KR102562271B1 - AI-based hair loss diagnosis system and method thereof - Google Patents

Abstract

The present invention relates to an artificial intelligence-based hair loss diagnosis system and method, and relates to a user terminal that installs and executes a hair loss diagnosis application to request a hair loss diagnosis service, and when a request for a hair loss diagnosis service is received from the user terminal, AI-based hair loss diagnosis A user's hair loss condition is diagnosed using at least one of a first hair loss diagnosis mode for diagnosing hair loss based on a survey using a program and a second hair loss diagnosis mode for diagnosing hair loss based on a scalp image, and the diagnosis result is sent to the user. It includes an artificial intelligence-based hair loss diagnosis server provided to the terminal.

Description

AI-based hair loss diagnosis system and method {AI-based hair loss diagnosis system and method thereof}

The present invention relates to an artificial intelligence-based hair loss diagnosis system and method, and more particularly, to an artificial intelligence-based hair loss diagnosis system and method capable of providing a hair loss diagnosis service based on a survey or scalp image.

Hair loss refers to a state in which there is no hair in an area where hair should normally exist, and generally means that terminal hair (thick and black hair) of the scalp is lost. In the case of Koreans, who have a lower hair density than Westerners, there are about 100,000 hairs, and it is normal to lose about 50 to 100 hairs a day.

Hair loss diseases include male pattern baldness, female pattern baldness, alopecia areata, telogen effluvium, etc. with high frequency, and efforts to find the cause of telogen effluvium are important. Methods of treating male pattern baldness and female pattern baldness include drugs applied to the scalp, medications taken, and hair transplant surgery. As methods for treating alopecia areata , topical steroid preparations, systemic steroid preparations, immunotherapy, and the like are used. In telogen alopecia, the hair is likely to recover when the cause is removed, so it is important to identify and treat the cause as soon as possible.

However, in the past, in order to determine the progress of hair loss, there is a hassle of visiting a hospital, so even if hair loss progresses, the problem of preventing the hair loss state for a considerable period of time frequently occurs.

Domestic registered patent 10-2445248

In order to solve the above problems, the technical problem to be achieved by the present invention is to provide an artificial intelligence-based hair loss diagnosis system and method capable of diagnosing the user's hair loss condition without place and time restrictions upon user request and providing the diagnosis result as a service. is to present

The problems of the present invention are not limited to those mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

As a means for solving the above-described technical problem, an artificial intelligence-based hair loss diagnosis system according to an embodiment of the present invention includes a user terminal for requesting a hair loss diagnosis service by installing and executing a hair loss diagnosis application; And when a hair loss diagnosis service request is received from the user terminal, a first hair loss diagnosis mode for diagnosing hair loss based on a survey using an artificial intelligence-based hair loss diagnosis program, and a second hair loss diagnosis for diagnosing hair loss based on a scalp image An artificial intelligence-based hair loss diagnosis server for diagnosing a user's hair loss condition using at least one of the modes and providing a diagnosis result to the user terminal; may include.

The artificial intelligence-based hair loss diagnosis server, when the first hair loss diagnosis mode is selected from the user terminal and the user's response to the survey item for diagnosing hair loss is received, the user's response to the received survey item A first hair loss condition diagnosis unit for diagnosing the hair loss condition by analyzing based on artificial intelligence; may include.

The survey items may include a question for understanding the progress of the user's hair loss, selection of the user's scalp type, selection of scalp condition, selection of concerns related to hair loss, and the user's gender and age.

The first hair loss diagnosis unit calculates a hair loss progress score by applying the weight score set for each survey item to the user's response to the survey item using the weight score set for each survey item, and the calculated A first hair loss diagnosis result may be output by analyzing the hair loss progress score and the user's gender and age based on artificial intelligence.

The artificial intelligence-based hair loss diagnosis server may include a pre-processing unit that obtains a temporary scalp region by pre-processing a hair loss image obtained by photographing a hair loss progression region input from the user terminal when the second hair loss diagnosis mode is selected from the user terminal; a region correction unit located in the temporary scalp region of the preprocessed hair loss image and generating a corrected hair loss image by correcting a hair region in which loss occurs due to light reflection when a photo is taken by the user terminal; and a first hair loss condition diagnosis unit configured to analyze the corrected hair loss image to calculate a possibility of hair loss progression.

The region correction unit may analyze the temporary scalp region in the preprocessed hair loss image to set a hair region including the broken hair region, and correct the hair region by connecting the broken hair region.

The pre-processing unit extracts a temporary scalp region using a Gaussian Mixture Model (GMM), and the artificial intelligence-based hair loss diagnosis server extracts the remaining regions of the corrected hair loss image, excluding the corrected hair region, the light-reflected region, and the background region. a region separating unit separating the hair loss as a hair loss region of interest; and a scalp region re-learning unit re-learning the hair loss-of-interest region separated by the region separation unit and generating a scalp color model to be used by the GMM in the pre-processing unit.

