KR102413404B1 - skin condition analyzing and skin disease diagnosis device - Google Patents

Abstract

The present invention relates to a skin condition analysis and skin disease diagnosis device.
The skin condition analysis and skin disease diagnosis device according to the present invention is an all-in-one type device capable of performing not only cosmetic skin condition analysis but also medical diagnosis of skin disease items.

Description

Skin condition analyzing and skin disease diagnosis device

The present invention relates to a skin condition analysis and skin disease diagnosis device.

Specifically, the present invention relates to an all-in-one type skin condition analysis and skin disease diagnosis device capable of performing not only cosmetic skin condition analysis but also medical skin condition diagnosis in a single device.

In recent years, as interest in beauty increases, interest in skin beauty of a user's face is increasing. In particular, a skin condition measuring device for photographing a user's facial skin and analyzing various skin items (eg, wrinkles, pores, acne, etc.) on the user's face has been developed.

A conventional skin condition analysis apparatus and system transmits a user's face image measured through a portable or non-portable device to a cloud server, and then provides a result analyzed through an analysis model in the cloud system to the user, a cloud-based two-layer Systems are the mainstream, and the cloud-based skin condition analysis device and system have various problems, such as a security problem such as exposure of personal information or a problem of a decrease in data processing speed.

On the other hand, the conventional skin condition measurement device and system can only measure pores, sebum, wrinkles, pigmentation or redness, which are cosmetic skin condition analysis items, and can be used for major skin disease items such as atopic dermatitis, psoriasis, acne or vitiligo. The analysis has technical limitations.

In addition, devices that measure or analyze skin conditions by applying artificial intelligence models such as deep learning or machine learning are being developed recently. There is still a technical challenge in the development of a skin disease analysis device that can be used in places such as a specialized dermatology clinic by providing a comprehensive guide.

Republic of Korea Patent No. 10-2052722

The present invention relates to a skin condition analysis and skin disease diagnosis device capable of performing not only cosmetic skin condition analysis but also diagnosis of medical skin disease items and providing the results to a user.

The present invention also relates to a skin condition analysis and skin disease diagnosis device capable of providing a user with a visualization of the temperature for each region of the user's face through skin temperature mapping.

The present invention further relates to a skin condition analysis and skin disease diagnosis device that can visually provide a user with a simulation of improvement and deterioration for each item according to a skin condition analysis result and a skin disease diagnosis result.

The present invention has been devised to solve the above problems, and relates to a skin condition analysis and skin disease diagnosis device.

The skin condition analysis and skin disease diagnosis device includes: a facial interpolation region having a space in which a user's face is inserted and fixed; a photographing module located in the facial interpolation region and designed to acquire first and second user face images for analysis of a skin condition analysis item and a skin disease diagnosis item; Analysis of skin condition analysis items including wrinkles, pigmentation, redness, sebum and pores is performed using the first user's face image, and acne, atopic dermatitis, and psoriasis are performed using the first and second user's face images. , a processor designed to perform a diagnosis for skin disease diagnostic items including urticaria, dermatitis and vitiligo; and a rotatable touch display that provides the user with the analysis result of the skin condition analysis item and the skin disease diagnosis item performed by the processor. In addition, the photographing module includes a photographing unit including a first camera designed to acquire the first user's face image and having a first polarizing film positioned in front, and a second camera designed to acquire the second user's face image; A 1a white light emitting light source having a second polarizing film having the same polarization axis as the first polarizing film on the front side and a 1b white light emitting light source having a third polarizing film having a different polarization axis from the first polarizing film on the front side a module front region including a first white light emitting light source including a first white light emitting light source unit, and a UV light emitting light source unit including a UV light emitting light source having a visible light blocking filter positioned on the front thereof; and a module rear region including a second white light emitting light source including a second white light emitting light source positioned to correspond to the first white light emitting light source and the UV light emitting light source as a rear region of the front surface of the module. In addition, the processor includes an auto-focusing unit for controlling the first camera to form an auto-focus on the user's face; a blur detection unit for determining whether the first user's face image exceeds a preset blur value; After extracting a face region from the first user face image obtained by the first camera and passing through the blur detection unit, encoding for feature extraction is performed on the extracted face region, and a first face with a class value for each pixel set a first deep learning unit performing decoding to obtain a feature map; After extracting a face region from the first user face image obtained by the first camera, the encoding result for extracting features of the extracted face region and extracting the face region from the second user face image obtained by the second camera a second deep learning unit for forming a second facial feature map through a feature combination of a result of encoding for a later extracted facial region; a skin condition analysis unit that analyzes skin condition analysis items including wrinkles, pigmentation, redness, sebum, and pores using the first facial feature map; and a skin disease diagnosis unit that diagnoses skin disease diagnosis items including acne, atopic dermatitis, psoriasis, urticaria, dermatitis, and vitiligo by using the second facial feature map.

