TWI775229B - A skin condition detection system and method thereof - Google Patents

Abstract

The present invention provides a skin condition detection system and method thereof. The method comprises storing multiple sets of facial feature model data and irradiation reference files in a remote server, establishing a 3D facial model data of the subject through the subject’s face color image data and face depth image data, distributing the skin condition detection and determination computing to a skin condition detection device and a skin irradiation device by the edge computing model to automatically detect and determine the skin condition feature image and to irradiate and repair the skin.

Description

Skin beauty detection system and method thereof

The invention relates to a skin beautifying detection system and a method thereof, in particular to a system capable of running a facial feature model through a remote servo device in an edge computing mode, and the skin state detection, judgment and irradiation related operations are distributed in peripheral devices system and method thereof.

Phototherapy is a method of using electromagnetic radiation to generate light of different wavelengths to treat skin diseases. Phototherapy is the oldest and most effective physical treatment for skin diseases. The use of sunbathing to treat skin diseases has been documented as far back as the ancient Greeks, but it wasn't until the early 20th century that a systematic study of phototherapy patterns began.

However, the structure of today's phototherapy devices is too large, which makes it inconvenient to use and cannot be used with you. For example, a box-type phototherapy device is used by users to receive whole body light therapy. However, due to its large size, it needs to be assembled in a specific location. and cannot move freely.

Another relatively small phototherapy device is positioned on the ground with a support frame, and then irradiates the human body with light. However, this type of phototherapy device must be manually adjusted and operated, so the operation is still complicated, and how to accurately target the patient to be treated. Irradiation is still a difficulty that needs to be solved.

Therefore, this case provides a system and method, which can train multiple sets of facial feature model data through deep learning, and then use the color image data of the subject's face and the depth data of the face image to create a 3D image of the subject. Face model data, and in order to detect and judge skin quality, the response speed will not be affected by network conditions and the number of required devices, so the models are distributed on the edge (device side), so that the device can automatically Judging, detecting the target object and performing radiation therapy can greatly increase the availability of the device. Therefore, this case should be an optimal solution.

The skin beautifying detection system of the present invention includes a remote servo device, and includes: a target feature creator, which is based on a plurality of photos with human faces to classify objects according to different skin state feature images, so as to establish a Multiple sets of facial feature model data; an irradiation parameter storage for storing multiple sets of irradiation reference files according to different skin condition feature patterns; a 3D model builder based on color image data of a subject's face and face image depth data, to establish the 3D face model data of the subject; one or more photographic devices for shooting to obtain the subject's face color image data and face image depth The data is sent to the remote server device; one or more skin beauty detection devices are connected to the remote server device to receive multiple sets of facial feature model data and the subject's 3D face model data, and compare them to indicate the skin condition feature pattern on one or more coordinates of the subject's face; one or more skin beautifying irradiation devices are connected to the remote servo device and the The skin beautifying detection device is connected, and the skin beautifying irradiation device includes: an irradiation data adjuster, which is used for targeting one or a plurality of coordinates on the subject's face according to a plurality of sets of irradiation reference files. The characteristic pattern of the skin state is used to set an irradiation parameter data; and an LED irradiation body has a plurality of irradiation units and a control unit connected with the irradiation data adjuster, and each irradiation unit includes at least one LED light, Wherein, the control unit can control different LED lights to irradiate one or more coordinates on the subject's face according to the irradiation parameter data.

More specifically, the skin condition characteristic pattern can be an image of skin texture or/and skin defects.

More specifically, the skin type image is an image of normal skin type, oily skin type, dry skin type or/and combination skin type.

More specifically, the image of the skin defect is an image of spots, wrinkles, acne, acne scars, keratosis or/and moles, and the image of the skin defect is not limited to spots, wrinkles, acne , acne scars, keratinization or/and images of moles.

More specifically, the target feature builder can use a deep learning algorithm to classify targets for different skin state feature images, so as to create multiple sets of facial feature model data.

More specifically, the skin beautifying detection device can determine the type of the skin condition feature pattern on one or more coordinates on the subject's face according to multiple sets of facial feature model data.

