
    • A61B5/444—Evaluating skin marks, e.g. mole, nevi, tumour, scar
    • 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
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition
    • G06V30/18—Extraction of features or characteristics of the image
    • G06V30/18105—Extraction of features or characteristics of the image related to colour
    • 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
    • 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
    • G06V40/172—Classification, e.g. identification
    • 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
    • G06V40/18—Eye characteristics, e.g. of the iris
    • G06V40/193—Preprocessing; Feature extraction

Definitions

  • Embodiments of the present application relate to the field of image processing, and more specifically, to a method and device for detecting and evaluating dark circles.
  • facial skin quality can be evaluated by the eight dimensions of fine lines, pores, blackheads, red areas, spots, muscle age, decree lines, and dark circles.
  • the dark eye is ranked top.
  • dark circles have always been very concerned about facial skin problems.
  • the present application provides a method and device for detecting and evaluating dark circles, which can effectively detect and evaluate dark circles in an image.
  • a method for detecting and evaluating dark circles includes: acquiring an image to be processed; extracting dark circles of interest in the to-be-processed image; and performing color aggregation on the dark circles of interest Class to obtain n types of colors in the dark circle of interest region, where n is a positive integer; according to the n types of colors, identify the dark circles in the dark circle of interest region; obtain based on the dark circles region Evaluation results of dark circles.
  • extract the black eye area of interest in the to-be-processed image perform color clustering on the black eye area of interest to obtain n types of colors, and accurately identify according to the n types of colors
  • the dark circle area in the dark circle area of interest is obtained, and the dark circle evaluation result is obtained based on the dark circle area, so that the dark circle area in the image can be effectively detected and evaluated.
  • identifying the black eye area within the black eye area of interest according to the n-type colors includes: determining the black eye color and the reference skin color according to the n-type colors; The dark circle color and the reference skin color determine the dark circle area in the dark circle interest area.
  • the first color of the n-type colors is determined as the color of the dark circles in the dark circle of interest region
  • the second color of the n-type colors is determined as the reference skin color in the dark circle of interest area.
  • the first color is the darkest color among the n-type colors after removing noise
  • the second color is the lightest color among the n-type colors after removing bright light.
  • the determining the black eye area in the black eye area of interest according to the black eye color and the reference skin color includes: if the black eye area of interest One pixel satisfies:
  • the first pixel belongs to the dark circle area, where C is the first pixel, CD is the color of the dark circle, CS is the reference skin color, and T1 is the relative of the first pixel The contrast between the color of the dark circles and the color of the reference skin color.
  • the first pixel in the black eye area of interest satisfies:
  • C is the first pixel
  • CS is the reference skin color
  • T2 represents the lowest contrast of the two colors that can be distinguished by the human eye.
  • the reference skin color satisfies:
  • CS is the reference skin color
  • CS belongs to the i- th color among the n - type colors
  • CS i-1 is the i-1- th color among the n-type colors
  • i is A positive integer greater than or equal to 1 and less than or equal to n
  • T3 represents the minimum value of the color in the bright area
  • T4 represents the difference between the i-th color and the i-1th color.
  • the extracting the dark eye area of interest in the image to be processed includes: removing the eyelash area in the image to be processed; and extracting the image to be processed with the eyelash area removed The region of interest in the dark circles.
  • the obtaining the evaluation results of the dark circles based on the dark circles includes: extracting features of the dark circles, the features including the contrast of the dark circles and the dark circles At least one of the area of the area or the variance of the dark circles; according to the characteristics, the severity of the dark circles is evaluated by a pattern recognition method.
  • the obtaining of a dark circle evaluation result based on the dark circle region includes: determining black according to the position of the darkest region in the n-type colors and/or the color of the dark circle region Types of eye circles.
  • the black eye area includes j areas, and the j areas correspond to the j colors of the n colors, and j is an integer greater than or equal to 1 and less than n
  • the method further includes: extracting the Y value, CR value, and CB value of each color of the j-type colors, wherein the Y value represents the brightness of each color and the CR value represents the The difference between the red component and the brightness of the color, the CB value represents the difference between the blue component and the brightness of each color; according to the Y value, the CR value, the CB value, it is determined that the type of dark circles included in the area corresponding to each type of color is pigmentary dark circles or vascular dark circles, wherein the vascular dark circles include red eyes, dark cyan dark circles, and light cyan dark circles , Light red dark circles or blue circles.
  • the method further includes: if the color of the middle region of the dark circle of interest region is darker than the color of the dark circle, the color of the middle region is darker than the colors around it, and If the intermediate area is a non-discrete area, then the dark circle area includes a structured dark circle.
  • the method further includes: if the dark circles area includes at least two types of dark circles in vascular dark circles, pigmented dark circles, or structural dark circles, the black The dark circles in the eye circle area are mixed.
  • an apparatus for detecting and evaluating dark circles includes: an acquisition module for acquiring an image to be processed; a processing module for extracting a dark circle of interest region in the image to be processed; The processing module is used to perform color clustering on the black eye area of interest to obtain n types of colors in the black eye area of interest, where n is a positive integer; Similar colors to identify the black eye area in the black eye area of interest; the processing module is used to obtain a black eye evaluation result based on the black eye area.
  • the device in the embodiment of the present application extract the black eye area of interest in the image to be processed, perform color clustering on the black eye area of interest to obtain n types of colors, and accurately identify the n types of colors
  • the dark circle area in the dark circle area of interest is obtained, and the dark circle evaluation result is obtained based on the dark circle area, so that the dark circle area in the image can be effectively detected and evaluated.
  • the processing module is specifically configured to: determine a dark circle color and a reference skin color according to the n-type colors; determine the dark circle based on the black circle color and the reference skin color The dark circles in the area of interest.
  • the processing module is specifically configured to: determine the first color of the n-type colors as the dark-eye color in the dark-eye area of interest; The second color of is determined as the reference skin color in the black eye area of interest.
  • the first color is the darkest color among the n-type colors after removing noise
  • the second color is the lightest color among the n-type colors after removing bright light.
  • the processing module is specifically configured to: if the first pixel in the dark circle of interest area satisfies:
  • the first pixel belongs to the dark circle area, where C is the first pixel, CD is the color of the dark circle, CS is the reference skin color, and T1 is the relative of the first pixel The contrast between the color of the dark circles and the color of the reference skin color.
  • the first pixel in the black eye area of interest satisfies:
  • C is the first pixel
  • CS is the reference skin color
  • T2 represents the lowest contrast of the two colors that can be distinguished by the human eye.
  • the reference skin color satisfies:
  • CS is the reference skin color
  • CS belongs to the i- th color among the n - type colors
  • CS i-1 is the i-1- th color among the n-type colors
  • i is A positive integer greater than or equal to 1 and less than or equal to n
  • T3 represents the minimum value of the color in the bright area
  • T4 represents the difference between the i-th color and the i-1th color.
  • the processing module is specifically configured to: remove the eyelash area in the image to be processed; extract the black eye area of interest in the image to be processed with the eyelash area removed.
  • the processing module is specifically configured to: extract features of the dark circle area, the features include contrast of the dark circle area, area of the dark circle area, or the dark circle area At least one of the variances of the regions; according to the characteristics, the severity of dark circles is evaluated by a pattern recognition method.
  • the processing module is specifically configured to determine the type of dark circles according to the position of the darkest region in the n-type colors and/or the color of the dark circles region.
  • the black eye area includes j areas, and the j areas correspond to the j colors of the n colors, and j is an integer greater than or equal to 1 and less than n
  • the processing module is also used to: extract the Y value, CR value, and CB value of each color in the j-type colors, where the Y value represents the brightness of each color and the CR value represents The difference between the red component and the brightness of each type of color, and the CB value represents the difference between the blue component and the brightness of each type of color; according to the Y value, the CR value,
  • the CB value determines that the type of dark circles included in the area corresponding to each type of color is pigmentary dark circles or vascular dark circles, wherein the vascular dark circles include red eyes, dark cyan dark circles, and light cyan Dark circles, light red dark circles or blue circles.
  • the dark circles area includes structured dark circles.
  • the dark circles area includes at least two types of dark circles in vascular type dark circles, pigmented dark circles or structural dark circles, the dark circles in the dark circles area It is a hybrid dark eye.
  • Each module included in the device in the second aspect may be implemented in software and/or hardware.
  • each module included in the apparatus in the second aspect may be implemented by a processor, that is, the apparatus in the second aspect may include a processor, which is used to execute program instructions to implement each module that can be implemented by each module included in the apparatus Features.
  • the device in the second aspect may include a memory for storing program instructions executed by the processor, and even for storing various data.
  • the device in the second aspect may be a chip that can be integrated in a smart device.
  • the device may further include a communication interface.
  • the present application provides a computer-readable storage medium.
  • the computer-readable storage medium stores program codes executed by a device for detecting and evaluating dark circles.
  • the program code includes instructions for performing the method in the first aspect or any one of the possible implementation manners.
  • the present application provides a computer program product containing instructions.
  • the computer program product runs on the device for detecting and evaluating dark circles, the device is caused to perform the method in the first aspect or any one of the possible implementation manners.
  • extract the black eye area of interest in the to-be-processed image perform color clustering on the black eye area of interest to obtain n types of colors, and accurately identify according to the n types of colors