In one example, the face interpolation region may further include a reflective mirror present on a surface of the module rear region. In this case, the second white light emission light source may be reflected by the reflective mirror to be indirectly irradiated to the user.

In one example, the photographing unit of the photographing module includes: a first camera designed to acquire a two-dimensional user face image having a number of pixels of 2 million or more as the first user face image; and a second camera designed to acquire an IR temperature map image as the second user face image.

In one example, the front area of the module includes a photographing unit located in the center; the 1a white light emitting light source and the 1b white light emitting light source positioned as a pair on the left and right sides of the photographing unit, respectively; and the UV light emitting light source positioned above and below the photographing unit, wherein the module rear region is positioned to correspond to the 1a white light emitting light source, the 1b white light emitting light source and the UV light emitting light source A second white light emission light source may be included.

In one example, the second deep learning unit performs pre-processing by performing any one or more of pixel value transformation, spatial domain transformation, frequency domain transformation, and morphological operation on the first user face image and the second user face image, respectively, and then encoding and decoding can be performed.

In one example, the processor extracts a face contour from the first user face image to form a first face contour image, and extracts a face contour from the second user face image to form a second face contour image, The method may further include a skin temperature mapping unit configured to align a skin temperature analysis result according to the second facial contour image to the first facial contour image to form a skin temperature visualization image.

In one example, the skin temperature mapping unit extracts a facial contour from the first user's facial image through landmark detection to form a first facial contour image, and uses a similar har feature from the second user's facial image to extract a facial contour may be extracted to form a second face contour image.

In one example, the processor generates a simulation image of skin improvement and deterioration along with a solution related to skin condition improvement and prescription based on the skin condition analysis and skin disease diagnosis results in the skin condition analysis unit and the skin disease diagnosis unit It may further include a simulation unit to generate.

In one example, the simulation unit processes individual pixel values in the first user's face image according to item-by-item improvement and exacerbation step weights for the skin condition analysis and skin disease diagnosis results to generate a simulation image of skin improvement and deterioration. can create

In one example, the rotatable touch display may be designed to rotate 360 degrees by providing a rotating means at the lower end thereof.

The skin condition analysis and skin disease diagnosis device according to the present invention can perform an analysis of a user's cosmetic skin condition and a diagnosis of a medical skin disease item more accurately than a specialist's examination and palpation, and immediately visualize it to the user can be provided.

The device for analyzing a skin condition and diagnosing a skin disease according to the present invention may visually provide a temperature distribution or condition for each face region to a user.

The skin condition analysis and skin disease diagnosis device according to the present invention may visually provide a user with a simulation of improvement and deterioration for each item according to the skin condition analysis result and the skin disease diagnosis result.

Of course, the scope of the present invention is not limited by these effects.

1 is a block diagram of a structure for a skin condition analysis and skin disease diagnosis device according to the present invention.
2 and 3 are diagrams for explaining in more detail the structure of the front area (FIG. 2) and the rear area (FIG. 3) of the imaging module according to the present invention, respectively.
4 is a block diagram for explaining in detail a process in which a skin condition analysis and skin disease diagnosis device according to the present invention performs an analysis on a skin condition analysis item.
5A and 5B are a flow block diagram (a) and an exemplary flow diagram (b) for specifically explaining a process in which a skin condition analysis and skin disease diagnosis device according to the present invention performs an analysis on a skin disease diagnosis item, respectively; )to be.
6A and 6B are one block flowchart (a) and one flow example diagram (b) for specifically explaining a process in which the skin temperature mapping unit in the processor forms a skin temperature visualization image according to the present invention, respectively.
7 is a block diagram illustrating in detail a process of generating a simulation image for skin improvement and deterioration by a simulation unit within a processor according to the present invention.

Hereinafter, the present invention will be described in more detail with reference to drawings and examples.

In this specification, the singular expression includes the plural expression unless otherwise specified.