On the other hand, the AI-based hair loss diagnosis method includes: (A) requesting a hair loss diagnosis service by the user terminal installing and executing a hair loss diagnosis application; And (B) a first hair loss diagnosis mode in which the artificial intelligence-based hair loss diagnosis server diagnoses hair loss based on a questionnaire survey using an artificial intelligence-based hair loss diagnosis program when the hair loss diagnosis service request is received from step (A), and , diagnosing the user's hair loss state by using at least one of the second hair loss diagnosis modes for diagnosing hair loss based on the scalp image, and providing the diagnosis result to the user terminal;

The step (B) may include (B1) selecting a first hair loss diagnosis mode from the user terminal and receiving a user's response to a survey item for diagnosing hair loss; (B2) diagnosing a hair loss state by analyzing a user's response to a survey item received from the user terminal based on artificial intelligence; and (B3) providing a result of diagnosing the hair loss condition generated in step (B2) (hereinafter, referred to as 'first hair loss diagnosis result') to the user terminal.

The survey items may include a question for understanding the progress of the user's hair loss, selection of the user's scalp type, selection of scalp condition, selection of concerns related to hair loss, and the user's gender and age.

In step B2, a hair loss progress score is calculated by applying the weight score set for each survey item to the user's response to the survey item using the weight score set for each survey item, and the calculated hair loss progress A first hair loss diagnosis result may be output by analyzing the score and the user's gender and age based on artificial intelligence.

The step (B) may include: (B4) obtaining a temporary scalp area by pre-processing a hair loss image obtained by photographing a hair loss progression region input from the user terminal when the second hair loss diagnosis mode is selected from the user terminal; (B5) generating a corrected hair loss image by correcting a hair region located in a temporary scalp region of the preprocessed hair loss image and in which loss occurs due to light reflection when a photo is taken by the user terminal; and (B6) calculating a possibility of progressing hair loss from the corrected hair loss image.

In the step (B5), the temporary scalp region in the preprocessed hair loss image is analyzed to set a hair region including the broken hair region, and the hair region can be corrected by connecting the broken hair region.

The step (B4) extracts a temporary scalp region using a Gaussian Mixture Model (GMM), and the step (B) includes, (B7) the corrected hair region, the light-reflected region, and the background region among the corrected hair loss images. Separating the rest of the region except for the hair loss region of interest; and (B8) generating a scalp color model to be used by the GMM in step (B4) by relearning the separated hair loss region of interest in step (B7).

According to the present invention, it is possible to diagnose and treat hair loss at an early stage by allowing a user to use a hair loss diagnosis service by using a survey method or a scalp image without directly visiting a hospital, thereby slowing down or preventing hair loss. can do.

The effects of the present invention are not limited to those mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

1 is a diagram showing an AI (Artificial Intelligent) based hair loss diagnosis system according to an embodiment of the present invention;
2 is an example of a hair loss diagnosis initial screen 200 displayed on the user terminal 10;
3A to 3E are examples of query screens 300a to 300e for grasping the progress of hair loss of a user;
4 is an exemplary diagram of a user's scalp type selection screen 400;
5 is an exemplary diagram of a user's scalp condition selection screen 500;
6 is an example of a user's trouble selection screen 600 related to hair loss;
7 is an exemplary view of an image capturing screen 700;
8 is an example of a first hair loss diagnosis result screen 800 displayed on the user terminal 10;
9 is a diagram showing a process of diagnosing hair loss according to a second hair loss diagnosis mode in a second hair loss diagnosis unit 120;
10 is an example of a second hair loss diagnosis result screen 1000 displayed on the user terminal 10;
11 is an exemplary view of a search result screen 1100;
12a to 12c are examples of screens 1200a to 1200c for providing non-face-to-face medical treatment services;
13 is a flowchart schematically illustrating an artificial intelligence-based hair loss diagnosis method according to an embodiment of the present invention;
14 is a flowchart showing steps S1310 to S1330 shown in FIG. 13 in detail, and
FIG. 15 is a flowchart showing steps S1340 to S1360 shown in FIG. 13 in detail.

The above objects, other objects, features and advantages of the present invention will be easily understood through the following preferred embodiments in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments described herein and may be embodied in other forms. Rather, the embodiments introduced herein are provided so that the disclosed content will be thorough and complete and the spirit of the present invention will be sufficiently conveyed to those skilled in the art.

In some cases, it is mentioned in advance that parts that are commonly known in describing the invention and are not greatly related to the invention are not described in order to prevent confusion for no particular reason in explaining the present invention.

In this specification, when terms such as first and second are used to describe components, these components should not be limited by these terms. These terms are only used to distinguish one component from another.

Further, when a first component is referred to as being operated or executed on (ON) a second component, the first component is operated or executed in an environment in which the second component is operated or executed, or with the second component. It should be understood that it is operated or executed through direct or indirect interaction.