  • the dark circle area in the dark circle area of interest is obtained, and the dark circle evaluation result is obtained based on the dark circle area, so that the dark circle area in the image can be effectively detected and evaluated.
  • FIG. 1 is a schematic flowchart of a method for detecting and evaluating dark circles in an embodiment of the present application.
  • FIG. 2 is a schematic block diagram of a display interface according to an embodiment of the present application.
  • FIG. 3 is a schematic block diagram of clustering processing according to an embodiment of the present application.
  • FIG. 4 is a schematic block diagram of removing bright light according to an embodiment of the present application.
  • FIG. 5 is a schematic block diagram of segmenting a black eye area according to an embodiment of the present application.
  • FIG. 6 is a schematic block diagram of segmenting a black eye area according to another embodiment of the present application.
  • FIG. 7 is a schematic block diagram of a display interface according to an embodiment of the present application.
  • FIG. 8 is a schematic block diagram of a type of dark circles according to an embodiment of the present application.
  • FIG. 9 is a schematic structural diagram of an apparatus for detecting and evaluating dark circles according to an embodiment of the present application.
  • the technical solutions of the embodiments of the present application may be applied to various terminal devices capable of performing image processing.
  • the terminal devices may specifically be cameras, smart phones, or other terminal devices or devices capable of performing image processing. limited.
  • the terminal device or the network device includes a hardware layer, an operating system layer running on the hardware layer, and an application layer running on the operating system layer.
  • the hardware layer includes central processing unit (CPU), memory management unit (memory management unit, MMU), and memory (also called main memory) and other hardware.
  • the operating system may be any one or more computer operating systems that implement business processes through processes, for example, a Linux operating system, a Unix operating system, an Android operating system, an iOS operating system, or a windows operating system.
  • the application layer includes browser, address book, word processing software, instant messaging software and other applications.
  • the embodiment of the present application does not specifically limit the specific structure of the execution body of the method provided in the embodiment of the present application, as long as it can run the program that records the code of the method provided by the embodiment of the present application to provide according to the embodiment of the present application
  • the method may be used for communication.
  • the execution body of the method provided in the embodiments of the present application may be a terminal device or a network device, or a functional module in the terminal device or network device that can call a program and execute the program.
  • the term "article of manufacture” as used in this application encompasses a computer program accessible from any computer-readable device, carrier, or medium.
  • the computer-readable medium may include, but is not limited to: magnetic storage devices (for example, hard disks, floppy disks, or magnetic tapes, etc.), optical disks (for example, compact discs (CDs), digital universal discs (digital discs, DVDs)) Etc.), smart cards and flash memory devices (for example, erasable programmable read-only memory (EPROM), cards, sticks or key drives, etc.).
  • various storage media described herein may represent one or more devices and/or other machine-readable media for storing information.
  • machine-readable medium may include, but is not limited to, wireless channels and various other media capable of storing, containing, and/or carrying instructions and/or data.
  • FIG. 1 is a schematic flowchart of a method 100 for detecting and evaluating dark circles in an embodiment of the present application. It should be understood that FIG. 1 shows steps or operations of the communication method, but these steps or operations are merely examples, and other operations or variations of the operations in FIG. 1 may be performed in the embodiments of the present application, or not all steps are Need to perform, or these steps can be performed in other order.
  • the image to be processed should include the eye area of the user.
  • the image to be processed may be a facial image of the user.
  • the terminal device for performing the method 100 for detecting and evaluating dark circles may be a terminal device with a camera function such as a camera or a smart phone.
  • the user can use the terminal device to take pictures to obtain the image to be processed.
  • the terminal device may use facial recognition technology to identify facial feature points in the image to be processed; according to the position of the facial feature points, extract the sense of dark circles in the image to be processed Region of interest (ROI).
  • ROI Region of interest
  • the black eye circle region of interest may be extracted according to key eye feature points included in the facial feature points.
  • the dark eye area of interest may include all dark eye areas of the lower eyelid and the reference skin color area.
  • the eyelash area in the image to be processed can be removed.
  • different skin colors may be corresponding to different thresholds; the threshold corresponding to the skin color in the dark circle of interest region is determined, and then the eyelashes are removed according to the threshold.
  • the black eye area of interest in the to-be-processed image from which the eyelash area has been removed may be extracted.
  • the image to be processed may be filtered to remove noise in the image to be processed.
  • a specific filtering method reference may be made to the prior art, and details are not described in the embodiments of the present application.
  • S130 Perform color clustering on the dark circle of interest region to obtain n types of colors in the dark circle of interest region, where n is a positive integer.
  • the n-type colors obtained after clustering may be as shown in FIG. 3, that is, from top to bottom, in order from the first type of color to the nth type of color.
  • the n-type colors obtained after clustering may also be from the bottom to the top, in order from the first type of color to the n-type color.
  • the n-type colors may be distributed from the first type of color to the n-type color, from dark to light.
  • the pixels included in the i-th color among the n-type colors have the same or similar colors, that is, the difference of the gray values of different pixels in the i-th color meets a certain threshold range, i is a positive integer greater than or equal to 1 and less than or equal to n.
  • a k-means clustering algorithm or a fuzzy c-means algorithm may be used to perform color clustering on the black eye area of interest.
  • the application examples are not limited to this.
  • S140 Identify the black eye area within the black eye area of interest according to the n-type colors.
  • the dark circle color and the reference skin color can be determined according to the n-type colors.
  • the first color of the n-type colors may be determined as the color of the dark circles in the dark circle of interest region. It should be understood that the first color may be the darkest color in the region of interest of the dark circles after removing noise, where the noise may refer to areas that are darker than the dark circles such as spots, moles, or eyelashes.
  • the first color may be the darkest color among the n types of colors.
  • the darkest region in the upper left corner of the image to be processed may be used as the starting position of the dark circles.
  • the darkest part of the color needs to have a certain number of pixels to prevent the influence of a few noises.
  • the second color of the n-type colors may be determined as the reference skin color in the dark circle of interest area. It should be understood that the second color may be the lightest color in the dark eye area of interest after removing bright light. Generally, the bright light may refer to a lighter color than the reference skin color.
  • the second color may be the lightest color among the n types of colors.
  • the reference skin color needs to satisfy:
  • CS is the reference skin color
  • CS belongs to the i- th color among the n - type colors
  • CS i-1 is the i-1- th color among the n-type colors
  • i is A positive integer greater than or equal to 1 and less than or equal to n
  • T3 represents the minimum value of the color in the bright area
  • T4 represents the difference between the i-th color and the i-1th color.
  • the T4 may be the maximum difference between the i-th color and the i-1th color.
  • the bright light can be removed by the following methods:
  • the difference of the cluster colors is greater than T4, indicating that the lightest color is currently bright light, then remove the bright light (ie, the lightest color currently), at this time, take the lightest current in the black eye area of interest after removing the bright light Color, and then judge whether the current lightest color is bright (iterate until the current lightest color is not bright); if the lightest color does not satisfy greater than or equal to T3, or does not satisfy the
  • the difference between the lightest color and the adjacent cluster color is greater than T4, indicating that the current lightest color is the reference skin color (that is, the current lightest color is not bright, and the iteration ends), then the removal of bright light ends.
  • the reference skin color needs to satisfy:
  • the light needs to satisfy:
  • C bright light is the color of bright light
  • C bright light belongs to the i-th color among the n-type colors.
  • the left image in FIG. 4 is the black eye area of interest without removing bright light
  • the right image in FIG. 4 is the black eye area of interest after removing bright light using the above formula. It can be seen that the above formula can be used to remove the bright light area, which can prevent the bright light or low light area from being recognized as the reference skin color area due to the lighter color, thereby improving the accuracy of dark circle recognition.
  • T3 and T4 may be preset.
  • T3 may represent the minimum value that bright light in the image to be processed may take. That is, T3 may be the darkest color in the bright area, and the reference skin color is darker than the bright area, that is, CS ⁇ T3 needs to be satisfied.
  • the black eye area in the black eye area of interest may be determined according to the black eye color and the reference skin color.
  • the lowest contrast threshold it may be determined whether the first pixel in the dark circle of interest region is a pixel that can be distinguished by the human eye.
  • the first pixel in the dark circle of interest area is a pixel that can be distinguished by the human eye, and the first pixel may belong to the dark circle area; or, the first pixel may belong to the reference skin color area.
  • C is the first pixel
  • CS is the reference skin color
  • T2 represents the lowest contrast of the two colors that can be distinguished by the human eye.
  • T2 may be preset.
  • the first pixel in the dark circle of interest region is a pixel that cannot be distinguished by the human eye, and the first pixel may belong to the reference skin color region.
  • C is the first pixel
  • CS is the reference skin color
  • T2 represents the lowest contrast of the two colors that can be distinguished by the human eye.
  • threshold judgment may be performed on the first pixel in the dark circle of interest region to determine whether the first pixel belongs to the dark circle or the reference skin color region.
  • the first pixel belongs to the dark circle area, where C is the first pixel, CD is the color of the dark circle, CS is the reference skin color, and T1 is the relative of the first pixel The contrast between the color of the dark circles and the color of the reference skin color.
  • T1 may be preset.
  • the color of the first pixel is closer to the color of the dark circle area.
  • the first pixel belongs to the reference skin color area, where C is the first pixel, CD is the dark circle color, CS is the reference skin color, and T1 is the relative of the first pixel The contrast between the color of the dark circles and the color of the reference skin color.
  • the left image in FIG. 5 is the black eye area of interest
  • the right image in FIG. 5 is the black eye area divided by the method in the embodiment of the present application. It can be seen that the dark circles area can be accurately identified by the method in the embodiment of the present application.