The terms used in the present specification, while considering the functions in the present invention, currently widely used general terms are selected, which may vary depending on the intention or precedent of a person skilled in the art, the emergence of new technology, and the like. In addition, in a specific case, there is a term arbitrarily selected by the applicant, and in this case, the meaning will be described in detail in the description of the corresponding invention. Therefore, the term used in the present invention should be defined based on the meaning of the term and the overall content of the present invention, rather than the name of a simple term.

Embodiments of the present invention can apply various transformations and can have various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the scope of the specific embodiments, and it should be understood to include all modifications, equivalents and substitutes included in the spirit and scope of the invention. In describing the embodiments, if it is determined that a detailed description of a related known technology may obscure the subject matter, the detailed description thereof will be omitted.

In this specification, terms such as first, second, etc. may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

In this specification, terms such as "comprises" or "consisting of" are intended to designate that the features, numbers, steps, operations, components, parts, or combinations thereof described in the specification exist, but one or more other It should be understood that this does not preclude the possibility of addition or presence of features or numbers, steps, operations, components, parts, or combinations thereof.

In this specification, each component of the term 'module' or 'unit' performs at least one function or operation, and may be implemented as hardware or software, or a combination of hardware and software. In addition, the plurality of ‘modules’ or ‘units’ may be operated and controlled by at least one processor, except for ‘modules’ or ‘units’ that need to be implemented with specific hardware.

Hereinafter, a skin condition analysis and skin disease diagnosis device according to the present invention will be described in more detail with reference to the accompanying drawings.

A skin condition analysis and skin disease diagnosis device according to the present invention is a device for analyzing a skin condition and disease based on a face image captured by a camera after a user fixes a face inside a device having a predetermined shape.

The skin condition analysis and skin disease diagnosis device according to the present invention provides cosmetic skin based on a plurality of user face images acquired by a plurality of cameras and a convolutional neural network-based deep learning model that analyzes them. In addition to analyzing the condition, medical skin diagnosis can also be performed, and the skin condition analysis and skin diagnosis results are provided in a single device.

The skin condition analysis and skin disease diagnosis device according to the present invention can also provide skin temperature mapping results and simulation results according to improvement/deterioration, which can be visualized and provided to the user immediately.

1 is a block diagram of a structure for a skin condition analysis and skin disease diagnosis device according to the present invention.

As shown in FIG. 1 , the device for skin condition analysis and skin disease diagnosis according to the present invention includes a photographing and diagnosis unit 1 and a photographing and diagnosis unit for performing skin condition analysis and skin disease diagnosis by photographing a user's face. It may include a display unit 2 that visually provides the user with the result in (1), and a support member 3 that integrates the imaging and diagnosis unit 1 and the display unit 2 . Although the imaging and diagnosis unit 1 is shown on the left and the display unit 2 is shown as being located on the right in FIG. 1, this is only an example according to the present invention. The position and structure of the part (2) can be changed, and the position or structure of FIG. 1 is not limitedly interpreted.

The imaging and diagnosis unit 1, the display unit 2, and the support member 3 each include individual components of the skin condition analysis and skin disease diagnosis device according to the present invention.

Specifically, the imaging and diagnosis unit 1 includes the facial interpolation region 100 and the imaging module 200 , and the display unit 2 includes a rotatable touch display 400 . In addition, the processor 300 may be included in the imaging and diagnosis unit 1 or the display unit 2 .

The facial interpolation region 100 is a region serving to fix the user's face, and has a space in which the user's face of a general non-portable skin condition analysis device is inserted and fixed.

In one example, the facial interpolation region 100 may include a head fixing region and a chin fixing region, wherein the head fixing region and the chin fixing region serve to fix the upper part of the user's head (forehead region) and the chin region, respectively. can be done

On the other hand, the module rear region 202 of the imaging module 200 according to the present invention includes the second white light emission light source unit 2021, and the facial interpolation region 100 is the second white light emission light source unit 2021. The light source is indirect As a light source, it may further include a reflective mirror to be indirectly irradiated when photographing the user's face.

In one example, the facial interpolation region 100 may further include a reflective mirror present at the face of the module back region 202 . In this case, the second white light emission light source 2021a may be reflected by the reflective mirror to be indirectly irradiated to the user.