When a component, device, or system is referred to as comprising a component consisting of a program or software, even if not explicitly stated otherwise, the component, device, or system refers to hardware (necessary for the program or software to execute or operate). For example, memory, CPU, etc.) or other programs or software (eg, operating system or driver required to drive hardware) should be understood as including.

In addition, it should be understood that, unless otherwise specified, the component may be implemented in any form of software, hardware, or both software and hardware.

In addition, terms used in this specification are for describing embodiments and are not intended to limit the present invention. In this specification, singular forms also include plural forms unless specifically stated otherwise in a phrase. The terms 'comprises' and/or 'comprising' used in the specification do not exclude the presence or addition of one or more other elements.

Also, in this specification, terms such as 'unit', 'module', 'server', 'system', 'platform', 'device' or 'terminal' refer to hardware and software driven by the hardware or for driving the hardware. It may be intended to refer to functional and structural combinations of For example, the hardware herein may be a data processing device including a CPU or other processor. Also, software driven by hardware may refer to a running process, an object, an executable file, a thread of execution, a program, and the like.

In addition, the terms may mean a predetermined code and a logical unit of hardware resources for executing the predetermined code, and do not necessarily mean physically connected codes or one type of hardware in the present invention. can be easily deduced to the average expert in the art.

Hereinafter, with reference to the accompanying drawings for the specific technical content to be carried out in the present invention will be described in detail.

Each component shown in FIG. 1 indicates that it can be functionally and logically separated, and does not necessarily mean that each component is classified as a separate physical device or written as a separate code. An average expert would be able to reason with ease.

1 is a diagram showing an AI (Artificial Intelligent) based hair loss diagnosis system according to an embodiment of the present invention.

Referring to FIG. 1 , the AI-based hair loss diagnosis system according to an embodiment of the present invention includes a user terminal 10 and an AI-based hair loss diagnosis server 100.

The user terminal 10 installs and executes the hair loss diagnosis application provided by the AI-based hair loss diagnosis server 100 to generate a plurality of screens necessary for the hair loss diagnosis service, and transmits information input from the user to the AI-based hair loss diagnosis server 100. ) to request hair loss diagnosis service. Alternatively, the user terminal 10 may access a web page provided by the AI-based hair loss diagnosis server 100 and use the hair loss diagnosis service. The user terminal 10 may be an electronic device capable of wired/wireless communication, such as a portable mobile phone, desktop, or laptop. In an embodiment of the present invention, a case in which the user terminal 10 requests a hair loss diagnosis service using a hair loss diagnosis application will be described as an example.

As shown in FIG. 1 , the user terminal 10 may include a first hair loss diagnosis request unit 11 , a second hair loss diagnosis request unit 12 and a diagnosis result output unit 13 .

When the first hair loss diagnosis mode menu 210 is selected, the first hair loss diagnosis request unit 11 displays a plurality of survey screens (300a to 300e, 400, 500, 600) necessary for diagnosing hair loss based on the first hair loss survey. ), and transmits the user's response input through the plurality of survey screens (300a to 300e, 400, 500, 600) to the AI-based hair loss diagnosis server 100 to request the first hair loss diagnosis service. .

When the second hair loss diagnosis mode menu 220 is selected, the second hair loss diagnosis request unit 12 creates a photographing screen 700 necessary for diagnosing hair loss based on the scalp image, and transmits the photographed hair loss image to an AI-based hair loss diagnosis server. (100) to request a second hair loss diagnosis service.

The diagnosis result output unit 13 receives the hair loss diagnosis result performed for the first hair loss diagnosis mode and the hair loss diagnosis result performed for the second hair loss diagnosis mode from the AI-based hair loss diagnosis server 100, and the user terminal 10 It is possible to generate screens 800 and 1000 that can be displayed on , and display the generated screens 800 and 1000 on the user terminal 10 .

Hereinafter, an operation of using the hair loss diagnosis service by the user terminal 10 will be described with reference to FIGS. 2 to 12 .

2 is an exemplary diagram of a hair loss diagnosis initial screen 200 displayed on the user terminal 10 .

Referring to FIG. 2 , the user terminal 10 generates and displays a hair loss diagnosis initial screen 200 by executing a hair loss diagnosis application according to a user request. The initial screen for diagnosing hair loss 200 includes a first hair loss diagnosis mode menu 210 for diagnosing hair loss based on a questionnaire survey, and a second hair loss diagnosis mode menu 220 for diagnosing hair loss based on a scalp image.

When the user selects the first hair loss diagnosis mode menu 210, the first hair loss diagnosis request unit 11 asks questions to determine the progress of the user's hair loss, selects the user's scalp type, selects the scalp condition, and issues related to hair loss. Survey items including selection and the user's gender and age may be created as shown in FIGS. 3A to 6 .

3A to 3E are examples of query screens 300a to 300e for identifying the user's hair loss progress, FIG. 4 is an example of the user's scalp type selection screen 400, and FIG. 5 is a user's scalp condition selection An exemplary view of the screen 500, FIG. 6 is an exemplary view of the user's trouble selection screen 600 related to hair loss.