  • dark circles can be divided into structural dark circles, pigmented dark circles and vascular dark circles.
  • the structure-type dark circles are generally formed by the growth of eye bags, and their positions usually appear in the middle of the dark circle of interest area. Therefore, the structure of dark circles can be judged by the position.
  • the middle area of the black eye area of interest is the black eye area.
  • the black eye circle region of interest includes a structure type black eye circle.
  • the black eye area of interest includes a structured black eye area
  • the left image in FIG. 6 is the black eye area of interest
  • the right image in FIG. 6 is segmented using the method in the embodiment of the present application Out of the dark circles area. It can be seen that when the dark circles are structural dark circles, the dark circles area can also be accurately identified by the method in the embodiment of the present application.
  • the feature of the dark circle area may be extracted, and the characteristic includes at least one of the contrast of the dark circle area, the area of the dark circle area, or the variance of the dark circle area; According to the characteristics, the severity of dark circles can be evaluated by a pattern recognition method.
  • the evaluation result of the dark circles may be obtained based on the dark circles area and the reference skin color area.
  • the characteristics of the reference skin color area may be extracted, and according to the characteristics of the dark eye area and the reference skin color area, the severity of the dark eye area may be evaluated by a pattern recognition method.
  • the characteristics of the reference skin color area include at least one of the contrast of the reference skin color area, the area of the reference skin color area, or the variance of the reference skin color area.
  • the pattern recognition method may be linear regression (linear regression) or support vector machine (support vector machine, SVM) regression, etc., which is not limited in the embodiments of the present application.
  • a pattern recognition method may be used to obtain a score indicating the severity of dark circles.
  • a dark circle score map can be established, and the dark circle maps of different gray values corresponding to the corresponding scores, for example, the score interval indicating the severity of the dark circle can be 60-100 points, the dark circle map and the score interval correspond.
  • the terminal device may output score information of dark circles. As shown in FIG. 7, the terminal device may output the score information on its display interface. It should be understood that the display interface in FIG. 7 is only an example and not a limitation.
  • the type of dark circles is determined according to the position of the darkest region in the n-type colors and/or the color of the dark circles region.
  • the dark circles are pigment-type dark circles; if the pixels in the dark circle area are reddish or bluish, it can be It is determined that the dark circles are vascular dark circles.
  • the dark circle area may include j areas, and the j areas are in one-to-one correspondence with the color j of the n colors, and j is an integer greater than or equal to 1 and smaller than n.
  • the Y value, CR value, and CB value of each type of color in the j types of colors may be extracted, where the Y value indicates the brightness of each color and the CR value indicates The difference between the red component of each color and the brightness, the CB value represents the difference between the blue component of each color and the brightness.
  • the type of dark circles included in the area corresponding to each type of color is pigment type dark circles or vascular type Eye circles, wherein the blood vessel type dark circles include red eyes, dark cyan dark circles, light cyan dark circles, light red dark circles or blue circles.
  • the dark circle area includes a structured dark circle.
  • the dark circles area includes at least two types of dark circles in vascular dark circles, pigmented dark circles, or structural dark circles
  • the dark circles in the dark circles area are hybrid dark circles .
  • the terminal device may output type information of dark circles. As shown in FIG. 7, the terminal device may output the type information on its display interface. It should be understood that the display interface in FIG. 7 is only an example and not a limitation.
  • the types of the above-mentioned dark circles are shown in FIG. 8, the shaded area in the upper diagram in FIG. 8 is the recognized vascular type dark circles, the shaded area in the middle diagram in FIG. 8 is the identified pigmented dark circles, in FIG. 8
  • the picture below shows the structure of dark circles.
  • FIG. 9 is a schematic block diagram of an apparatus 900 for detecting and evaluating dark circles under an embodiment of the present application. It should be understood that the device 900 for detecting and evaluating dark circles is only an example. The device in the embodiment of the present application may further include other modules or units, or include modules with similar functions to the modules in FIG. 9, or not necessarily include all modules in FIG. 9.
  • the obtaining module 910 is used to obtain an image to be processed
  • the processing module 920 is configured to extract a black eye region of interest in the image to be processed
  • the processing module 920 is configured to perform color clustering on the dark circle of interest region to obtain n types of colors in the dark circle of interest region, where n is a positive integer;
  • the processing module 920 is configured to identify a black eye area within the black eye area of interest according to the n-type colors
  • the processing module 920 is configured to obtain a dark circle evaluation result based on the dark circle area.
  • the processing module 920 is specifically configured to: determine a dark circle color and a reference skin color according to the n-type colors; and determine a Dark circles area.
  • the processing module 920 is specifically configured to: determine the first color in the n-type colors as the black eye color in the black-eye area of interest; and determine the second color in the n-type colors Determined as the reference skin color in the dark circle of interest area.
  • the first color is the darkest color among the n-type colors after removing noise
  • the second color is the lightest color among the n-type colors after removing bright light.
  • the processing module 920 is specifically configured to: if the first pixel in the dark circle of interest area satisfies:
  • the first pixel belongs to the dark circle area, where C is the first pixel, CD is the color of the dark circle, CS is the reference skin color, and T1 is the relative of the first pixel The contrast between the color of the dark circles and the color of the reference skin color.
  • the first pixel in the dark circle of interest area satisfies:
  • C is the first pixel
  • CS is the reference skin color
  • T2 represents the lowest contrast of the two colors that can be distinguished by the human eye.
  • the reference skin color meets:
  • CS is the reference skin color
  • CS belongs to the i- th color among the n - type colors
  • CS i-1 is the i-1- th color among the n-type colors
  • i is A positive integer greater than or equal to 1 and less than or equal to n
  • T3 represents the minimum value of the color in the bright area
  • T4 represents the difference between the i-th color and the i-1th color.
  • the processing module 920 is specifically configured to: remove the eyelash area in the image to be processed; and extract the black eye area of interest in the image to be processed with the eyelash area removed.
  • the processing module 920 is specifically configured to: extract features of the dark circle area, the features include contrast of the dark circle area, area of the dark circle area, or variance of the dark circle area At least one of; according to the characteristics, the pattern recognition method to evaluate the severity of dark circles.
  • the processing module 920 is specifically configured to determine the type of dark circles according to the position of the darkest area in the n-type colors and/or the color of the dark circles area.
  • the dark circle area includes j areas, and the j areas are in one-to-one correspondence with the j colors of the n colors, j is an integer greater than or equal to 1 and less than n, and the processing module 920 is also used to: extract the Y value, CR value, and CB value of each type of color in the j types of colors, where the Y value represents the brightness of each color and the CR value represents each color The difference between the red component and the brightness of the color, the CB value represents the difference between the blue component of each color and the brightness; according to the Y value, the CR value, the CB value of each color , Determine that the type of dark circles included in the area corresponding to each type of color is pigmentary dark circles or vascular dark circles, wherein the vascular dark circles include red circles, dark cyan dark circles, light cyan dark circles, light circles Red dark circles or blue circles.
  • the dark circle area includes a structured dark circle.
  • the dark circles area includes at least two types of dark circles in vascular dark circles, pigmented dark circles, or structural dark circles
  • the dark circles in the dark circles area are hybrid dark circles .
  • the processor in the embodiments of the present application may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSPs), and application specific integrated circuits (application specific integrated circuit, ASIC), ready-made programmable gate array (field programmable gate array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
  • the memory in the embodiments of the present application may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.
  • the non-volatile memory can be read-only memory (read-only memory, ROM), programmable read-only memory (programmable ROM, PROM), erasable programmable read-only memory (erasable PROM, EPROM), electronically Erasable programmable read-only memory (electrically EPROM, EEPROM) or flash memory.
  • the volatile memory may be a random access memory (random access memory, RAM), which is used as an external cache.
  • random access memory random access memory
  • static random access memory static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access Access memory
  • SDRAM double data rate synchronous dynamic random access memory
  • double data Srate double data Srate
  • DDR SDRAM enhanced synchronous dynamic random access memory
  • ESDRAM synchronous connection dynamic random access memory Take memory (synchlink DRAM, SLDRAM) and direct memory bus random access memory (direct rambus RAM, DR RAM).
  • the above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination.
  • the above-described embodiments may be implemented in whole or in part in the form of computer program products.
  • the computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions according to the embodiments of the present application are generated in whole or in part.
  • the computer may be a general-purpose computer, a dedicated computer, a computer network, or other programmable devices.
  • the computer instructions may be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be from a website site, computer, server or data center Transmit to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.).
  • the computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more collections of available media.
  • the usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium.
  • the semiconductor medium may be a solid state drive.
  • the disclosed system, device, and method may be implemented in other ways.
  • the device embodiments described above are only schematic.
  • the division of the units is only a division of logical functions.
  • there may be other divisions for example, multiple units or components may be combined or Can be integrated 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, devices or units, and may be in electrical, mechanical 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 may be distributed on multiple 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.
  • each functional 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 functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.
  • the technical solution of the present application essentially or part of the contribution to the existing technology 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 are used to enable a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (random access memory, RAM), magnetic disk or optical disk and other media that can store program codes .

