Method and device for providing alopecia information
According to an aspect of the present disclosure, there is provided a method of providing hair loss state information, the method comprising: obtaining a head image of a user; extracting a plurality of feature points from the head image, wherein the plurality of feature points include a first group of feature points and a second group of feature points; extracting a plurality of boundary points corresponding to boundaries of a hair and a forehead of the user from the head image; selecting a matching boundary point corresponding to the first group of feature points from among the plurality of boundary points; obtaining a top-face portion calculation value determined based on a distance between the first group of feature points and a matching boundary point corresponding to the first group of feature points with respect to a first axis on the head image; obtaining a middle-bottom-face portion calculation value determined based on a distance between the first group of feature points and the second group of feature points with respect to the first axis; and providing hair loss state information of the user based on a ratio of the top-face portion calculation value and the middle-bottom-face portion calculation value.
1 . A method of selecting images for analysis to provide hair loss diagnosis assistance information, the method comprising:
obtaining a plurality of scalp images including a first scalp image and a second scalp image;
obtaining pore region information included in each scalp image;
based on the pore region information of the first scalp image, obtaining first quantitative information related to the number of pore regions included in the first scalp image;
based on the pore region information of the second scalp image, obtaining second quantitative information related to the number of pore regions included in the second scalp image; and
selecting a target image by comparing the first quantitative information and the second quantitative information,
wherein obtaining pore region information, further comprising:
obtaining initial pore region information related to the pore region included in the scalp image; and
obtaining the pore region information by correcting the initial pore region information,
wherein the correcting the initial pore region information, further comprising:
obtaining first location information and second location information related to the pore region included in the initial pore region information;
verifying the validity of the first location information and the second location information based on whether the separation distance between the first location information and the second location information is within a predetermined distance; and
determining at least one of first location information and the second location information to be invalid if the separation distance between the first location information and the second location information is within a predetermined distance;
determining the first location information and the second location information to be valid if the separation distance between the first location information and the second location information is not within the predetermined distance;
determining a pore region corresponding to the location information determined to be valid as the pore region information when it is determined that at least one of the first location information and the second location information is invalid.
2 . The method of claim 1 ,
wherein selecting the target image, further comprising:
selecting a valid scalp image based on a result of comparing the first quantitative information and the second quantitative information;
obtaining a manually captured scalp image;
obtaining third quantitative information related to the number of pore regions included in the manually captured scalp image; and
determining the target image based on quantitative information related to the valid scalp image and the third quantitative information related to the manually captured scalp image.
3 . The method of claim 2 ,
wherein determining the target image, further comprising:
determining an image having a greater value of quantitative information as the target image by comparing the quantitative information related to the valid scalp image and the third quantitative information related to the manually captured scalp image.
4 . The method of claim 1 ,
wherein obtaining pore region information is performed by an artificial neural network configured to obtain the pore region information based on the scalp image.
5 . The method of claim 1 ,
wherein the correcting the initial pore region information, further comprising:
obtaining first information related to the pore region included in the scalp image;
obtaining second information related to the hair region included in the scalp image; and
correcting the initial pore region information based on whether the first information and the second information included in the common region range of the scalp image correspond to each other.
6 . A scalp measuring device for obtaining scalp images comprising:
a camera configured to capture a plurality of scalp images; and
an at least one processor configured to select images for analysis to provide hair loss diagnosis assistance information based on the plurality of scalp images;
wherein the at least one processor is configured to:
obtain the plurality of scalp images including a first scalp image and a second scalp image;
obtain pore region information included in each scalp image;
based on the pore region information of the first scalp image, obtain first quantitative information related to the number of pore regions included in the first scalp image;
based on the pore region information of the second scalp image, obtain second quantitative information related to the number of pore regions included in the second scalp image; and
select a target image by comparing the first quantitative information and the second quantitative information,
wherein the at least one processor is configured to obtain initial pore region information related to the pore region included in the scalp image; and obtain the pore region information by correcting the initial pore region information,
wherein the at least one processor is configured to:
obtain first location information and second location information related to the pore region included in the initial pore region information;
verify the validity of the first location information and the second location information based on whether the separation distance between the first location information and the second location information is within a predetermined distance; and
determining at least one of first location information and the second location information to be invalid if the separation distance between the first location information and the second location information is within a predetermined distance;
determining the first location information and the second location information to be valid if the separation distance between the first location information and the second location information is not within the predetermined distance;
determine a pore region corresponding to the location information determined to be valid as the pore region information when it is determined that at least one of the first location information and the second location information is invalid.
7 . The scalp measuring device of claim 6 ,
wherein the at least one processor is configured to select the target image by performing the steps below:
selecting a valid scalp image based on a result of comparing the first quantitative information and the second quantitative information;
obtaining a manually captured scalp image;
obtaining third quantitative information related to the number of pore regions included in the manually captured scalp image; and
determining the target image based on quantitative information related to the valid scalp image and the third quantitative information related to the manually captured scalp image.
8 . The scalp measuring device of claim 7 ,
wherein the at least one processor is configured to determine an image having a greater value of quantitative information as the target image by comparing the quantitative information related to the valid scalp image and the third quantitative information related to the manually captured scalp image.
9 . The scalp measuring device of claim 6 ,
wherein the at least one processor is configured to obtain pore region information based on an artificial neural network configured to obtain the pore region information based on the scalp image.
10 . The scalp measuring device of claim 6 ,
wherein the at least one processor is configured to:
obtain first information related to the pore region included in the scalp image;
obtain second information related to the hair region included in the scalp image; and
correct the initial pore region information based on whether the first information and the second information included in the common region range of the scalp image correspond to each other.