Referring to FIG. 3A , the first hair loss diagnosis request unit 11 generates an inquiry screen 300a asking whether there is a feeling that the forehead or the crown of the head is widened.

Referring to FIG. 3B , the first hair loss diagnosis request unit 11 generates an inquiry screen 300b asking whether or not hair falls out more severely when washing or combing the hair.

Referring to FIG. 3C , the first hair loss diagnosis request unit 11 generates an inquiry screen 300c inquiring whether hair has become thinner than before.

Referring to FIG. 3D , the first hair loss diagnosis requesting unit 11 generates an inquiry screen 300d inquiring whether there is a person with hair loss in the family.

Referring to FIG. 3E , the first hair loss diagnosis request unit 11 may further generate an inquiry screen 300e asking whether the user has recently attempted a drastic diet or has a history of surgery.

Referring to FIG. 4 , when the user responds to the query screens 300a to 300e, the first hair loss diagnosis request unit 11 displays a scalp type selection screen 400 allowing the user to select a scalp type among oily, dry, and combination scalp types. ) to create

Referring to FIG. 5 , when the user selects a scalp type in FIG. 4 , the first hair loss diagnosis request unit 11 creates a scalp condition selection screen 500 allowing all current scalp conditions of the user to be selected.

Referring to FIG. 6 , when the user selects all of the current scalp conditions in FIG. 5 , the first hair loss diagnosis request unit 11 creates a worry selection screen 600 allowing the user to select all of the concerns related to hair loss.

In addition, the first hair loss diagnosis request unit 11 requests the user's basic information including the user's gender and age (or age group).

The first hair loss diagnosis request unit 11 may transmit the user's gender and age, and responses to survey items conducted through FIGS. 3A to 6 to the AI-based hair loss diagnosis server 100.

Meanwhile, when the user selects the second hair loss diagnosis mode menu 220 in FIG. 2 , the second hair loss diagnosis request unit 12 displays an image capture screen 700 requesting to capture the user's scalp including hair. 7 can be created. The user may photograph the scalp using the camera of the user terminal 10, and the second hair loss diagnosis request unit 12 may transmit the photographed picture to the AI hair loss diagnosis server 100 to request hair loss diagnosis.

Referring back to FIG. 1, the AI-based hair loss diagnosis server 100 first diagnoses hair loss based on a questionnaire survey using an AI-based hair loss diagnosis program when a request for hair loss diagnosis service is received from the user terminal 10. The user's hair loss condition may be diagnosed using at least one of the diagnosis mode and the second hair loss diagnosis mode for diagnosing hair loss based on the scalp image, and the diagnosis result may be provided to the user terminal 10 . The AI-based hair loss diagnosis server 100 may be an electronic device capable of wired/wireless communication such as a cloud server or a large-capacity computer.

The AI-based hair loss diagnosis server 100 according to an embodiment of the present invention includes a first hair loss diagnosis unit 110, a second hair loss diagnosis unit 120, a product recommendation unit 130, and a non-face-to-face treatment connection unit 140. can include

The first hair loss diagnosis unit 110 may include a first hair loss diagnosis unit 111 and a first hair loss information providing unit 113 .

The first hair loss diagnosis unit 111 may diagnose hair loss based on a survey using an AI-based hair loss diagnosis program. When the first hair loss diagnosis mode menu 210 is selected in the user terminal 10 and the user's response to the hair loss diagnosis survey item is received, the first hair loss condition diagnosis unit 111 determines information about the received survey item. The user's response can be analyzed based on AI to diagnose the hair loss condition.

The first hair loss diagnosis unit 111 calculates a hair loss progress score by applying the weighted score set for each survey item to the user's response to the survey item, and calculates the hair loss progress score and the user's gender and age group by AI. Based on the analysis, it is possible to generate the first hair loss diagnosis result.

For example, the first hair loss condition diagnosis unit 111 makes the five 'yes or no' selection items as shown in FIGS. 3A to 3E account for 50% of the total hair loss progress score, and the individual worry item 30% It takes up the specific gravity, and the scalp condition can be set to account for 20% of the specific gravity. Therefore, in the case of five 'yes' or 'no' choice items, a weighted score of 10 is given for each number of 'yes', and 50 points are calculated when five 'yes' are selected, and each item of personal concern is weighted at 6 points. Scores are given and 30 points are calculated when all five concerns are selected, and 20 points are calculated when all four items are selected by giving a weighted score of 5 points per scalp condition item. Thereafter, the first hair loss condition diagnosis unit 111 may diagnose the user's hair loss condition by inputting the weighted score for each item and the user's gender and age group into the AI program.

The first hair loss information providing unit 113 may provide the result of diagnosing the hair loss condition by the first hair loss condition diagnosis unit 111 (hereinafter referred to as 'first hair loss diagnosis result') to the user terminal 10. .

8 is an exemplary view of a first hair loss diagnosis result screen 800 displayed on the user terminal 10 .