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • General Physics & Mathematics (AREA)
  • Animal Behavior & Ethology (AREA)
  • Biomedical Technology (AREA)
  • Surgery (AREA)
  • Medical Informatics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Molecular Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Human Computer Interaction (AREA)
  • Dermatology (AREA)
  • Dentistry (AREA)
  • Ophthalmology & Optometry (AREA)
  • Image Analysis (AREA)

Abstract

一种检测和评价黑眼圈的方法和装置，该方法包括：获取待处理图像（S110）；提取所述待处理图像中的黑眼圈感兴趣区域（S120）；对所述黑眼圈感兴趣区域进行颜色聚类，以得到所述黑眼圈感兴趣区域中的n类颜色，n为正整数（S130）；根据所述n类颜色，识别所述黑眼圈感兴趣区域内的黑眼圈区域（S140）；基于所述黑眼圈区域获取黑眼圈评价结果（S150）。该方法能够有效地检测和评价图像中的黑眼圈区域。

Description

检测和评价黑眼圈的方法和装置 技术领域
本申请实施例涉及图像处理领域，并且更具体地，涉及一种检测和评价黑眼圈的方法和装置。
背景技术
随着科学技术的提升和人们生活质量的提高，面部皮肤质量作为人体生理健康状况的一面镜子，越来越受到人们的重视。
一般来说，面部皮肤质量可以通过细纹、毛孔、黑头、红区、色斑、肌龄、法令纹、黑眼圈八大维度进行评价。在用户对面部皮肤的关心程度中，黑眼圈这一项的关注度排名靠前。尤其对于女性用户而言，黑眼圈一直是其非常关注的面部皮肤问题。
因此，如何对黑眼圈进行有效地检测和评价成为一个亟需解决的问题。
发明内容
本申请提供一种检测和评价黑眼圈的方法和装置，能够有效地检测和评价图像中的黑眼圈区域。
第一方面，提供了一种检测和评价黑眼圈的方法，该方法包括：获取待处理图像；提取所述待处理图像中的黑眼圈感兴趣区域；对所述黑眼圈感兴趣区域进行颜色聚类，以得到所述黑眼圈感兴趣区域中的n类颜色，n为正整数；根据所述n类颜色，识别所述黑眼圈感兴趣区域内的黑眼圈区域；基于所述黑眼圈区域获取黑眼圈评价结果。
根据本申请实施例中的方法，提取所述待处理图像中的黑眼圈感兴趣区域，对所述黑眼圈感兴趣区域进行颜色聚类得到n类颜色，根据所述n类颜色，精确地识别出所述黑眼圈感兴趣区域内的黑眼圈区域，并基于所述黑眼圈区域获取黑眼圈评价结果，从而能够有效地检测和评价图像中的黑眼圈区域。
在一种可能的实现方式中，所述根据所述n类颜色，识别所述黑眼圈感兴趣区域内的黑眼圈区域，包括：根据所述n类颜色确定黑眼圈颜色和基准肤色颜色；根据所述黑眼圈颜色和所述基准肤色颜色，确定所述黑眼圈兴趣区域中的黑眼圈区域。
在一种可能的实现方式中，将所述n类颜色中的第一颜色确定为所述黑眼圈感兴趣区域中的黑眼圈颜色；
将所述n类颜色中的第二颜色确定为所述黑眼圈感兴趣区域中的基准肤色颜色。
在一种可能的实现方式中，所述第一颜色为所述n类颜色中去除噪声后最深的颜色，所述第二颜色为所述n类颜色中去除亮光后最浅的颜色。
在一种可能的实现方式中，所述根据所述黑眼圈颜色和所述基准肤色颜色，确定所述黑眼圈兴趣区域中的黑眼圈区域，包括：若所述黑眼圈感兴趣区域中的第一像素点满足：
Figure PCTCN2019118635-appb-000001
则所述第一像素点属于所述黑眼圈区域，其中，C为所述第一像素点，CD为所述黑眼圈颜色，CS为所述基准肤色颜色，T1为所述第一像素点相对所述黑眼圈颜色与所述基准肤色颜色的对比度。
在一种可能的实现方式中，所述黑眼圈感兴趣区域中的第一像素点满足：
CS-C＞＝T2
其中，C为所述第一像素点，CS为所述基准肤色颜色，T2表示人眼能够区分的两种颜色的最低对比度。
在一种可能的实现方式中，所述基准肤色颜色满足：
CS＜T3或CS-CS i-1＜＝T4
中至少一项，其中，CS为所述基准肤色颜色，CS属于所述n类颜色中的第i类颜色，CS i-1为所述n类颜色中的第i-1类颜色，i为大于或等于1且小于或等于n的正整数，T3表示亮光区域颜色的最小值，T4表示所述第i类颜色与所述第i-1类颜色的差值。
在一种可能的实现方式中，所述提取所述待处理图像中的黑眼圈感兴趣区域，包括：去除所述待处理图像中的睫毛区域；提取已去除睫毛区域的所述待处理图像中的所述黑眼圈感兴趣区域。
在一种可能的实现方式中，所述基于所述黑眼圈区域获取黑眼圈评价结果，包括：提取所述黑眼圈区域的特征，所述特征包括所述黑眼圈区域的对比度、所述黑眼圈区域的面积或所述黑眼圈区域的方差中的至少一项；根据所述特征，通过模式识别方法评价黑眼圈的严重程度。
在一种可能的实现方式中，所述基于所述黑眼圈区域获取黑眼圈评价结果，包括：根据所述n类颜色中颜色最深区域的位置和/或所述黑眼圈区域的颜色，确定黑眼圈的类型。
在一种可能的实现方式中，所述黑眼圈区域包括j个区域，所述j个区域与所述n类颜色中的j类颜色一一对应，j为大于或等于1且小于n的整数，所述方法还包括：提取所述j类颜色中每类颜色的Y值、CR值、CB值，其中，所述Y值表示所述每类颜色的亮度、所述CR值表示所述每类颜色的红色分量与亮度的差值，所述CB值表示所述每类颜色的蓝色分量与亮度的差值；根据所述每类颜色的所述Y值、所述CR值、所述CB值，确定所述每类颜色对应的区域包括的黑眼圈的类型为色素型黑眼圈或血管型黑眼圈，其中，所述血管型黑眼圈包括红眼圈、深青色黑眼圈、浅青色黑眼圈、浅红色黑眼圈或蓝眼圈。
在一种可能的实现方式中，所述方法还包括：若所述黑眼圈感兴趣区域的中间区域的颜色深于所述黑眼圈颜色，所述中间区域的颜色深于其周围的颜色，且所述中间区域为非离散区域，则所述黑眼圈区域包括结构型黑眼圈。
在一种可能的实现方式中，所述方法还包括：若所述黑眼圈区域包括血管型黑眼圈、色素型黑眼圈或结构型黑眼圈中的至少两种类型的黑眼圈，则所述黑眼圈区域内的黑眼圈为混合型黑眼圈。
第二方面，提供了一种检测和评价黑眼圈的装置，该装置包括：获取模块，用于获 取待处理图像；处理模块，用于提取所述待处理图像中的黑眼圈感兴趣区域；所述处理模块，用于对所述黑眼圈感兴趣区域进行颜色聚类，以得到所述黑眼圈感兴趣区域中的n类颜色，n为正整数；所述处理模块，用于根据所述n类颜色，识别所述黑眼圈感兴趣区域内的黑眼圈区域；所述处理模块，用于基于所述黑眼圈区域获取黑眼圈评价结果。
根据本申请实施例中的装置，提取所述待处理图像中的黑眼圈感兴趣区域，对所述黑眼圈感兴趣区域进行颜色聚类得到n类颜色，根据所述n类颜色，精确地识别出所述黑眼圈感兴趣区域内的黑眼圈区域，并基于所述黑眼圈区域获取黑眼圈评价结果，从而能够有效地检测和评价图像中的黑眼圈区域。
在一种可能的实现方式中，所述处理模块具体用于：根据所述n类颜色确定黑眼圈颜色和基准肤色颜色；根据所述黑眼圈颜色和所述基准肤色颜色，确定所述黑眼圈兴趣区域中的黑眼圈区域。
在一种可能的实现方式中，所述处理模块具体用于：将所述n类颜色中的第一颜色确定为所述黑眼圈感兴趣区域中的黑眼圈颜色；将所述n类颜色中的第二颜色确定为所述黑眼圈感兴趣区域中的基准肤色颜色。
在一种可能的实现方式中，所述第一颜色为所述n类颜色中去除噪声后最深的颜色，所述第二颜色为所述n类颜色中去除亮光后最浅的颜色。
在一种可能的实现方式中，所述处理模块具体用于：若所述黑眼圈感兴趣区域中的第一像素点满足：
Figure PCTCN2019118635-appb-000002
则所述第一像素点属于所述黑眼圈区域，其中，C为所述第一像素点，CD为所述黑眼圈颜色，CS为所述基准肤色颜色，T1为所述第一像素点相对所述黑眼圈颜色与所述基准肤色颜色的对比度。
在一种可能的实现方式中，所述黑眼圈感兴趣区域中的第一像素点满足：
CS-C＞＝T2
其中，C为所述第一像素点，CS为所述基准肤色颜色，T2表示人眼能够区分的两种颜色的最低对比度。
在一种可能的实现方式中，所述基准肤色颜色满足：
CS＜T3或CS-CS i-1＜＝T4
中至少一项，其中，CS为所述基准肤色颜色，CS属于所述n类颜色中的第i类颜色，CS i-1为所述n类颜色中的第i-1类颜色，i为大于或等于1且小于或等于n的正整数，T3表示亮光区域颜色的最小值，T4表示所述第i类颜色与所述第i-1类颜色的差值。
在一种可能的实现方式中，所述处理模块具体用于：去除所述待处理图像中的睫毛区域；提取已去除睫毛区域的所述待处理图像中的所述黑眼圈感兴趣区域。
在一种可能的实现方式中，所述处理模块具体用于：提取所述黑眼圈区域的特征，所述特征包括所述黑眼圈区域的对比度、所述黑眼圈区域的面积或所述黑眼圈区域的方差中的至少一项；根据所述特征，通过模式识别方法评价黑眼圈的严重程度。
在一种可能的实现方式中，所述处理模块具体用于：根据所述n类颜色中颜色最深区域的位置和/或所述黑眼圈区域的颜色，确定黑眼圈的类型。
在一种可能的实现方式中，所述黑眼圈区域包括j个区域，所述j个区域与所述n类颜色中的j类颜色一一对应，j为大于或等于1且小于n的整数，所述处理模块还用于：提取所述j类颜色中每类颜色的Y值、CR值、CB值，其中，所述Y值表示所述每类颜色的亮度、所述CR值表示所述每类颜色的红色分量与亮度的差值，所述CB值表示所述每类颜色的蓝色分量与亮度的差值；根据所述每类颜色的所述Y值、所述CR值、所述CB值，确定所述每类颜色对应的区域包括的黑眼圈的类型为色素型黑眼圈或血管型黑眼圈，其中，所述血管型黑眼圈包括红眼圈、深青色黑眼圈、浅青色黑眼圈、浅红色黑眼圈或蓝眼圈。
在一种可能的实现方式中，若所述黑眼圈感兴趣区域的中间区域的颜色深于所述黑眼圈颜色，所述中间区域的颜色深于其周围的颜色，且所述中间区域为非离散区域，则所述黑眼圈区域包括结构型黑眼圈。
在一种可能的实现方式中，若所述黑眼圈区域包括血管型黑眼圈、色素型黑眼圈或结构型黑眼圈中的至少两种类型的黑眼圈，则所述黑眼圈区域内的黑眼圈为混合型黑眼圈。
第二方面中的装置包括的各个模块可以通过软件和/或硬件方式实现。
例如，第二方面中的装置包括的各个模块可以通过处理器实现，即第二方面中的装置可以包括处理器，该处理器用于执行程序指令，以实现该装置包括的各个模块能够实现的各个功能。
可选地，第二方面中的装置开可以包括存储器，用于存储处理器执行的程序指令，甚至用于存储各种数据。
可选地，第二方面中的装置可以是能够集成在智能设备中的芯片，此时，该装置还可以包括通信接口。
第三方面，本申请提供了一种计算机可读存储介质。该计算机可读存储介质中存储用于检测和评价黑眼圈的装置执行的程序代码。该程序代码包括用于执行第一方面或其中任意一种可能的实现方式中的方法的指令。
第四方面，本申请提供了一种包含指令的计算机程序产品。当该计算机程序产品在检测和评价黑眼圈的装置上运行时，使得该装置执行第一方面或其中任意一种可能的实现方式中的方法。