More specifically, the operation of skin condition detection and judgment can be dispersedly deployed on the skin beauty detection device and the skin beauty irradiation device in an edge computing mode, so that the skin beauty detection device and the skin beauty irradiation device can be Perform self-judgment and detect the characteristic pattern of the skin state.

More specifically, after the skin beautifying irradiation device is irradiated, new face color image data and face image depth data can be captured by the photographing device, and sent to the remote servo device for updating. a new 3D face model data, and the skin beauty detection device can compare the 3D face model data of the subject before and after irradiation, and reset a new irradiation parameter data through the irradiation data adjuster.

More specifically, the LED light of the LED illuminating body is a variable spectrum LED light.

More specifically, the irradiation unit has one or more infrared receivers to assist in determining the distance between the LED light and one or more coordinates on the subject's face.

More specifically, the control unit can change the illumination depth, illumination angle or/and illumination position of the LED lamp through the displacement mechanism.

More specifically, the control unit can change the irradiation wavelength or/and the irradiation time of different LED lamps.

More specifically, each irradiation unit can change the position of the LED lamp through a displacement mechanism, and the displacement mechanism can be an active adjustment mechanism or a passive adjustment mechanism.

A skin beautifying detection method, comprising: (1) On a remote servo device, according to a plurality of photos with faces, target classification is performed to establish multiple sets of facial feature model data and irradiation reference files; (2) Through one or more photographing equipment, take pictures to obtain the color image data and the depth data of the face image of a subject, and create a 3D face model of the subject through the remote servo equipment data; (3) Send multiple sets of facial feature model data and the subject's 3D face model data to a skin beautifying detection device for comparison by the skin beautifying detection device and mark the subject's face a pattern of skin state characteristics on one or more coordinates; (4) Then, according to the marked skin condition characteristic pattern and the irradiation reference file, set an irradiation parameter data, so that an LED irradiation body with a plurality of irradiation units can respond to the subject's face according to the irradiation parameter data. One or more coordinates on the irradiated.

More specifically, the skin condition characteristic pattern can be an image of skin texture or/and skin defects.

More specifically, the skin type image is an image of normal skin type, oily skin type, dry skin type or/and combination skin type.

More specifically, the image of the skin defect is an image of spots, wrinkles, acne, acne scars, keratinization or/and moles.

More specifically, the remote servo device can use a deep learning algorithm to classify objects according to the characteristic images of different skin states, so as to establish multiple sets of facial feature model data.

More specifically, the skin beautifying detection device can determine the type of the skin condition feature pattern on one or more coordinates on the subject's face according to multiple sets of facial feature model data.

More specifically, the operation of skin condition detection and judgment can be dispersedly deployed on the skin beauty detection device and the skin beauty irradiation device in an edge computing mode, so that the skin beauty detection device and the skin beauty irradiation device can be Perform self-judgment and detect the characteristic pattern of the skin state.

More specifically, after the skin beautifying irradiation device is irradiated, new face color image data and face image depth data can be captured by the photographing device, and sent to the remote servo device for updating. a new 3D face model data, and the skin beauty detection device can compare the 3D face model data of the subject before and after irradiation, and reset a new irradiation parameter data through the irradiation data adjuster.

More specifically, the irradiation depth, irradiation angle, irradiation position, irradiation wavelength or/and irradiation time of the LED lamp can be changed.

Other technical contents, features and effects of the present invention will be clearly presented in the following detailed description of the preferred embodiments with reference to the drawings.

Please refer to Figures 1, 2, 3A~3C, which are the schematic diagram of the overall structure of the skin beautifying detection system and the method thereof, the schematic diagram of the structure of the remote servo device, the schematic diagram of the structure of the skin beautifying irradiation equipment, the schematic diagram of the structure of the irradiation unit, and the schematic diagram of the irradiation unit of the present invention. Schematic diagram of a simple structure, it can be seen from the figure that the skin beautifying detection system at least includes a remote servo device 1 , a photographing device 2 , a skin beautifying detection device 3 and a skin beautifying irradiation device 4 .

