Search

CN-120912531-B - Scalp skin identification method and system based on image classification

CN120912531BCN 120912531 BCN120912531 BCN 120912531BCN-120912531-B

Abstract

The embodiment of the invention relates to the technical field of image detection and discloses a scalp skin identification method based on image classification, which comprises the steps of acquiring first image information through a detector under a first set condition, and identifying the first image information by adopting an image segmentation algorithm to determine the position of a hair region in the first image information; determining pixel parameters of each pixel point in the first image information after the hair image is removed, matching the pixel point parameters of each pixel point with a preset reflective pixel database, if the pixel point parameters of the corresponding pixel points are matched with the reflective pixel database, updating the number of the reflective pixel points until all the pixel points are matched, and determining a corresponding scalp moisture result according to the number of the reflective pixel points and the first image information after the hair image is removed. The quantitative evaluation of scalp skin is realized through the identification method.

Inventors

  • GAO FUCHENG

Assignees

  • 广州美莱宝美容设备有限公司

Dates

Publication Date
20260512
Application Date
20250721

Claims (8)

  1. 1. A scalp skin recognition method based on image classification, comprising: acquiring first image information by a detector under a first set condition, wherein the first image information comprises scalp images, and the first set condition is that a handle with 100 times of the first set condition is adopted for bright shooting; Identifying the first image information by adopting an image segmentation algorithm to determine the hair region position in the first image information, and removing the corresponding hair image in the first image information according to the hair region position; Determining pixel parameters of each pixel point in the first image information after the hair image is removed, matching the pixel point parameters of each pixel point with a preset reflective pixel database, and if the pixel point parameters of the corresponding pixel points are matched with the reflective pixel database, updating the number of the reflective pixel points until all the pixel points are matched; Determining a corresponding scalp moisture result according to the number of the reflective pixel points and the first image information of the hair image; The identification method further comprises the following steps: Acquiring second image information by a detector under a second set condition, wherein the second image information includes a scalp image; Identifying the second image information by adopting an image segmentation algorithm to determine the hair region position in the second image information, and removing the corresponding hair image in the second image information according to the hair region position; determining pixel point parameters of each pixel point in the second image information after the hair image is removed, matching the pixel point parameters of each pixel point with a preset sensitive pixel database, and if the pixel point parameters of the corresponding pixel points are matched with the sensitive pixel database, updating the number of the sensitive pixel points until all the pixel points are matched; if the values of the three color channels in the pixel point parameters of the corresponding pixel points are all at the edge of the data interval, determining a corresponding pixel comparison interval in the sensitive pixel database according to the scalp state, wherein the pixel comparison interval comprises a first color interval, a second color interval and a third color interval; Determining weight parameters corresponding to the first color interval, the second color interval and the third color interval; Matching pixel point parameters of each pixel point with corresponding pixel comparison intervals to determine corresponding matching scores, and updating the number of sensitive pixel points when the matching scores exceed a set value; And determining a corresponding scalp detection result according to the number of the sensitive pixel points and the second image information of the hair image.
  2. 2. The method for scalp skin identification based on image classification as claimed in claim 1, further comprising, after said until all pixel point matching is completed: converting pixel point parameters of the pixel points matched with the reflective pixel database into set pixel parameters, and carrying out pixel updating operation on corresponding pixel points in the first image information according to the set pixel parameters and the pixel point positions matched with the reflective pixel database so as to obtain an updated display effect diagram; and displaying the display effect graph.
  3. 3. The method for scalp skin identification based on image classification as claimed in claim 2, wherein said determining a corresponding scalp moisture result from the number of the light reflecting pixels and the first image information of the hair image is removed, comprises: inputting the number of the reflective pixel points and the total pixel point amount of the first image information of the hair image to a moisture calculation formula to calculate so as to obtain a corresponding moisture detection score, wherein the moisture calculation formula is as follows: S is the number of the reflective pixel points, Z is the sum of the pixels of the picture after the example is divided, Y is a background configuration normal proportion value, and C is the moisture detection fraction at the time; the displaying the display effect graph comprises the following steps: and comprehensively displaying the display effect graph and the first image information.
  4. 4. The method for scalp skin identification based on image classification as claimed in claim 1, wherein if the pixel point parameter of the corresponding pixel point is matched with the reflective pixel database, further comprising: If the values of the three color channels in the pixel point parameters of the corresponding pixel points are all at the edge of the data interval, acquiring pixel point information in a set position range, and analyzing the pixel point information in the set position range to determine corresponding gradient change information; and if the gradient change information meets the set condition, determining that the pixel point parameters of the corresponding pixel points are matched with the reflective pixel database.
  5. 5. The scalp skin identifying method based on image classification as claimed in claim 1, further comprising, after said acquiring the first image information by the detector under the first setting condition: traversing all image pixel points in the first image information through a hair recognition model, recognizing hair areas with all set lengths, and generating a rotary rectangular frame for each hair area; Extracting width parameters in the rotary rectangular frame, wherein the width parameters are the number of pixels, and determining the actual physical diameter of the corresponding hairline according to the optical multiplying power of the detector and the width parameters; If the actual physical diameter of the corresponding hairline is in the first width interval, determining that the corresponding hairline is fine hair, and updating the number of the corresponding fine hair; If the actual physical diameter of the corresponding hair is in the second width interval, determining that the corresponding hair is common hair, and updating the number of the corresponding common hair; If the actual physical diameter of the corresponding hairline is in the third width interval, determining that the corresponding hairline is thick hairline, and updating the number of the corresponding thick hairline; The corresponding hair status score is determined based on the number of fine hair, the number of normal hair, and the number of coarse hair.
  6. 6. A scalp skin recognition system based on image classification, comprising: The acquisition module is used for acquiring first image information through a detector under a first setting condition, wherein the first image information comprises scalp images; the segmentation module is used for identifying the first image information by adopting an image segmentation algorithm to determine the hair region position in the first image information, and removing the corresponding hair image in the first image information according to the hair region position; The matching module is used for determining pixel parameters of each pixel point in the first image information after the hair image is removed, matching the pixel parameters of each pixel point with a preset reflective pixel database, and updating the number of reflective pixel points until all the pixel points are matched if the pixel parameters of the corresponding pixel points are matched with the reflective pixel database; The calculation module is used for determining a corresponding scalp moisture result according to the number of the reflective pixel points and the first image information of the hair image; the identification system further comprises: Acquiring second image information by a detector under a second set condition, wherein the second image information includes a scalp image; Identifying the second image information by adopting an image segmentation algorithm to determine the hair region position in the second image information, and removing the corresponding hair image in the second image information according to the hair region position; determining pixel point parameters of each pixel point in the second image information after the hair image is removed, matching the pixel point parameters of each pixel point with a preset sensitive pixel database, and if the pixel point parameters of the corresponding pixel points are matched with the sensitive pixel database, updating the number of the sensitive pixel points until all the pixel points are matched; if the values of the three color channels in the pixel point parameters of the corresponding pixel points are all at the edge of the data interval, determining a corresponding pixel comparison interval in the sensitive pixel database according to the scalp state, wherein the pixel comparison interval comprises a first color interval, a second color interval and a third color interval; Determining weight parameters corresponding to the first color interval, the second color interval and the third color interval; Matching pixel point parameters of each pixel point with corresponding pixel comparison intervals to determine corresponding matching scores, and updating the number of sensitive pixel points when the matching scores exceed a set value; And determining a corresponding scalp detection result according to the number of the sensitive pixel points and the second image information of the hair image.
  7. 7. An electronic device comprising a memory storing executable program code, a processor coupled to the memory, the processor invoking the executable program code stored in the memory for performing the image classification based scalp skin identification method of any of claims 1 to 5.
  8. 8. A computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute the scalp skin identifying method based on image classification as claimed in any one of claims 1 to 5.