根据本申请实施例中的方法，提取所述待处理图像中的黑眼圈感兴趣区域，对所述黑眼圈感兴趣区域进行颜色聚类得到n类颜色，根据所述n类颜色，精确地识别出所述黑眼圈感兴趣区域内的黑眼圈区域，并基于所述黑眼圈区域获取黑眼圈评价结果，从而能够有效地检测和评价图像中的黑眼圈区域。
附图说明
图1是本申请实施例的检测和评价黑眼圈的方法的示意性流程图。
图2是本申请一个实施例的显示界面的示意性框图。
图3是本申请实施例的聚类处理的示意性框图。
图4是本申请一个实施例的去除亮光的示意性框图。
图5是本申请一个实施例的分割黑眼圈区域的示意性框图。
图6是本申请另一个实施例的分割黑眼圈区域的示意性框图。
图7是本申请一个实施例的显示界面的示意性框图。
图8是本申请一个实施例的黑眼圈类型的示意性框图。
图9是本申请一个实施例的检测和评价黑眼圈的装置的示意性结构图。
具体实施方式
下面将结合附图，对本申请中的技术方案进行描述。
本申请实施例的技术方案可以应用于各种能够进行图像处理的终端设备，该终端设备具体可以为照相机、智能手机或者其他能够进行图像处理的终端设备或装置，本申请实施例对此并不限定。
在本申请实施例中，终端设备或网络设备包括硬件层、运行在硬件层之上的操作***层，以及运行在操作***层上的应用层。该硬件层包括中央处理器(central processing unit，CPU)、内存管理单元(memory management unit，MMU)和内存(也称为主存)等硬件。该操作***可以是任意一种或多种通过进程(process)实现业务处理的计算机操作***，例如，Linux操作***、Unix操作***、Android操作***、iOS操作***或windows操作***等。该应用层包含浏览器、通讯录、文字处理软件、即时通信软件等应用。并且，本申请实施例并未对本申请实施例提供的方法的执行主体的具体结构特别限定，只要能够通过运行记录有本申请实施例的提供的方法的代码的程序，以根据本申请实施例提供的方法进行通信即可，例如，本申请实施例提供的方法的执行主体可以是终端设备或网络设备，或者，是终端设备或网络设备中能够调用程序并执行程序的功能模块。
另外，本申请的各个方面或特征可以实现成方法、装置或使用标准编程和/或工程技术的制品。本申请中使用的术语“制品”涵盖可从任何计算机可读器件、载体或介质访问的计算机程序。例如，计算机可读介质可以包括，但不限于:磁存储器件(例如，硬盘、软盘或磁带等)，光盘(例如，压缩盘(compact disc，CD)、数字通用盘(digital versatile disc，DVD)等)，智能卡和闪存器件(例如，可擦写可编程只读存储器(erasable programmable read-only memory，EPROM)、卡、棒或钥匙驱动器等)。另外，本文描述的各种存储介质可代表用于存储信息的一个或多个设备和/或其它机器可读介质。术语“机器可读介质”可包括但不限于，无线信道和能够存储、包含和/或承载指令和/或数据的各种其它介质。
图1是本申请实施例的检测和评价黑眼圈的方法100的一个示意性流程图。应理解，图1示出了通信方法的步骤或操作，但这些步骤或操作仅是示例，本申请实施例还可以执行其他操作或者图1中的各个操作的变形，或者，并不是所有步骤都需要执行，或者，这些步骤可以按照其他顺序执行。
S110，获取待处理图像。
其中，所述待处理图像中应该包括用户的眼部区域，例如，如图2所示，所述待处理图像可以为用户的面部图像。
在本申请实施例中，用于执行所述检测和评价黑眼圈的方法100的终端设备可以为照相机或智能手机等具有拍照功能的终端设备。
可选地，用户可以使用所述终端设备进行拍摄，以获取所述待处理图像。
S120，提取所述待处理图像中的黑眼圈感兴趣区域。
在本申请实施例中，所述终端设备可以使用人脸识别技术识别出所述待处理图像中的面部特征点；根据所述面部特征点的位置，提取出所述待处理图像中黑眼圈感兴趣区域(region of interest，ROI)。具体的人脸识别方法可以参考现有技术，本申请实施例中不再赘述。
可选地，可以根据所述面部特征点包括的眼部关键特征点，提取所述黑眼圈感兴趣区域。所述黑眼圈感兴趣区域可以包括下眼睑全部黑眼圈区域及基准肤色区域。
在本申请实施例中，可以去除所述待处理图像中的睫毛区域。
可选地，可以将不同的肤色对应不同的阈值；确定所述黑眼圈感兴趣区域中的肤色对应的阈值，进而根据所述阈值去除眼睫毛。
可选地，可以提取已去除睫毛区域的所述待处理图像中的所述黑眼圈感兴趣区域。
在本申请实施例中，可以对所述待处理图像进行滤波，以去除所述待处理图像中的噪声。具体的滤波方法可以参考现有技术，本申请实施例中不再赘述。
S130，对所述黑眼圈感兴趣区域进行颜色聚类，以得到所述黑眼圈感兴趣区域中的n类颜色，n为正整数。
在本申请实施例中，聚类后得到的所述n类颜色可以如图3所示，即自上而下，依次为第1类颜色至第n类颜色。可选地，聚类后得到的所述n类颜色也可以自下而上，依次为第1类颜色至第n类颜色。
可选地，所述n类颜色可以从第1类颜色至第n类颜色，由深到浅层次分布。其中，所述n类颜色中第i类颜色中包括的像素点的颜色相同或相近，也就是说，所述第i类颜色中不同像素点的灰度值的差值满足一定的阈值范围，i为大于或等于1且小于或等于n的正整数。
在本申请实施例中，可以使用k-均值聚类(k-means clustering)算法或模糊c-均值聚类(fuzzy c-means)算法等对所述黑眼圈感兴趣区域进行颜色聚类，本申请实施例对此并不限定。
S140，根据所述n类颜色，识别所述黑眼圈感兴趣区域内的黑眼圈区域。
在本申请实施例中，可以根据所述n类颜色确定黑眼圈颜色和基准肤色颜色。
在本申请实施例中，可以将所述n类颜色中的第一颜色确定为所述黑眼圈感兴趣区域中的黑眼圈颜色。应理解，所述第一颜色可以为所述黑眼圈感兴趣区域中去除噪声后最深的颜色，其中，所述噪声可以指斑、痣或睫毛等比黑眼圈深的区域。
可选地，所述第一颜色可以为所述n类颜色中最深的颜色。
例如，在对所述黑眼圈感兴趣区域进行颜色聚类后，可以将待处理图像中左上角颜色最深区域作为黑眼圈的起始位置。其中，所述颜色最深部分需要具有一定像素数，以防止少数噪声的影响。
在本申请实施例中，可以将所述n类颜色中的第二颜色确定为所述黑眼圈感兴趣区域中的基准肤色颜色。应理解，所述第二颜色可以为所述黑眼圈感兴趣区域中去除亮光后最浅的颜色，通常，所述亮光可以指比所述基准肤色颜色浅的区域。
可选地，所述第二颜色可以为所述n类颜色中最浅的颜色。
在一种可能的实现方式中，为了避免亮光或微光区域内的像素点被确定为基准肤色颜色，所述基准肤色颜色需要满足：
CS＜T3或CS-CS i-1＜＝T4
中至少一项，其中，CS为所述基准肤色颜色，CS属于所述n类颜色中的第i类颜色，CS i-1为所述n类颜色中的第i-1类颜色，i为大于或等于1且小于或等于n的正整数，T3表示亮光区域颜色的最小值，T4表示所述第i类颜色与所述第i-1类颜色的差值。可选地，所述T4可以为所述第i类颜色与所述第i-1类颜色的最大差值。
在本申请实施例中，可以通过下述方法去除亮光：
取所述黑眼圈感兴趣区域中最浅的颜色，判断所述最浅的颜色是否为亮光，若所述最浅的颜色满足大于或等于T3，且同时满足所述最浅的颜色与相邻聚类颜色的差值大于T4，说明当前最浅的颜色为亮光，则去除该亮光(即当前最浅的颜色)，此时，取去除该亮光后所述黑眼圈感兴趣区域中当前最浅的颜色，再判断所述当前最浅的颜色是否为亮光(一直迭代，直到当前最浅的颜色不为亮光为止)；若所述最浅的颜色不满足大于或等于T3，或不满足所述最浅的颜色与相邻聚类颜色的差值大于T4，说明当前最浅的颜色为基准肤色颜色(即当前最浅的颜色不为亮光，此时迭代结束)，则去除亮光结束。
也就是说，所述基准肤色颜色需要满足：
CS＜T3或CS-CS i-1＜＝T4
中至少一项。
换句话说，亮光需要同时满足：
C 亮光＞＝T3且C 亮光-CS i-1＞T4
其中，C 亮光为亮光的颜色，C 亮光属于所述n类颜色中的第i类颜色。如图4所示，图4中左图为未去除亮光的所述黑眼圈感兴趣区域，图4中右图为利用上述公式去除亮光后的所述黑眼圈感兴趣区域。可以看出，利用上述公式能够很好的去除亮光区域，可以避免亮光或微光区域由于颜色较浅而被识别为基准肤色区域，从而提高黑眼圈识别的准确性。
可选地，T3、T4可以是预先设置的。
可选地，T3可以表示所述待处理图像中亮光可能取到的最小值。也就是说，T3可以为亮光区域的最深颜色，所述基准肤色颜色比所述亮光区域深，即需要满足CS＜T3。
在本申请实施例中，可以根据所述黑眼圈颜色和所述基准肤色颜色，确定所述黑眼圈兴趣区域中的黑眼圈区域。
可选地，可以根据最低对比度阈值，确定所述黑眼圈感兴趣区域中的第一像素点是否为人眼能够区分的像素点。
例如，若所述黑眼圈感兴趣区域中的第一像素点满足：
CS-C＞＝T2
则所述黑眼圈感兴趣区域中的第一像素点为人眼能够区分的像素点，所述第一像素可以属于所述黑眼圈区域；或者，所述第一像素点可以属于基准肤色区域。其中，C为所述第一像素点，CS为所述基准肤色颜色，T2表示人眼能够区分的两种颜色的最低对比度。
可选地，T2可以是预先设置的。
再例如，若所述黑眼圈感兴趣区域中的第一像素点不满足：
CS-C＞＝T2
则所述黑眼圈感兴趣区域中的第一像素点为人眼无法区分的像素点，所述第一像素点可以属于基准肤色区域。其中，C为所述第一像素点，CS为所述基准肤色颜色，T2表示人眼能够区分的两种颜色的最低对比度。
在本申请实施例中，可以对所述黑眼圈感兴趣区域中的第一像素点进行阈值判断，以确定所述第一像素点属于所述黑眼圈区域或所述基准肤色区域。
例如，若所述黑眼圈感兴趣区域中的第一像素点满足：
Figure PCTCN2019118635-appb-000003
则所述第一像素点属于所述黑眼圈区域，其中，C为所述第一像素点，CD为所述黑眼圈颜色，CS为所述基准肤色颜色，T1为所述第一像素点相对所述黑眼圈颜色与所述基准肤色颜色的对比度。
可选地，T1可以是预先设置的。
也就是说，当所述第一像素点满足
Figure PCTCN2019118635-appb-000004
时，所述第一像素点的颜色更接近黑眼圈区域的颜色。
再例如，若所述黑眼圈感兴趣区域中的第一像素点不满足：
Figure PCTCN2019118635-appb-000005
则所述第一像素点属于所述基准肤色区域，其中，C为所述第一像素点，CD为所述黑眼圈颜色，CS为所述基准肤色颜色，T1为所述第一像素点相对所述黑眼圈颜色与所述基准肤色颜色的对比度。
如图5所示，图5中左图为所述黑眼圈感兴趣区域，图5中右图为利用本申请实施例中的方法分割出的所述黑眼圈区域。可以看出，通过本申请实施例中的方法可以精确地识别出黑眼圈区域。
一般来说，黑眼圈可以分为结构型黑眼圈、色素型黑眼圈和血管型黑眼圈等。
其中，结构型黑眼圈一般是由于年龄增长造成的眼袋形成，其位置通常出现在所述黑眼圈感兴趣区域的中间位置。因此，对于结构型黑眼圈可以通过位置进行判断。