Referring to FIG. 8 , the first hair loss information providing unit 113 provides information about hair loss, type of hair loss, possibility of hair loss, and detailed diagnosis based on user responses input by the user.

Referring back to FIG. 1 , the second hair loss diagnosis unit 120 may diagnose hair loss based on a hair loss image taken by a user using an AI-based hair loss diagnosis program.

To this end, the second hair loss diagnosis unit 120 includes a preprocessing unit 121, a region correcting unit 122, a second hair loss condition diagnosis unit 123, a region separating unit 124, and a scalp region re-learning unit 125 And it may include a second hair loss information providing unit (126).

9 is a view showing a process of diagnosing hair loss according to the second hair loss diagnosis mode in the second hair loss diagnosis unit 120.

Referring to FIG. 9 , after the second hair loss diagnosis mode is selected from the user terminal 10, the pre-processing unit 121 pre-processes the hair loss image 910 input from the user terminal 10 and captures the hair loss progressing region to temporarily A preprocessing image 920 including the scalp region may be acquired.

The preprocessing unit 121 may perform a preprocessing operation to remove a background from the hair loss image 910 and leave a final scalp area. To this end, the pre-processing unit 121 may convert the hair loss image 910 in RGB format into a YCbCr format and use an adaptive skin color model based on a Gaussian Mixture Model (GMM) to obtain a temporary scalp area. there is.

[Equation 1] is an equation for obtaining a temporary scalp area.

[Equation 1] is a GMM-based adaptive skin color model, where I _skin (P) is a temporary scalp area, and T _skin is a YCbCr scalp color threshold for obtaining a temporary scalp area. Taking RGB as an example, if the pixel value of T _skin is (R, G, B) = (251,206,177), the RGB value of the pixel is 241<R<255, 196<G<216, 167<B<187 respectively, the scalp determined by the pixels of the region.

is the average, is the covariance matrix, and N is the normal distribution.

(mixing coefficient) is the probability that the kth Gaussian Distribution is selected, , It can be selected as a temporary scalp area only when two conditions are satisfied.

In the case of [Equation 1], the pre-processing unit 121 If is greater than T _skin , it is determined as a pixel corresponding to the temporary scalp area and set to 1. If it is smaller than T _skin , it is set to an area other than the temporary scalp area and set to 0 to generate a binary preprocessed hair loss image 920. .

The region correction unit 122 may generate a corrected hair loss image by correcting a hair region in which loss occurs due to light reflection when a photo is taken in the user terminal 10 . Accordingly, the hair region in which loss occurs due to light reflection may be located in the temporary scalp region of the preprocessed hair loss image 920 .

In detail, the region correction unit 122 extracts a linear component (ie, a straight region) from the preprocessed hair loss image 920, determines it as a hair region, and selects a pixel close to the edge of the temporary scalp region from among the linear components as an initial point. , and the opposite of the initial point can be set as the connection point. Accordingly, the initial point may be the starting point of the hair, and the connection point may be the ending point of the broken hair. The region correction unit 122 uses an algorithm such as Hough Transform to extract the linear component, and additional algorithms such as thinning and contour extraction may be further used as a preprocessing process for this. In the embodiment, straight line components are extracted, but the area correction unit 122 may further extract curve components having a first curvature or less as hair.

As shown in the image 930 of FIG. 9 , when both initial points and connection points are determined for a plurality of straight line components by analyzing the temporary scalp region, the region correction unit 122 calculates the area between the connection points generated by connecting the connection points. It is defined as the light reflection area generated by light reflection.

When the area correction unit 122 sets the hair area including the part where the hair is broken and the light reflection area, the straight line component is checked again after one straight line component, and when the two straight line components are connected, a straight line or a curve with a first curvature or more is formed. If it is determined that the two straight line components are broken hair, the hair region can be corrected by connecting two connection points of the two straight line components (ie, two connection points of the part where the hair is broken) with a line. As a result, the region correction unit 122 may generate a corrected hair loss image 940 including a final scalp region in which the broken hair is corrected in the preprocessed hair loss image 920 . The corrected hair loss image 940 is an image from which light reflection areas are removed.

The second hair loss diagnosis unit 123 may analyze the corrected hair loss image 940 to calculate the specific gravity of the scalp and the possibility of progressing hair loss. The second hair loss diagnosis unit 123 may calculate the specific gravity of the scalp using [Equation 2].

Referring to [Equation 2], the second hair loss diagnosis unit 123 extracts the rest of the corrected hair loss image 940 except for the background region and the corrected hair region as the final scalp region, and the corrected hair region and Among the pixels occupied by the final scalp area, a ratio of pixels occupied by the final scalp area may be calculated and determined as a ratio of visible scalp.

In addition, the second hair loss diagnosis unit 123 may calculate the possibility of progressing hair loss using [Equation 3].

In [Equation 3], the hair loss weight is a weight applied to calculate the possibility of hair loss progression, and 0.54 may be exemplified.