Description

Scalp skin identification method and system based on image classification Technical Field The invention relates to the technical field of image recognition, in particular to a scalp skin recognition method and system based on image classification. Background At present, scalp detection and evaluation mainly depend on manual observation and experience judgment, and the problems of inaccurate diagnosis, low efficiency, complex operation and the like exist, so that the requirements of rapid and safe scalp health management of the masses cannot be met. Traditional visual diagnosis has low reliability, and most customers rely on the objective data detected by the instrument. Existing scalp detection techniques have certain limitations in several respects. For example, a contact skin noninvasive measuring instrument can realize quantitative measurement of skin moisture content, but is limited by the contact area of a probe, and can only test skin moisture content of a very small area, so that the result cannot represent the overall skin moisture condition. For head skin, because the surface of the head skin is covered with dense hair, most of contact probes cannot be directly measured, the operation of removing local hair measurement is troublesome, the shaving process can also cause extrusion or damage to a test area, the test result is affected, the test result of the instrument is presented in a numerical value, and the skin moisture distribution cannot be visually presented. Although the partial contact type noninvasive measuring instrument adopts needle-shaped probe design aiming at scalp moisture content test to eliminate hair influence, the result is only the moisture content of the probe contact part, and cannot represent the whole scalp moisture condition and realize the visual presentation of scalp moisture. Therefore, designing a way to conveniently detect scalp skin condition is a technical problem to be solved by those skilled in the art. Disclosure of Invention Aiming at the defects, the embodiment of the invention discloses a scalp skin identification method based on image classification, which can realize accurate scalp skin quantitative evaluation and is convenient for users to intuitively know scalp moisture state. The first aspect of the embodiment of the invention discloses a scalp skin identification method based on image classification, which comprises the following steps: Acquiring first image information by a detector under a first set condition, wherein the first image information includes a scalp image; Identifying the first image information by adopting an image segmentation algorithm to determine the hair region position in the first image information, and removing the corresponding hair image in the first image information according to the hair region position; Determining pixel parameters of each pixel point in the first image information after the hair image is removed, matching the pixel point parameters of each pixel point with a preset reflective pixel database, and if the pixel point parameters of the corresponding pixel points are matched with the reflective pixel database, updating the number of the reflective pixel points until all the pixel points are matched; and determining a corresponding scalp moisture result according to the number of the reflective pixel points and the first image information of the hair image. As an optional implementation manner, in the first aspect of the embodiment of the present invention, after the matching of all the pixels is completed, the method further includes: converting pixel point parameters of the pixel points matched with the reflective pixel database into set pixel parameters, and carrying out pixel updating operation on corresponding pixel points in the first image information according to the set pixel parameters and the pixel point positions matched with the reflective pixel database so as to obtain an updated display effect diagram; and displaying the display effect graph. As an optional implementation manner, in the first aspect of the embodiment of the present invention, the determining the corresponding scalp moisture result according to the number of the reflective pixels and the first image information of the hair image includes: inputting the number of the reflective pixel points and the total pixel point amount of the first image information of the hair image to a moisture calculation formula to calculate so as to obtain a corresponding moisture detection score, wherein the moisture calculation formula is as follows: , S is the number of the reflective pixel points, Z is the sum of the pixels of the picture after the example is divided, Y is a background configuration normal proportion value, and C is the moisture detection fraction at the time; the displaying the display effect graph comprises the following steps: and comprehensively displaying the display effect graph and the first image information. In an optional implem