可选地，若所述黑眼圈感兴趣区域中间区域的颜色深于所述黑眼圈颜色，所述黑眼圈感兴趣区域中间区域的颜色深于其周围肤色一定阈值，且所述中间区域为非离散区域，则所述黑眼圈感兴趣区域的中间区域为黑眼圈区域。此时，也可以认为，所述黑眼圈感兴趣区域中间区域为黑眼圈的起始位置。相应地，可以确定所述黑眼圈感兴趣区域中包括结构型黑眼圈。
当所述黑眼圈感兴趣区域包括结构型黑眼圈时，如图6所示，图6中左图为所述黑眼圈感兴趣区域，图6中右图为利用本申请实施例中的方法分割出的所述黑眼圈区域。可以看出，当所述黑眼圈为结构型黑眼圈时，通过本申请实施例中的方法也可以精确地识别出黑眼圈区域。
S150，基于所述黑眼圈区域获取黑眼圈评价结果。
在本申请实施例中，可以提取所述黑眼圈区域的特征，所述特征包括所述黑眼圈区 域的对比度、所述黑眼圈区域的面积或所述黑眼圈区域的方差中的至少一项；可以根据所述特征，通过模式识别(pattern recognition)方法评价黑眼圈的严重程度。
在本申请实施例中，可以基于所述黑眼圈区域和所述基准肤色区域获取黑眼圈评价结果。
可选地，可以提取所述基准肤色区域的特征，根据所述黑眼圈区域的特征和所述基准肤色区域的特征，通过模式识别方法评价黑眼圈的严重程度。其中，所述基准肤色区域的特征包括所述基准肤色区域的对比度、所述基准肤色区域的面积或所述基准肤色区域的方差中的至少一项。
可选地，所述模式识别方法可以为线性回归(linear regression)或支持向量机(support vector machine，SVM)回归等，本申请实施例对此并不限定。
可选地，可以通过模式识别方法，得到表示黑眼圈严重程度的得分。
可选地，可以建立黑眼圈得分图谱，将不同灰度值的黑眼圈图谱对应相应的得分，例如，表示黑眼圈严重程度的得分区间可以为60-100分，将黑眼圈图谱与该得分区间对应。
可选地，所述终端设备可以输出黑眼圈的得分信息。如图7所示，所述终端设备可以在其显示界面输出所述得分信息。应理解，图7中的显示界面仅为示例而非限定。
在本申请实施例中，根据所述n类颜色中颜色最深区域的位置和/或所述黑眼圈区域的颜色，确定黑眼圈的类型。
可选地，若所述黑眼圈区域中的像素点偏黑、偏褐，则可以确定该黑眼圈为色素型黑眼圈；若所述黑眼圈区域中的像素点偏红、偏青，则可以确定该黑眼圈为血管型黑眼圈。
在本申请实施例中，所述黑眼圈区域可以包括j个区域，所述j个区域与所述n类颜色中的j类颜色一一对应，j为大于或等于1且小于n的整数。
例如，在颜色空间YCRCB中，可以提取所述j类颜色中每类颜色的Y值、CR值、CB值，其中，所述Y值表示所述每类颜色的亮度、所述CR值表示所述每类颜色的红色分量与亮度的差值，所述CB值表示所述每类颜色的蓝色分量与亮度的差值。
此时，可以根据所述每类颜色的所述Y值、所述CR值、所述CB值，确定所述每类颜色对应的区域包括的黑眼圈的类型为色素型黑眼圈或血管型黑眼圈，其中，所述血管型黑眼圈包括红眼圈、深青色黑眼圈、浅青色黑眼圈、浅红色黑眼圈或蓝眼圈。
可选地，若所述黑眼圈感兴趣区域的中间区域的颜色深于所述黑眼圈颜色，所述中间区域的颜色深于其周围的颜色，且所述中间区域为非离散区域，则所述黑眼圈区域包括结构型黑眼圈。
可选地，若所述黑眼圈区域包括血管型黑眼圈、色素型黑眼圈或结构型黑眼圈中的至少两种类型的黑眼圈，则所述黑眼圈区域内的黑眼圈为混合型黑眼圈。
在本申请实施例中，所述终端设备可以输出黑眼圈的类型信息。如图7所示，所述终端设备可以在其显示界面输出所述类型信息。应理解，图7中的显示界面仅为示例而非限定。上述黑眼圈的类型如图8所示，图8中的上图的阴影区域为识别出的血管型黑眼圈，图8中的中图的阴影区域为识别出的色素型黑眼圈，图8中的下图为结构型黑眼圈。
图9是本申请实施例的检测和评价黑眼圈的装置900的示意性框图。应理解，检测和评价黑眼圈的装置900仅是一种示例。本申请实施例的装置还可以包括其他模块或单元，或者包括与图9中的各个模块的功能相似的模块，或者并非要包括图9中的所有模块。
获取模块910，用于获取待处理图像；
处理模块920，用于提取所述待处理图像中的黑眼圈感兴趣区域；
所述处理模块920，用于对所述黑眼圈感兴趣区域进行颜色聚类，以得到所述黑眼圈感兴趣区域中的n类颜色，n为正整数；
所述处理模块920，用于根据所述n类颜色，识别所述黑眼圈感兴趣区域内的黑眼圈区域；
所述处理模块920，用于基于所述黑眼圈区域获取黑眼圈评价结果。
可选地，所述处理模块920具体用于：根据所述n类颜色确定黑眼圈颜色和基准肤色颜色；根据所述黑眼圈颜色和所述基准肤色颜色，确定所述黑眼圈兴趣区域中的黑眼圈区域。
可选地，所述处理模块920具体用于：将所述n类颜色中的第一颜色确定为所述黑眼圈感兴趣区域中的黑眼圈颜色；将所述n类颜色中的第二颜色确定为所述黑眼圈感兴趣区域中的基准肤色颜色。
可选地，所述第一颜色为所述n类颜色中去除噪声后最深的颜色，所述第二颜色为所述n类颜色中去除亮光后最浅的颜色。
可选地，所述处理模块920具体用于：若所述黑眼圈感兴趣区域中的第一像素点满足：
Figure PCTCN2019118635-appb-000006
则所述第一像素点属于所述黑眼圈区域，其中，C为所述第一像素点，CD为所述黑眼圈颜色，CS为所述基准肤色颜色，T1为所述第一像素点相对所述黑眼圈颜色与所述基准肤色颜色的对比度。
可选地，所述黑眼圈感兴趣区域中的第一像素点满足：
CS-C＞＝T2
其中，C为所述第一像素点，CS为所述基准肤色颜色，T2表示人眼能够区分的两种颜色的最低对比度。
可选地，所述基准肤色颜色满足：
CS＜T3或CS-CS i-1＜＝T4
中至少一项，其中，CS为所述基准肤色颜色，CS属于所述n类颜色中的第i类颜色，CS i-1为所述n类颜色中的第i-1类颜色，i为大于或等于1且小于或等于n的正整数，T3表示亮光区域颜色的最小值，T4表示所述第i类颜色与所述第i-1类颜色的差值。
可选地，所述处理模块920具体用于：去除所述待处理图像中的睫毛区域；提取已去除睫毛区域的所述待处理图像中的所述黑眼圈感兴趣区域。
可选地，所述处理模块920具体用于：提取所述黑眼圈区域的特征，所述特征包括所述黑眼圈区域的对比度、所述黑眼圈区域的面积或所述黑眼圈区域的方差中的至少一 项；根据所述特征，通过模式识别方法评价黑眼圈的严重程度。
可选地，所述处理模块920具体用于：根据所述n类颜色中颜色最深区域的位置和/或所述黑眼圈区域的颜色，确定黑眼圈的类型。
可选地，所述黑眼圈区域包括j个区域，所述j个区域与所述n类颜色中的j类颜色一一对应，j为大于或等于1且小于n的整数，所述处理模块920还用于：提取所述j类颜色中每类颜色的Y值、CR值、CB值，其中，所述Y值表示所述每类颜色的亮度、所述CR值表示所述每类颜色的红色分量与亮度的差值，所述CB值表示所述每类颜色的蓝色分量与亮度的差值；根据所述每类颜色的所述Y值、所述CR值、所述CB值，确定所述每类颜色对应的区域包括的黑眼圈的类型为色素型黑眼圈或血管型黑眼圈，其中，所述血管型黑眼圈包括红眼圈、深青色黑眼圈、浅青色黑眼圈、浅红色黑眼圈或蓝眼圈。
可选地，若所述黑眼圈感兴趣区域的中间区域的颜色深于所述黑眼圈颜色，所述中间区域的颜色深于其周围的颜色，且所述中间区域为非离散区域，则所述黑眼圈区域包括结构型黑眼圈。
可选地，若所述黑眼圈区域包括血管型黑眼圈、色素型黑眼圈或结构型黑眼圈中的至少两种类型的黑眼圈，则所述黑眼圈区域内的黑眼圈为混合型黑眼圈。
应理解，本申请实施例中的处理器可以为中央处理单元(central processing unit，CPU)，该处理器还可以是其他通用处理器、数字信号处理器(digital signal processor，DSP)、专用集成电路(application specific integrated circuit，ASIC)、现成可编程门阵列(field programmable gate array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
还应理解，本申请实施例中的存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的随机存取存储器(random access memory，RAM)可用，例如静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synchlink DRAM，SLDRAM)和直接内存总线随机存取存储器(direct rambus RAM，DR RAM)。
上述实施例，可以全部或部分地通过软件、硬件、固件或其他任意组合来实现。当使用软件实现时，上述实施例可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令或计算机程序。在计算机上加载或执行所述计算机指令或计算机程序时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以为通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个 计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集合的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质。半导体介质可以是固态硬盘。
应理解，本文中术语“和/或”，仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本文中字符“/”，一般表示前后关联对象是一种“或”的关系。
应理解，在本申请的各种实施例中，上述各过程的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的***、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的***、装置和方法，可以通过其它的方式实现。例如，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，仅仅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***，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，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
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网络单元上。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。
所述功能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(可以是个人计算机，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、移动硬盘、只读存储器(read-only memory，ROM)、随机存取存储器(random access memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (26)