The shape, shape, and specification of the reflection mirror follow the structure or shape of the face interpolation region 100 and exist on the module rear region 202 surface so that the second white light emission light source 2021a can be indirectly irradiated to the user. As long as it is a degree, the form, shape, and standard are not particularly limited.

The imaging module 200 is positioned inside the facial interpolation region 100 .

The imaging module 200 is a component for performing an analysis on a skin condition analysis item and a diagnosis for a skin disease diagnosis item, and is located in the facial interpolation area 100 for analysis of the skin condition analysis item and the skin disease diagnosis item. It is designed to acquire first and second user face images. In the above, the first and second user face images are captured and obtained by different devices (cameras), respectively, and are used for analysis of skin condition analysis items as a single image or diagnosis for skin condition diagnosis items as a combination image image used for

The photographing module 200 includes a module front area 201 positioned in the user's face direction and a module rear area 202 positioned in a direction opposite to the user's face direction.

2, the module front area 201 is described in more detail with reference to FIG. 2, the module front area 201 is designed to acquire a first user's face image and the first polarizing film is located in the front of the first camera 2011a ) and a second camera (2011b) designed to acquire a second user's face image; A first white light emission light source unit 2012 including a light source 2012a and a first white light emission light source 2012b having a third polarizing film having a different polarization axis from the first polarizing film located on the front surface, and a visible light blocking filter on the front surface and a UV light emission light source unit 2013 including a UV light emission light source 2013a positioned in the .

The photographing unit 2011 includes a first camera 2011a and a second camera 2011b designed to acquire first and second user face images.

The first camera 2011a may be a camera having 2 million pixels or 8 million pixels or more in order to obtain a high-quality two-dimensional user face image of FHD or UHD level or higher. The high-quality two-dimensional user's face image obtained by the first camera 2011a may be used for analysis of a skin condition analysis item or a skin disease diagnosis item as the first user's face image.

The second camera 2011b may be an infrared camera that acquires an IR temperature map image of a user's face. The IR temperature map image obtained from the infrared camera may be used for analysis of skin disease diagnosis items that can be distinguished from heat, such as atopy, psoriasis, acne, and urticaria, along with a two-dimensional user's face image.

The photographing unit 2011 may be located in the center or one side of the module front area 201 , and preferably, as shown in FIG. 2 , may be located in the center of the module front area 201 .

The first white light emitting light source unit 2012 includes a 1a white light emitting light source 2012a having a second polarizing film having the same polarization axis as the first polarizing film on the front side, and a third polarized light having a different polarization axis from the first polarizing film. The film includes a 1b white light emitting light source 2012b positioned on the front side.

The first white light emitting light source 2012a is a light source emitting light when a second polarizing film having the same polarization axis as the first polarizing film is positioned on the front surface of the first polarizing film in normal mode shooting by the first camera 2011a, so-called day light illuminant ) may be a light source having a color temperature of 3500 to 6500K.

The 1b white light emission light source 2012b is a light source emitting light when a third polarizing film having a different polarization axis from the first polarizing film is positioned on the front side and shooting in the polarization mode or light reflection removal mode by the first camera 2011a. It may be a light source having the same color temperature as the white light emission light source 2012a. The angles of the polarization axes of the third polarizing film and the first polarizing film may be orthogonal to each other.

The number or positions of the 1a white light emission light sources 2012a and the 1b white light emission light sources 2012b are not particularly limited. For example, as shown in FIG. 2 , the 1a white light emission light sources 2012a and 1b are not particularly limited. The white light emission light sources 2012b may be located in pairs on the left and right sides of the photographing unit 2011 located in the center of the module front area 201 , respectively.

The UV light emitting light source unit 2013 includes a UV light emitting light source 2013a having a visible light blocking filter positioned on the front thereof.

The UV light emission light source 2013a is a light source that is emitted during UV pass mode shooting by the first camera 2011a, and may be, for example, a light source that emits UVA light or UVB light.

The number or position of the UV light emitting light sources 2013a is not particularly limited. For example, as shown in FIG. 2 , on the upper and lower sides of the photographing unit 2011 located in the center of the module front region 201 . Each may be located in a cluster.

As the opposite surface of the module front area 201 , the module rear area 202 is located in a direction opposite to the user's face direction.

Referring to FIG. 3 , the module rear region 202 is described in more detail, wherein the module rear region 202 includes a first white light emitting light source 2012 and a second white light emitting light source corresponding to the UV light emitting light source 2013 . and a second white light emission light source unit 2021 including a light emission light source 2021a. That is, the module rear region 202 may include the second white light emission light sources 2021a forming a cluster on the top, bottom, left, and right sides of the region corresponding to the photographing unit 2011 located in the center of the module front region 201 .