For example, if the area corresponding to the scalp or head excluding the background area in the corrected hair loss image 940 has 10,000 pixels and the number of pixels identified as the final scalp area through correction is 5,000, the percentage of visible scalp is It is 50%. And, the possibility of progressing hair loss, that is, the density meaning the degree of hair loss is 50% Х 0.54 = 27%.

The region separating unit 124 may separate the rest of the corrected hair loss image 940 except for the corrected hair region, the light reflection region, and the background region as the final scalp region, that is, the hair loss interest region.

The scalp region relearning unit 125 may relearn the hair loss interest region separated by the region separation unit 124 to generate a scalp color model to be used by the GMM in the preprocessing unit 121 .

The second hair loss information providing unit 126 is a result of diagnosing the hair loss condition in the second hair loss condition diagnosis unit 123, that is, the second hair loss diagnosis result including the specific gravity of the scalp, the possibility of progressing hair loss, and detailed diagnosis contents. It can be provided to the user terminal 10 .

10 is an exemplary view of a second hair loss diagnosis result screen 1000 displayed on the user terminal 10 . Referring to FIG. 10 , the second hair loss information providing unit 126 provides a result of diagnosing hair loss based on an image 910 of a user's scalp.

Referring back to FIG. 1 , the product recommendation unit 130 selects a recommended management product search menu from the first hair loss diagnosis result screen 800 shown in FIG. City) or a hair loss product sales server (not shown) associated with the server 100 may provide search results through a search result screen 1100 shown in FIG. 11 .

The non-face-to-face medical treatment connection unit 140 is configured when the menu for receiving medicine after treatment is selected on the second hair loss diagnosis result screen 1000 shown in FIG. 10 or the non-face-to-face prescription diagnosis menu is selected on the search result screen 1100 of FIG. 11 , the hospital list screen 1200a shown in FIG. 12A is provided to the user terminal 10 .

Referring to FIG. 12A , the hospital list screen 1200a includes a hospital list by region and recommended hospitals. When the user selects a hospital, the non-face-to-face treatment connection unit 140 provides a hospital introduction screen 1200b shown in FIG. 12B. When the non-face-to-face treatment request menu is selected from the hospital introduction screen 1200b, the non-face-to-face treatment connection unit 140 provides the treatment request screen 1200c shown in FIG. 12c. When the user inputs personal information, hair loss symptoms, images, etc. through the medical treatment request screen 1200c on the user terminal 10 and requests medical treatment, the non-face-to-face treatment linking unit 140 sends a server (not shown) to the selected hospital. and request non-face-to-face treatment. A server (not shown) of the hospital issues a prescription and provides it to the user terminal 10 .

13 is a flowchart schematically illustrating an artificial intelligence-based hair loss diagnosis method according to an embodiment of the present invention.

Referring to FIG. 13 , the AI-based hair loss diagnosis server 100 receives a user response to a survey item for diagnosing hair loss after the first hair loss diagnosis mode is selected from the user terminal 10 (S1310).

The AI-based hair loss diagnosis server 100 analyzes the user response received in step S1310 based on AI to diagnose the hair loss condition and generates a first hair loss condition diagnosis result (S1320).

The AI-based hair loss diagnosis server 100 provides the first hair loss diagnosis result generated in step S1320 to the user terminal 10 (S1330).

In addition, the AI-based hair loss diagnosis server 100 receives a hair loss image obtained by photographing the user's scalp after the second hair loss diagnosis mode is selected from the user terminal 10 (S1340).

The AI-based hair loss diagnosis server 100 pre-processes the hair loss image received in step S1340 and analyzes the hair loss image based on AI to generate a second hair loss condition diagnosis result including hair loss density (S1350).

The AI-based hair loss diagnosis server 100 provides the second hair loss condition diagnosis result generated in step S1350 to the user terminal 10 (S1360).

The AI-based hair loss diagnosis server 100 provides a non-face-to-face medical treatment service when a non-face-to-face medical treatment service is requested based on the first hair loss condition diagnosis result or the second hair loss condition diagnosis result from the user terminal 10 (S1370) ( S1380).

FIG. 14 is a flowchart showing steps S1310 to S1330 shown in FIG. 13 in detail.

Referring to Figure 14, the user terminal 10 displays the initial screen 200 by installing and executing the hair loss diagnosis application (S1405).

When the user selects the first hair loss diagnosis mode menu 210 from the initial screen 200 (S1410), the user terminal 10 asks questions to determine the progress of the user's hair loss, selects the user's scalp type, selects the scalp condition, A user response may be received by generating survey items including selection of concerns related to hair loss and the user's gender and age as shown in FIGS. 3A to 6 (S1415).

The user terminal 10 may transmit the user's gender and age input in step S1415 and user responses to survey items to the AI-based hair loss diagnosis server 100 (S1420).

The AI-based hair loss diagnosis server 100 calculates a hair loss progress score by applying a weight score set for each survey item to a user response (S1425).