  1. 一种检测和评价黑眼圈的方法，其特征在于，包括：
    获取待处理图像；
    提取所述待处理图像中的黑眼圈感兴趣区域；
    对所述黑眼圈感兴趣区域进行颜色聚类，以得到所述黑眼圈感兴趣区域中的n类颜色，n为正整数；
    根据所述n类颜色，识别所述黑眼圈感兴趣区域内的黑眼圈区域；
    基于所述黑眼圈区域获取黑眼圈评价结果。
  2. 根据权利要求1所述的方法，其特征在于，所述根据所述n类颜色，识别所述黑眼圈感兴趣区域内的黑眼圈区域，包括：
    根据所述n类颜色确定黑眼圈颜色和基准肤色颜色；
    根据所述黑眼圈颜色和所述基准肤色颜色，确定所述黑眼圈兴趣区域中的黑眼圈区域。
  3. 根据权利要求2所述的方法，其特征在于，所述根据所述n类颜色确定黑眼圈颜色和基准肤色颜色，包括：
    将所述n类颜色中的第一颜色确定为所述黑眼圈感兴趣区域中的黑眼圈颜色；
    将所述n类颜色中的第二颜色确定为所述黑眼圈感兴趣区域中的基准肤色颜色。
  4. 根据权利要求3所述的方法，其特征在于，所述第一颜色为所述n类颜色中去除噪声后最深的颜色，所述第二颜色为所述n类颜色中去除亮光后最浅的颜色。
  5. 根据权利要求2至4中任一项所述的方法，其特征在于，所述根据所述黑眼圈颜色和所述基准肤色颜色，确定所述黑眼圈兴趣区域中的黑眼圈区域，包括：
    若所述黑眼圈感兴趣区域中的第一像素点满足：
    Figure PCTCN2019118635-appb-100001
    则所述第一像素点属于所述黑眼圈区域，其中，C为所述第一像素点，CD为所述黑眼圈颜色，CS为所述基准肤色颜色，T1为所述第一像素点相对所述黑眼圈颜色与所述基准肤色颜色的对比度。
  6. 根据权利要求2至4中任一项所述的方法，其特征在于，所述黑眼圈感兴趣区域中的第一像素点满足：
    CS-C＞＝T2
    其中，C为所述第一像素点，CS为所述基准肤色颜色，T2表示人眼能够区分的两种颜色的最低对比度。
  7. 根据权利要求2至6中任一项所述的方法，其特征在于，所述基准肤色颜色满足：
    CS＜T3或CS-CS i-1＜＝T4
    中至少一项，其中，CS为所述基准肤色颜色，CS属于所述n类颜色中的第i类颜色，CS i-1为所述n类颜色中的第i-1类颜色，i为大于或等于1且小于或等于n的正整数，T3表示亮光区域颜色的最小值，T4表示所述第i类颜色与所述第i-1类颜色的差值。
  8. 根据权利要求1至7中任一项所述的方法，其特征在于，所述提取所述待处理图像中的黑眼圈感兴趣区域，包括：
    去除所述待处理图像中的睫毛区域；
    提取已去除睫毛区域的所述待处理图像中的所述黑眼圈感兴趣区域。
  9. 根据权利要求1至8中任一项所述的方法，其特征在于，所述基于所述黑眼圈区域获取黑眼圈评价结果，包括：
    提取所述黑眼圈区域的特征，所述特征包括所述黑眼圈区域的对比度、所述黑眼圈区域的面积或所述黑眼圈区域的方差中的至少一项；
    根据所述特征，通过模式识别方法评价黑眼圈的严重程度。
  10. 根据权利要求1至9中任一项所述的方法，其特征在于，所述黑眼圈区域包括j个区域，所述j个区域与所述n类颜色中的j类颜色一一对应，j为大于或等于1且小于n的整数，所述方法还包括：
    提取所述j类颜色中每类颜色的Y值、CR值、CB值，其中，所述Y值表示所述每类颜色的亮度、所述CR值表示所述每类颜色的红色分量与亮度的差值，所述CB值表示所述每类颜色的蓝色分量与亮度的差值；
    根据所述每类颜色的所述Y值、所述CR值、所述CB值，确定所述每类颜色对应的区域包括的黑眼圈的类型为色素型黑眼圈或血管型黑眼圈，其中，所述血管型黑眼圈包括红眼圈、深青色黑眼圈、浅青色黑眼圈、浅红色黑眼圈或蓝眼圈。
  11. 根据权利要求10所述的方法，其特征在于，所述方法还包括：
    若所述黑眼圈感兴趣区域的中间区域的颜色深于所述黑眼圈颜色，所述中间区域的颜色深于其周围的颜色，且所述中间区域为非离散区域，则所述黑眼圈区域包括结构型黑眼圈。
  12. 根据权利要求10或11所述的方法，其特征在于，所述方法还包括：
    若所述黑眼圈区域包括血管型黑眼圈、色素型黑眼圈或结构型黑眼圈中的至少两种类型的黑眼圈，则所述黑眼圈区域内的黑眼圈为混合型黑眼圈。
  13. 一种检测和评价黑眼圈的装置，其特征在于，包括：
    获取模块，用于获取待处理图像；
    处理模块，用于提取所述待处理图像中的黑眼圈感兴趣区域；
    所述处理模块，用于对所述黑眼圈感兴趣区域进行颜色聚类，以得到所述黑眼圈感兴趣区域中的n类颜色，n为正整数；
    所述处理模块，用于根据所述n类颜色，识别所述黑眼圈感兴趣区域内的黑眼圈区域；
    所述处理模块，用于基于所述黑眼圈区域获取黑眼圈评价结果。
  14. 根据权利要求13所述的装置，其特征在于，所述处理模块具体用于：
    根据所述n类颜色确定黑眼圈颜色和基准肤色颜色；
    根据所述黑眼圈颜色和所述基准肤色颜色，确定所述黑眼圈兴趣区域中的黑眼圈区域。
  15. 根据权利要求14所述的装置，其特征在于，所述处理模块具体用于：
    将所述n类颜色中的第一颜色确定为所述黑眼圈感兴趣区域中的黑眼圈颜色；
    将所述n类颜色中的第二颜色确定为所述黑眼圈感兴趣区域中的基准肤色颜色。
  16. 根据权利要求15所述的装置，其特征在于，所述第一颜色为所述n类颜色中去除噪声后最深的颜色，所述第二颜色为所述n类颜色中去除亮光后最浅的颜色。
  17. 根据权利要求14至16中任一项所述的装置，其特征在于，所述处理模块具体用于：
    若所述黑眼圈感兴趣区域中的第一像素点满足：
    Figure PCTCN2019118635-appb-100002
    则所述第一像素点属于所述黑眼圈区域，其中，C为所述第一像素点，CD为所述黑眼圈颜色，CS为所述基准肤色颜色，T1为所述第一像素点相对所述黑眼圈颜色与所述基准肤色颜色的对比度。
  18. 根据权利要求14至16中任一项所述的装置，其特征在于，所述黑眼圈感兴趣区域中的第一像素点满足：
    CS-C＞＝T2
    其中，C为所述第一像素点，CS为所述基准肤色颜色，T2表示人眼能够区分的两种颜色的最低对比度。
  19. 根据权利要求14至18中任一项所述的装置，其特征在于，所述基准肤色颜色满足：
    CS＜T3或CS-CS i-1＜＝T4
    中至少一项，其中，CS为所述基准肤色颜色，CS属于所述n类颜色中的第i类颜色，CS i-1为所述n类颜色中的第i-1类颜色，i为大于或等于1且小于或等于n的正整数，T3表示亮光区域颜色的最小值，T4表示所述第i类颜色与所述第i-1类颜色的差值。
  20. 根据权利要求13至19中任一项所述的装置，其特征在于，所述处理模块具体用于：
    去除所述待处理图像中的睫毛区域；
    提取已去除睫毛区域的所述待处理图像中的所述黑眼圈感兴趣区域。
  21. 根据权利要求13至20中任一项所述的装置，其特征在于，所述处理模块具体用于：
    提取所述黑眼圈区域的特征，所述特征包括所述黑眼圈区域的对比度、所述黑眼圈区域的面积或所述黑眼圈区域的方差中的至少一项；
    根据所述特征，通过模式识别方法评价黑眼圈的严重程度。
  22. 根据权利要求13至21中任一项所述的装置，其特征在于，所述黑眼圈区域包括j个区域，所述j个区域与所述n类颜色中的j类颜色一一对应，j为大于或等于1且小于n的整数，所述处理模块还用于：提取所述j类颜色中每类颜色的Y值、CR值、CB值，其中，所述Y值表示所述每类颜色的亮度、所述CR值表示所述每类颜色的红色分量与亮度的差值，所述CB值表示所述每类颜色的蓝色分量与亮度的差值；根据所述每类颜色的所述Y值、所述CR值、所述CB值，确定所述每类颜色对应的区域包括的黑眼圈的类型为色素型黑眼圈或血管型黑眼圈，其中，所述血管型黑眼圈包括红眼圈、深青色黑眼圈、浅青色黑眼圈、浅红色黑眼圈或蓝眼圈。
  23. 根据权利要求22所述的装置，其特征在于，若所述黑眼圈感兴趣区域的中间区域的颜色深于所述黑眼圈颜色，所述中间区域的颜色深于其周围的颜色，且所述中间区域为非离散区域，则所述黑眼圈区域包括结构型黑眼圈。
  24. 根据权利要求22或23所述的装置，其特征在于，若所述黑眼圈区域包括血管型黑眼圈、色素型黑眼圈或结构型黑眼圈中的至少两种类型的黑眼圈，则所述黑眼圈区域内的黑眼圈为混合型黑眼圈。
  25. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质中存储用于检测和评价黑眼圈的装置执行的程序代码，所述程序代码包括用于执行权利要求1至12中任一项所述的方法的指令。
  26. 一种计算机程序产品，其特征在于，所述计算机程序产品包括用于执行权利要求1至12中任一项所述的方法的指令。
PCT/CN2019/118635 2018-11-29 2019-11-15 检测和评价黑眼圈的方法和装置 WO2020108322A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US17/297,684 US11779264B2 (en) 2018-11-29 2019-11-15 Dark circle detection and evaluation method and apparatus
EP19891460.8A EP3879437B1 (en) 2018-11-29 2019-11-15 Method and apparatus for detecting and evaluating dark circles