On the other hand, the second white light emission light source 2021a included in the module rear region 202 may be reflected by the aforementioned reflective mirror as indirect light to be indirectly irradiated to the user when photographing the user's face.

The first and second user face images obtained by the photographing module 200 are analyzed by the processor 300 .

The processor 300 controls the operation of the individual components according to the present invention, and performs an analysis on a skin condition analysis item and a diagnosis on a skin disease diagnosis item using the first and second user face images.

Specifically, the processor 300 analyzes the skin condition analysis items including wrinkles, pigmentation, redness, sebum, and pores by using the first user's face image, and the first and second user's face images It is designed to diagnose skin diseases including acne, atopic dermatitis, psoriasis, urticaria, dermatitis and vitiligo.

More specifically, the processor 300 includes an auto-focusing unit 301 that controls the first camera 2011a to automatically focus on the user's face and shoot; a blur detection unit 302 for determining whether the first user's face image exceeds a preset blur value; After extracting a face region from the first user face image obtained by the first camera 2011a and passing through the blur detection unit 302, encoding for feature extraction is performed on the extracted face region, and class values for each pixel a first deep learning unit 303 that performs decoding to obtain the set first facial feature map; After extracting a face region from the first user's face image acquired by the first camera 2011a, from the encoding result for feature extraction of the extracted facial region and the second user's face image acquired by the second camera 2011b a second deep learning unit 304 that extracts the facial region and forms a second facial feature map through a feature combination of the extracted facial region encoding result; a skin condition analysis unit 305 that analyzes skin condition analysis items including wrinkles, pigmentation, redness, sebum, and pores using the first facial feature map; and a skin disease diagnosis unit 306 that diagnoses skin disease diagnosis items including acne, atopic dermatitis, psoriasis, urticaria, dermatitis, and vitiligo by using the second facial feature map.

The auto-focusing unit 301 performs a control role so that the user's first camera 2011a automatically focuses on the user's face when photographing the user's face for skin condition analysis. Through the auto-focusing unit 301 , only the user's face of the processor 300 may be effectively extracted and analyzed.

The blur detection unit 302 is configured to determine whether the image acquired by the first camera 2011a exceeds a preset blur value when the user's face is photographed. It plays a role in inducing retake. Therefore, when a user's face image exceeding the preset blur value of the blur detection unit 302 is captured, the image is not used for skin condition analysis, and the processor 300 operates the first camera through the auto-focusing unit 301 . (2011a) induces reshooting with autofocus formation.

The first deep learning unit 303 passes through the blur detection unit 302 and obtains a first facial feature map based on a first user face image used for skin condition analysis. ) based deep learning model.

Specifically, the first deep learning unit 303 may apply a modified reception encoding module as an encoding module and a DeepLabv3+ decoding module as a decoding module.

More specifically, the first deep learning unit 303 extracts a face area from the first user's face image, and performs a regular convolution operation on the user's face image data for feature extraction of the extracted face area; Encoding is performed to generate a feature map by performing an Xception operation and a Rethinker operation, and an Atrous Spatial Pyramid Pooling (ASP) operation and up-sampling of the feature map. Then, the first user facial feature map is formed through a Convolutional Neural Network-based deep learning algorithm that performs decoding to concatenate a feature map having a corresponding dimension. can In addition, the Rethinker operation is a locally designed convolutional LSTM layer or 3D convolutional layer together with an SE block including a squeeze operation unit and an excitation operation unit. may include, and through this, it is possible to collect wide-area spatial information and extract a correlation to the skin condition analysis item object in the local space. On the other hand, regular convolution operation, Xception operation, Rethinker operation, locally designed convolutional LSTM layer and 3D convolutional layer (Conv3D) used in this specification layer) is a technical term widely used among those skilled in the art of convolutional neural networks, and unless otherwise specified in this specification, the term is commonly understood by those skilled in the art It has definition and meaning.

Accurate skin condition analysis is possible through the configuration of the first deep learning unit 303 as described above, and in particular, a deep learning algorithm that considers the correlation between detection target objects is applied to derive results close to expert examination and palpation. Image formation for skin condition analysis is possible.