The remote servo device 1 at least includes at least one processor and at least one computer-readable recording medium. The computer-readable recording medium further stores computer-readable instructions and a plurality of application units. When the processor executes the computer-readable instructions, it is capable of executing the application unit, and the application unit includes: (1) A target feature generator 11, which is based on a plurality of photos with human faces, classifying targets according to different skin state feature images, and then using a deep learning algorithm (such as Object Detection) to establish multiple sets of facial feature models Data (wherein the skin condition characteristic pattern can be an image of skin skin type or/and skin defects, and the skin type image is normal skin type, oily skin type, dry skin type or/and combination skin type and the image of the skin defect is an image of spots, wrinkles, acne, acne scars, keratosis or/and moles, wherein the image of skin defects is not limited to spots, wrinkles, acne, images of acne scars, keratinization or/and moles); (2) An irradiation parameter storage 12, which stores multiple sets of irradiation reference files according to different skin condition characteristic patterns. The irradiation reference files can be created manually, and can also store existing skin condition characteristics of different colors and ranges. The pattern and the corresponding treatment irradiation parameters are trained through deep learning to create an irradiation reference model, which can provide different irradiation adjustment parameters for different skin conditions. For example, if you want to treat age spots, according to According to the color depth and/or range of age spots, adjust the irradiation depth (the distance between the LED irradiation head and the symptom), irradiation wavelength, irradiation time, irradiation angle, and irradiation range; (3) A 3D model generator 13 for creating 3D face model data of a subject according to the color image data of the subject's face and the depth data of the face image.

In one embodiment, the photographing device 2 , the skin beautifying detection device 3 and the skin beautifying irradiation device 4 can be installed in a medical institution at the same time (but can also be installed at different positions respectively), and the remote servo device 1 is to connect the photographing equipment 2, the skin-beautifying detection equipment 3 and the skin-beautifying irradiation equipment 4 in multiple different medical institutions at the same time;

However, if multiple skin beautifying detection devices 3 and skin beautifying irradiation devices 4 all request the remote servo device 1 to perform skin quality detection and judgment with the remote servo device 1 at the same time, the response may be slow at light, and the system may crash at worst. Therefore, in order to avoid being affected by the network conditions and the number of required devices, the model is trained on the remote server equipment 1 in the edge computing mode in this case. Then, when the skin condition is detected and judged, the models are distributed at the edge (for example, the skin beauty detection equipment 3 and the skin beauty irradiation equipment 4 in different medical institutions), so that the device can judge and detect the target object and perform irradiation treatment by itself. , to maximize device availability;

The above edge computing model is a distributed computing architecture that allows enterprise applications to be closer to data sources such as IoT devices or local edge servers. Proximity to the data source delivers powerful business benefits: accelerated insights, improved response times, and higher bandwidth availability.

The photographing device 2 is capable of capturing color image data with an RGB lens, and capturing depth data of a face image with an infrared lens (in one embodiment, a 3D depth sensor composed of an infrared transmitter and an infrared CMOS camera is used) The camera is used to capture depth data), and after the shooting is completed, it can be uploaded to the remote servo device 1 through the photographing device 2 itself, or the color image data of the subject’s face and the depth data of the face image can be taken out. , and uploaded to the remote server device 1 by other electronic devices (such as a desktop computer, a notebook computer, a smart phone or a tablet computer) that can be connected to the remote server device 1 .

The skin beautifying detection device 3 at least includes at least one processor and at least one computer-readable recording medium. The computer-readable recording medium further stores computer-readable instructions and a plurality of detection application units. When the processors execute the computer-readable instructions, they can execute the detection application unit, and the detection application unit can receive a plurality of sets of facial feature model data and the 3D facial model data of the subject, and pass Multiple sets of facial feature model data are used to determine whether there is a skin condition characteristic pattern on the subject's 3D face model data, and its position and category (category is the type and severity (color depth) of the skin condition characteristics. or/and range)) are marked.