The AI-based hair loss diagnosis server 100 may generate a first hair loss diagnosis result by analyzing the hair loss progress score calculated in step S1425 and the gender and age group of the user based on AI (S1430).

The AI-based hair loss diagnosis server 100 may transmit the first hair loss diagnosis result generated in step S1430 to the user terminal 10 (S1435).

The user terminal 10 may display a screen 800 showing the received first hair loss diagnosis result (S1440).

FIG. 15 is a flowchart showing steps S1340 to S1360 shown in FIG. 13 in detail.

Referring to Figure 15, the user terminal 10 displays the initial screen 200 by installing and executing the hair loss diagnosis application (S1505).

When the user selects the second hair loss diagnosis mode menu 220 from the initial screen 200 (S1510), the user terminal 10 displays a screen for capturing the scalp, and the hair loss image of the scalp (or previously stored hair loss image) is displayed. image) is obtained and transmitted to the AI hair loss diagnosis server 100 (S1515, S1520).

The AI-based hair loss diagnosis server 100 may obtain a pre-processed image 920 including a temporary scalp region by pre-processing the received hair loss image 910 photographing the hair loss progression region (S1525).

The AI-based hair loss diagnosis server 100 may generate a corrected hair loss image 940 by correcting a hair region in which loss occurs due to light reflection when taking a photo (S1525). In step S1525, the temporary scalp region of the preprocessed hair loss image 920 is analyzed, both the initial point and the connection point are determined for a plurality of straight line components, and the area between the connection points generated by connecting the connection points is determined as a light reflection area. In addition, if it is determined that the two closest straight line components are broken hair, a hair loss image 940 corrected for the hair area may be generated by connecting two connection points of the two straight line components with a line.

The AI-based hair loss diagnosis server 100 may analyze the hair loss image 940 corrected in step S1525 to calculate the specific gravity of the scalp and the possibility of progressing hair loss (S1530).

The AI-based hair loss diagnosis server 100 may generate a second hair loss diagnosis result including the specific gravity of the scalp calculated in step S1530, the possibility of progression of hair loss, and detailed diagnosis information, and transmit it to the user terminal 10 (S1535, S1540). ).

The user terminal 10 may display a screen 1000 showing the received second hair loss diagnosis result (S1545).

In addition, the AI-based hair loss diagnosis server 100 separates the rest of the hair loss image 940 corrected in step S1525 except for the corrected hair region, the light reflection region, and the background region as the final scalp region, that is, the hair loss interest region. It can (S1550).

The AI-based hair loss diagnosis server 100 may relearn the separated hair loss region of interest in step S1550 to generate a scalp color model to be used by the GMM in step S1525 and store it in a database (not shown) (S1555).

In the above, even though all the components constituting the embodiment of the present invention have been described as being combined or operated as one, the present invention is not necessarily limited to these embodiments. That is, within the scope of the object of the present invention, all of the components may be selectively combined with one or more to operate. In addition, although all of the components may be implemented as a single independent piece of hardware, some or all of the components are selectively combined to perform some or all of the combined functions in one or a plurality of pieces of hardware. It may be implemented as a computer program having. Codes and code segments constituting the computer program may be easily inferred by a person skilled in the art. Such a computer program may implement an embodiment of the present invention by being stored in a computer readable storage medium, read and executed by a computer.

On the other hand, although the above has been described and illustrated in relation to preferred embodiments for illustrating the technical idea of the present invention, the present invention is not limited to the configuration and operation as shown and described in this way, and departs from the scope of the technical idea. It will be apparent to those skilled in the art that many changes and modifications can be made to the present invention without modification. Accordingly, all such appropriate alterations and modifications and equivalents are to be regarded as falling within the scope of the present invention. Therefore, the true technical protection scope of the present invention should be determined by the technical spirit of the attached claims.

10: user terminal
11: first hair loss diagnosis request unit
12: second hair loss diagnosis request unit
13: diagnosis result output unit
100: AI-based hair loss diagnosis server
110: first hair loss diagnosis unit
120: second hair loss diagnosis unit
130: product recommendation unit
140: Non-face-to-face treatment linkage

Claims (14)