The first facial feature map image obtained by the first deep learning unit 303 is transmitted to the skin condition analysis unit 305 for analysis of skin condition analysis items including wrinkles, pigmentation, redness, sebum and pores. In this case, in order to improve or enhance the image, the first deep learning unit 303 may additionally perform a predetermined pre-processing.

In one example, the first deep learning unit 303 performs encoding and decoding after pre-processing by performing any one or more of pixel value transformation, spatial domain transformation, frequency domain transformation, and morphological operation on the first user's face image. can

Referring to FIG. 4 , when the skin condition analysis and skin disease diagnosis device according to the present invention describes the processor for performing the analysis on the skin condition analysis item, auto-focusing is performed through the control of the auto-focusing unit 301 . A first user's face image is obtained by the first camera 2011a, a face region is extracted from the obtained first user's face image, and the first user's face image obtained through the blur detection unit 302 is set to a preset blur value. When it is determined that the first user face image does not exceed the preset blur value after determining whether it exceeds is carried out Next, analysis of skin condition analysis items including pores, wrinkles, sebum, pigmentation, and redness is performed by the skin condition analysis unit 305 , and the results are provided to the user through the rotary touch display 400 . do. On the other hand, when the blur detection unit 302 determines that the first user's face image exceeds a preset blur value, the auto-focusing unit 301 controls the first camera 2011a to automatically focus and re-shoot. induce

The second deep learning unit 304 implements a convolutional neural network-based deep learning model that acquires a second facial feature map based on the first and second user face images used for diagnosing skin diseases. is a configuration that

Like the first deep learning unit 303 , a deep learning algorithm in consideration of the correlation between detection target objects may be applied to the second deep learning unit 304 .

Specifically, the second deep learning unit 304 performs a convolutional operation, an Xception operation, and a Rethinker operation on each of the first user face image and the second user face image to perform each image. After each encoding is performed, an Atrous Spatial Pyramid Pooling (ASPP) operation and up-sampling and then concatenating decoding are performed to combine the features of each encoding result. Algorithm to obtain a second facial feature map can be applied, and the detailed process of each operation module may be the same as or different from that performed in the first deep learning unit 303 .

In addition, the second deep learning unit 304 may additionally perform a predetermined pre-processing process before encoding for feature extraction.

In one example, the second deep learning unit 304 preprocesses the first user face image and the second user face image by performing any one or more of pixel value transformation, spatial domain transformation, frequency domain transformation, and morphological operation. After encoding and decoding can be performed.

The second facial feature map obtained from the second deep learning unit 304 is transmitted to the skin disease diagnosis unit 306 and used for diagnosis of skin disease diagnosis items including acne, atopy, psoriasis, urticaria, dermatitis and vitiligo. do.

Referring to FIGS. 5A and 5B , a process for performing a diagnosis for a skin disease diagnosis item will be described in more detail. The first and second user face images acquired by the first camera 2011a and the second camera 2011b, respectively. After a predetermined preprocessing is performed, it goes through a separate encoding process for feature extraction. Then, the encoded individual results are decoded for feature combination and then converted into one second facial feature map, and the second facial feature map is transmitted to the skin disease diagnosis unit 306 for diagnosis of skin disease diagnosis items. used

The skin condition analysis and skin disease diagnosis device according to the present invention can perform skin condition analysis and skin disease diagnosis at the level of a specialist through the above-described configurations, and can directly visualize and provide this to the user.

The device for analyzing a skin condition and diagnosing a skin disease according to the present invention may also provide the user with the degree of warmth of the skin as additional information, which may be implemented by the skin temperature mapping unit in the processor 300 .

Specifically, the processor 300 extracts a face contour from the first user face image to form a first face contour image, and extracts a face contour from the second user face image to form a second face contour image, The method may further include a skin temperature mapping unit configured to form a skin temperature visualization image by aligning a skin temperature analysis result according to the second facial contour image to the first facial contour image.

More specifically, the skin temperature mapping unit extracts a facial contour from the first user's facial image through landmark detection to form a first facial contour image, and extracts a facial contour from the second user's facial image through similar har features Thus, a second face contour image may be formed. Meanwhile, in the above description, the pseudo-Har feature refers to a technique for filtering an input image using a rectangular filter mask composed of black and white, and then analyzing it through a cascade classifier to extract facial contours.