And the 4 series of skin beauty irradiation equipment includes: (1) An irradiation data adjuster 41, obtains the irradiation reference file established by the remote servo device 1, and sets the position and type of the marked skin condition characteristic pattern according to multiple sets of irradiation reference files to set an irradiation parameter data ; (2) An LED illuminates the body 42, which has: (a) a plurality of irradiation units 421, each irradiation unit 421 at least includes: (a1) One or more LED lights 4211, which are variable spectrum LED lights, which can emit light of the following wavelengths (the illumination unit 421 in this case can use light-emitting bodies with special wavelengths including LEDs but is not limited to use LEDs): (a11) Red light (622~780nm), red is the most penetrating color. The penetration rate of red light to tissues is stronger than that of other wavelengths of color light. Red light can stimulate collagen proliferation, activate fibroblasts, promote blood circulation, resist inflammation, and accelerate wound healing. (a12) Orange light (between red light and yellow light), can prevent skin dryness, increase skin elasticity, stretch wrinkles and stretch marks, delay skin aging, etc. (a13) Yellow light (575~600nm), can effectively repair skin red marks, erythema, relieve redness, redness and swelling of pigment, etc., can promote melanin metabolism, brighten skin tone, soothe sensitive skin, improve skin’s absorption of skin care products, promote Skin regeneration. (a14) Green light (between yellow and blue light), green light is the most common color in nature and is considered a neutral color. It has a balancing and stabilizing effect and can improve rough skin, wrinkles, blackheads, acne and acne caused by fatigue and mental stress. (a15) Blue light (475~500nm), blue light is a calm color and is regarded as a cool color. Blue light can inhibit oil secretion, anti-inflammatory and sterilization, eliminate acne bacillus, treat acne, scars, inhibit the growth of contusion bacillus, reduce the secretion of skin sebaceous glands, and improve acne and acne problems. (a16) Ultraviolet light (100~400nm), including: 1. UVA band, wavelength 320~400nm, also known as long-wave dark spot effect ultraviolet. It has strong penetrating power and can penetrate most transparent glass and plastic. UVA can reach the dermis layer of the skin, destroy elastic fibers and collagen fibers, and tan our skin. 2. UVB band, wavelength 275~320nm, also known as medium wave erythema effect ultraviolet. Medium penetrating power, especially strong in summer and afternoon. UVB ultraviolet rays have an erythematous effect on the human body, which can promote the metabolism of minerals and the formation of vitamin D in the body, but long-term or excessive exposure will tan the skin and cause redness and peeling. 3. UVC band, wavelength 200~275nm, also known as short-wave sterilization ultraviolet. It has the weakest penetrating ability and cannot penetrate most transparent glass and plastics. Short-wave ultraviolet rays are very harmful to the human body. Short-term exposure can burn the skin. Long-term or high-intensity exposure can also cause skin cancer. Ultraviolet germicidal lamps emit UVC short-wave ultraviolet rays. 4. UVD band, wavelength 100~200nm, also known as vacuum ultraviolet. (a2) A displacement mechanism 4212 capable of changing the position of the LED light 4211, and a motor or a similar driving mechanism is further connected inside to drive the displacement mechanism 4212 to control the forward, backward, left and right of the LED light Or/and tilt displacement, displacement amplitude, and displacement angle to conduct treatment at specific parts, specific depths, small-angle irradiation, and specific wavelengths through the LED light 4211. (a3) an infrared receiver 4213, the infrared receiver 4213 is used to assist in determining the distance between the LED light 4211 and one or more coordinates on the subject's face; (b) A control unit 422 connected to the irradiation data adjuster 41, wherein the control unit 422 can control one or more coordinates of different LED lights on the subject's face according to the irradiation parameter data Irradiate.

In addition, the LED illumination body 42 is provided with a plurality of illumination units 421, and the LED illumination body 42 can be composed of a plurality of LEDs to form a rectangular array (for example, 20*20=400 pieces, each of which can independently control the forward distance and the backward distance. , distance to the left, distance to the right, tilt displacement, displacement amplitude or/and displacement angle), and each illumination unit 421 as shown in FIG. 3C has LED lights 4211 and infrared rays on the front of each illumination unit 421 Receiver 4213.