A user terminal for requesting a hair loss diagnosis service by installing and running a hair loss diagnosis application; and
When a request for a hair loss diagnosis service is received from the user terminal, a first hair loss diagnosis mode diagnoses hair loss based on a survey using an artificial intelligence-based hair loss diagnosis program, and a second hair loss diagnosis mode diagnoses hair loss based on a scalp image. An artificial intelligence-based hair loss diagnosis server for diagnosing a user's hair loss state using at least one of the above and providing a diagnosis result to the user terminal;
The artificial intelligence-based hair loss diagnosis server,
a pre-processing unit that obtains a temporary scalp area by pre-processing a hair loss image obtained by photographing a hair loss progressing region input from the user terminal when the second hair loss diagnosis mode is selected from the user terminal;
a region correction unit located in the temporary scalp region of the preprocessed hair loss image and generating a corrected hair loss image by correcting a hair region in which loss occurs due to light reflection when a photo is taken by the user terminal; and
A first hair loss condition diagnosis unit for analyzing the corrected hair loss image to calculate the possibility of hair loss progression;
The pre-processing unit extracts a temporary scalp region using a Gaussian Mixture Model (GMM),
The artificial intelligence-based hair loss diagnosis server,
a region separating unit separating the rest of the corrected hair loss image, except for the corrected hair region, the light-reflected region, and the background region, as a region of interest for hair loss; and
a scalp region re-learning unit re-learning the hair loss-of-interest region separated by the region separation unit and generating a scalp color model to be used by the GMM in the pre-processing unit;
Artificial intelligence-based hair loss diagnosis system characterized in that it further comprises.
According to claim 1,
The artificial intelligence-based hair loss diagnosis server,
When the first hair loss diagnosis mode is selected from the user terminal and the user's response to the survey item for diagnosing hair loss is received, the user's response to the received survey item is analyzed based on artificial intelligence to determine the hair loss state. a first hair loss diagnosis unit for diagnosing;
Artificial intelligence-based hair loss diagnosis system comprising a.
According to claim 2,
The survey items are,
An artificial intelligence-based hair loss diagnosis system, characterized in that it includes questions for identifying the progress of the user's hair loss, selection of the user's scalp type, selection of the scalp condition, selection of concerns related to hair loss, and the user's gender and age.
According to claim 3,
The first hair loss condition diagnosis unit,
A hair loss progress score is calculated by applying the weight score set for each survey item to the user's response to the survey item using the weight score set for each survey item, and the calculated hair loss progress score and the user's gender Artificial intelligence-based hair loss diagnosis system, characterized in that for outputting the first hair loss diagnosis result by analyzing the age and age based on artificial intelligence.
delete According to claim 1,
The area correction unit,
Artificial intelligence-based hair loss diagnosis system, characterized in that by analyzing the temporary scalp region of the preprocessed hair loss image to set a hair region including the hair breakage portion, and to correct the hair region by connecting the hair breakage portion.
delete (A) requesting a hair loss diagnosis service by the user terminal installing and executing a hair loss diagnosis application; and
(B) a first hair loss diagnosis mode in which the artificial intelligence-based hair loss diagnosis server diagnoses hair loss based on a questionnaire survey using an artificial intelligence-based hair loss diagnosis program when the hair loss diagnosis service request is received from step (A); Diagnosing the user's hair loss condition using at least one of the second hair loss diagnosis modes for diagnosing hair loss based on the scalp image, and providing a diagnosis result to the user terminal;
In step (B),
(B4) obtaining a temporary scalp area by pre-processing a hair loss image obtained by photographing a hair loss progression region input from the user terminal when the second hair loss diagnosis mode is selected from the user terminal;
(B5) generating a corrected hair loss image by correcting a hair region located in a temporary scalp region of the preprocessed hair loss image and in which loss occurs due to light reflection when a photo is taken by the user terminal; and
(B6) calculating the possibility of hair loss progression from the corrected hair loss image;
In the step (B4), a temporary scalp region is extracted using a Gaussian Mixture Model (GMM),
In step (B),
(B7) separating an area other than the corrected hair area, the light-reflected area, and the background area from the corrected hair loss image as a hair loss interest area; and
(B8) generating a scalp color model to be used by the GMM in step (B4) by re-learning the separated hair loss region of interest in step (B7);
Artificial intelligence-based hair loss diagnosis method characterized in that it further comprises.
According to claim 8,
In step (B),
(B1) selecting a first hair loss diagnosis mode from the user terminal and receiving a user's response to a survey item for diagnosing hair loss;
(B2) diagnosing a hair loss state by analyzing a user's response to a survey item received from the user terminal based on artificial intelligence; and
(B3) providing a result of diagnosing the hair loss condition generated in step (B2) (hereinafter, referred to as 'first hair loss diagnosis result') to the user terminal;
Artificial intelligence-based hair loss diagnosis method comprising a.
According to claim 9,
The survey items are,
A method for diagnosing hair loss based on artificial intelligence, characterized in that it includes a question for identifying the progress of the user's hair loss, selection of the user's scalp type, selection of the scalp condition, selection of concerns related to hair loss, and the user's gender and age.
According to claim 10,
The (B2 step,
A hair loss progress score is calculated by applying the weight score set for each survey item to the user's response to the survey item using the weight score set for each survey item, and the calculated hair loss progress score and the user's gender Artificial intelligence-based hair loss diagnosis method, characterized in that for outputting the first hair loss diagnosis result by analyzing the age and age based on artificial intelligence.
delete According to claim 8,
In the step (B5),
Artificial intelligence-based hair loss diagnosis method, characterized in that by analyzing the temporary scalp region of the pre-processed hair loss image to set a hair region including the part where the hair is broken, and correcting the hair region by connecting the part where the hair is broken. delete

Images