When the process of the skin temperature mapping unit forming a skin temperature visualization image is described in detail with reference to FIGS. 6A and 6B , the skin temperature mapping unit first and second cameras 2011a and 2011b acquired by the first and second cameras 2011a and 2011b, respectively. First and second facial contour images are formed by applying landmark detection and similar har features to each user's face image, and the skin temperature analysis data analyzed based on the second facial contour image and the first facial contour image Alignment forms a skin temperature visualization image, which is presented to the user via the rotatable touch display 400 .

The skin condition analysis and skin disease diagnosis device according to the present invention may provide a skin condition analysis and skin disease diagnosis result to the user, and at the same time provide a solution for skin condition improvement and prescription and a simulation image accordingly, which This may be done in the simulation unit within the processor.

Specifically, the processor 300 improves and worsens skin condition along with a solution related to skin condition improvement and prescription based on the skin condition analysis and skin disease diagnosis results in the skin condition analysis unit 305 and the skin disease diagnosis unit 306 . It may further include a simulation unit for generating a simulation image of the.

More specifically, the simulation unit generates a simulation image of skin improvement and deterioration by processing individual pixel values in the first user's face image according to item-by-item improvement and exacerbation step weights for skin condition analysis and skin disease diagnosis results. can

Referring to FIG. 7 , when the simulation unit generates a simulation image for skin improvement and deterioration, the simulation unit first and second user faces acquired by the first and second cameras 2011a and 2011b are described in detail. Based on the image, the improvement or deterioration level for the simulation is set according to the skin condition analysis and skin disease diagnosis results performed by the skin condition analysis unit 305 and the skin disease diagnosis unit 306, and the improvement or worsening weight is set accordingly. After calculation, the result of skin condition analysis and skin disease diagnosis may be reflected in the image to be visualized and provided to the user through the rotary touch display 400 .

The above-described skin condition analysis result, skin disease diagnosis result, skin temperature mapping result, and simulation result for improving/deteriorating skin condition may be visualized and provided to the user through the rotary touch display 400 with the naked eye.

On the other hand, the rotary touch display 400 may be designed to be rotated forward and backward, and may include a rotating means at the lower end thereof.

Specifically, the rotatable touch display 400 may be designed to be rotated by 360 degrees by providing a rotating means at the lower end thereof. The rotating means may be fixedly positioned on the supporting member 3 that integrates the imaging and diagnosis unit 1 and the display unit 2 , for example.

Including the above-described components, the skin condition analysis and skin disease diagnosis device according to the present invention can diagnose not only cosmetic skin condition analysis but also medical skin disease items, and improve/repair skin condition and mapping information for skin warmth. It is advantageous in that it is a single device that can provide a simulation image of exacerbation, and can immediately visualize and provide it to the user.

As described above, although the present disclosure has been described with reference to the limited examples and drawings, the present disclosure is not limited to the above examples, and those skilled in the art to which the present disclosure pertains various modifications and variations from these descriptions. This is possible. Therefore, the scope of the present disclosure should not be limited to the described embodiments, but should be defined by the following claims as well as the claims and equivalents.

1: Photography and Diagnosis Department
2: display unit
3: support member
100: facial interpolation area
200: shooting module
201: module front area
2011 : Cinematography
2011a: 1st camera
2011b: 2nd camera
2012: first white light emitting light source unit
2012a,b: 1a,b white light emission light source
2013 : UV light emitting light source part
2013a: UV light emitting light source
202: module rear area
2021: second white light emitting light source unit
2021a: second white light emitting light source
300 : processor
301: auto focusing unit
302: blur detection unit
303: first deep learning unit
304: second deep learning unit
305: skin condition analysis unit
306: skin disease diagnosis unit
400: rotatable touch display

Claims (10)