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811446411.8A CN111241889B (zh) 2018-11-29 2018-11-29 检测和评价黑眼圈的方法和装置
CN201811446411.8 2018-11-29

Publications (1)

Publication Number Publication Date
WO2020108322A1 true WO2020108322A1 (zh) 2020-06-04

Family

ID=70854358

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/118635 WO2020108322A1 (zh) 2018-11-29 2019-11-15 检测和评价黑眼圈的方法和装置

Country Status (4)

Country Link
US (1) US11779264B2 (zh)
EP (1) EP3879437B1 (zh)
CN (1) CN111241889B (zh)
WO (1) WO2020108322A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113128377B (zh) * 2021-04-02 2024-05-17 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
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

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160125228A1 (en) * 2014-11-04 2016-05-05 Samsung Electronics Co., Ltd. Electronic device, and method for analyzing face information in electronic device
CN107392841A (zh) * 2017-06-16 2017-11-24 广东欧珀移动通信有限公司 人脸区域中黑眼圈消除方法、装置及终端
CN107730456A (zh) * 2016-08-10 2018-02-23 卡西欧计算机株式会社 图像处理方法以及图像处理装置
CN108830184A (zh) * 2018-05-28 2018-11-16 厦门美图之家科技有限公司 黑眼圈识别方法及装置

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100354875C (zh) * 2005-09-29 2007-12-12 上海交通大学 基于人脸检测的红眼去除方法
US20150313532A1 (en) * 2007-01-05 2015-11-05 Sava Marinkovich Method and system for managing and quantifying sun exposure
TW201106919A (en) * 2009-08-21 2011-03-01 Jian-Han Chen Method for measuring and identifying dark cycles under eyes
JP6297941B2 (ja) * 2014-07-18 2018-03-20 富士フイルム株式会社 うるおい感評価装置、うるおい感評価装置の作動方法およびうるおい感評価プログラム
US9760762B2 (en) * 2014-11-03 2017-09-12 Anastasia Soare Facial structural shaping
WO2017112913A1 (en) * 2015-12-23 2017-06-29 Gauss Surgical, Inc. System and method for estimating an amount of a blood component in a volume of fluid
US20170246473A1 (en) * 2016-02-25 2017-08-31 Sava Marinkovich Method and system for managing treatments
CN105844242A (zh) * 2016-03-23 2016-08-10 湖北知本信息科技有限公司 图像中的肤色检测方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160125228A1 (en) * 2014-11-04 2016-05-05 Samsung Electronics Co., Ltd. Electronic device, and method for analyzing face information in electronic device
CN107730456A (zh) * 2016-08-10 2018-02-23 卡西欧计算机株式会社 图像处理方法以及图像处理装置
CN107392841A (zh) * 2017-06-16 2017-11-24 广东欧珀移动通信有限公司 人脸区域中黑眼圈消除方法、装置及终端
CN108830184A (zh) * 2018-05-28 2018-11-16 厦门美图之家科技有限公司 黑眼圈识别方法及装置

Also Published As

Publication number Publication date
EP3879437B1 (en) 2024-02-07
EP3879437A1 (en) 2021-09-15
EP3879437A4 (en) 2021-12-29
CN111241889A (zh) 2020-06-05
US20220019765A1 (en) 2022-01-20
US11779264B2 (en) 2023-10-10
CN111241889B (zh) 2023-05-12

Similar Documents

Publication Publication Date Title
US10504624B2 (en) System and method for segmentation and automated measurement of chronic wound images
Thanh et al. A skin lesion segmentation method for dermoscopic images based on adaptive thresholding with normalization of color models
Olugbara et al. Segmentation of melanoma skin lesion using perceptual color difference saliency with morphological analysis
US20180330522A1 (en) System and method for assessing wound
Abbas et al. A perceptually oriented method for contrast enhancement and segmentation of dermoscopy images
Mookiah et al. Automated detection of optic disk in retinal fundus images using intuitionistic fuzzy histon segmentation
CN107844736B (zh) 虹膜定位方法和装置
Thakur et al. Face detection using skin tone segmentation
WO2020108322A1 (zh) 检测和评价黑眼圈的方法和装置
Mporas et al. Color models for skin lesion classification from dermatoscopic images
CN107256543A (zh) 图像处理方法、装置、电子设备及存储介质
Oukil et al. Automatic segmentation and melanoma detection based on color and texture features in dermoscopic images
Thapar et al. The epidemiology of automatic skin cancer detection by comparative analysis of pre-processing and segmentation techniques
Mohaimin et al. Automated method for the detection and segmentation of drusen in colour fundus image for the diagnosis of age‐related macular degeneration
Chen et al. Hybrid saliency detection for images
Zhou et al. Automated detection of red lesions using superpixel multichannel multifeature
Sun et al. Optic disc segmentation by balloon snake with texture from color fundus image
Tan et al. Classification of left and right eye retinal images
Rout et al. Transition region based approach for skin lesion segmentation
LeAnder et al. Differentiation of melanoma from benign mimics using the relative‐color method
Sharma et al. A survey on classification of malignant melanoma and benign skin lesion by using machine learning techniques
Bernart et al. Segmentation and classification of melanocytic skin lesions using local and contextual features
Tabatabaie et al. Extraction of skin lesion texture features based on independent component analysis
Powar et al. Skin detection for forensic investigation
CN107403133A (zh) 确定设备和确定方法

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: 19891460

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2019891460

Country of ref document: EP

Effective date: 20210609