a facial interpolation region having a space in which the user's face is inserted and fixed;
a photographing module located in the facial interpolation region and designed to acquire first and second user face images for analysis of a skin condition analysis item and a skin disease diagnosis item;
Analysis of skin condition analysis items including wrinkles, pigmentation, redness, sebum and pores is performed using the first user's face image, and acne, atopic dermatitis, and psoriasis are performed using the first and second user's face images. , a processor designed to perform a diagnosis for skin disease diagnostic items including urticaria, dermatitis and vitiligo; and
A skin condition analysis and skin disease diagnosis device comprising a; a rotational touch display that provides a user with an analysis result for a skin condition analysis item and a diagnosis result for a skin disease diagnosis item performed by the processor,
The photographing module is
A photographing unit including a first camera designed to acquire the first user's face image and having a first polarizing film positioned in front, and a second camera designed to acquire the second user's face image, the first polarizing film and A first white light emission including a 1a white light emitting light source having a second polarizing film having the same polarization axis on the front side and a 1b white light emitting light source having a third polarizing film having a different polarization axis on the front side of the first polarizing film a module front area including a light source unit and a UV light emission light source unit including a UV light emission light source having a visible light blocking filter positioned on the front side; and
a module rear region including a second white light emitting light source including a second white light emitting light source positioned corresponding to the first white light emitting light source and the UV light emitting light source as a rear region of the module front region; and
The processor is
an auto-focusing unit controlling the first camera to automatically focus on a user's face and take pictures;
a blur detection unit for determining whether the first user's face image exceeds a preset blur value;
After extracting a face region from the first user face image obtained by the first camera and passing through the blur detection unit, encoding for feature extraction is performed on the extracted face region, and a first face with a class value for each pixel set a first deep learning unit performing decoding to obtain a feature map;
After extracting a face region from the first user face image obtained by the first camera, the encoding result for extracting features of the extracted face region and extracting the face region from the second user face image obtained by the second camera a second deep learning unit for forming a second facial feature map through a feature combination of a result of encoding for a later extracted facial region;
a skin condition analysis unit that analyzes skin condition analysis items including wrinkles, pigmentation, redness, sebum, and pores using the first facial feature map; and
A skin condition analysis and skin disease diagnosis device comprising a; a skin disease diagnosis unit that diagnoses skin disease diagnosis items including acne, atopic dermatitis, psoriasis, urticaria, dermatitis, and vitiligo by using the second facial feature map. The method of claim 1,
The facial interpolation region is
Further comprising a reflective mirror present on the surface of the module rear area,
The second white light emitting light source is reflected by the reflective mirror to indirectly irradiate the user with a skin condition analysis and skin disease diagnosis device. The method of claim 1,
The photographing unit of the photographing module,
a first camera designed to acquire a two-dimensional user face image having a number of pixels of 2 million or more as the first user face image; and
and a second camera designed to acquire an IR temperature map image as the second user face image. The method of claim 1,
The front area of the module is
a photographing unit located in the center;
the 1a white light emitting light source and the 1b white light emitting light source positioned as a pair on the left and right sides of the photographing unit, respectively; and
The UV light emitting light source located above and below the photographing unit;
The module rear area is
A skin condition analysis and skin disease diagnosis device comprising the 1a white light emitting light source, the 1b white light emitting light source, and a second white light emitting light source positioned to correspond to the UV light emitting light source. The method of claim 1,
The second deep learning unit,
Skin condition analysis and skin disease diagnosis in which encoding and decoding are performed after preprocessing by performing any one or more of pixel value conversion, spatial domain conversion, frequency domain conversion, and morphological operation on the first user face image and the second user face image device. The method of claim 1,
The processor is
extracting a face contour from the first user face image to form a first face contour image; extracting a face contour from the second user face image to form a second face contour image; The skin condition analysis and skin disease diagnosis device further comprising a skin temperature mapping unit aligning a skin temperature analysis result with the first facial contour image to form a skin temperature visualization image. 7. The method of claim 6,
The skin temperature mapping unit,
A first facial contour image is formed by extracting a facial contour from the first user's facial image through landmark detection, and a second facial contour image is formed by extracting a facial contour from the second user's facial image through similar har features. A device for analyzing skin conditions and diagnosing skin diseases. The method of claim 1,
The processor is
The skin condition analysis unit and the skin disease diagnosis unit further comprising a simulation unit that generates a simulation image of skin improvement and deterioration together with a solution related to skin condition improvement and prescription based on the skin condition analysis and skin disease diagnosis results in the skin condition analysis unit and skin disease diagnosis unit Condition analysis and skin disease diagnosis device. 9. The method of claim 8,
The simulation unit,
Skin condition analysis and skin disease for generating a simulation image of skin improvement and deterioration by processing individual pixel values in the first user's face image according to the weights for each item improvement and exacerbation step for the skin condition analysis and skin disease diagnosis results diagnostic device. The method of claim 1,
The rotatable touch display,
A skin condition analysis and skin disease diagnosis device designed to be rotated 360 degrees by providing a rotating means at the lower end.

Images