The LED lights 4211 of the irradiation unit 421 can adjust the front and rear depths according to the 3D face model data of the subject, as shown in FIG. 4 , so that each irradiation unit 421 and the subject's face can remain The same distance, in addition, in order to respond to the position, color depth or/and range of different skin condition characteristics, the illumination depth, illumination angle illumination position or/and illumination range of the LED light can be changed, and can also be controlled through the control The unit 422 directly changes the irradiation wavelength, irradiation brightness or/and irradiation time of different LED lamps 4211, so as to strengthen the irradiation treatment of the skin on one or several coordinate positions.

The displacement mechanism 4212 can be an active mechanism or a passive mechanism. If it is an active mechanism, the control unit 422 can change the irradiation depth, irradiation angle, irradiation position or/and irradiation of the LED light through the displacement mechanism 4212. range to account for the location, colour depth or/and extent of features of different skin conditions;

However, if it is a passive mechanism, although the control unit 422 cannot control the displacement mechanism 4212 to displace or change the direction, since each LED lamp (tube) has a high friction coefficient, it is only necessary to combine a plurality of LED lamps 4211 to form an LED. After the matrix touches the face, the passive mechanism will passively expand and contract, so the face model can be obtained naturally, and the control unit 422 can control the LED light corresponding to each passive mechanism position according to the 3D face model data of the subject. Control exposure.

The following provides an example description: (1) After the 3D face model data of the subject 5 is established, it can be judged whether there is a skin condition feature pattern, and its position and category (the category includes the skin condition feature) can be determined. Type and severity (color depth or/and range)) are indicated; (2) As shown in Figure 5, after judging the presence of eczema 51 within the range of coordinates A1, A2, A3, and A4, for eczema 51, judge and provide a first irradiation parameter data (should be 2 cm away from the skin, the irradiation wavelength is 365 nm, and the irradiation time is 10 seconds); (3) As shown in Figure 5, at coordinates B1, B2, After judging the presence of age spots 52 within the range of B3 and B4, according to the age spots 52, determine and provide a second irradiation parameter data (at 1 cm from the skin, the irradiation wavelength is 280 nm, and the irradiation time is 5 seconds) (4) After that, the irradiation unit 421 (Led cylinder) corresponding to the position can be based on the depth data of the face modeling (the 3D face model data of the subject) and the aid of infrared ranging, and a plurality of corresponding positions The irradiation unit 421 (corresponding to the irradiation unit 421 of the coordinates A1, A2, A3, A4 and the coordinates B1, B2, B3, and B4) extends forward to the corresponding position, and emits light of the corresponding wavelength and time for treatment.

After the skin beautifying irradiation device 4 is irradiated, new face color image data and face image depth data can be captured again through the photographing device 2, and sent to the remote servo device 1 for updating a new one. 3D face model data, and the skin beauty detection device 3 can compare the The 3D face model data before and after irradiation, and the irradiation data adjuster is used to reset a new irradiation parameter data (for example, if a nail has eczema, it needs to be treated by irradiating ultraviolet rays with a wavelength of 350~390nm, but it is not a one-time treatment. It can be cured. After each irradiation, photos will be taken to record and analyze the degree of eczema, and compared with the previous data as the reference and formulation of the treatment plan for the next treatment).

As shown in Fig. 6, the skin beautifying detection method of this case includes: (1) 601 On a remote servo device, classifying objects according to a plurality of photos with human faces, so as to create a plurality of sets of facial features model data and irradiation reference file; (2) 602, through one or more photographing equipment, take pictures to obtain the color image data and depth data of the face image of a subject, and create the image data through the remote servo equipment 3D face model data of the subject; (3) 603 transmits multiple sets of facial feature model data and the subject's 3D face model data to a skin beautifying detection device for comparison by the skin beautifying detection device; Yes, and mark the skin condition characteristic pattern on one or more coordinates on the subject's face; (4) 604 and then set an irradiation parameter according to the marked skin condition characteristic pattern and the irradiation reference file data, so that an LED irradiating body with a plurality of irradiating units can irradiate one or a plurality of coordinates on the subject's face according to the irradiation parameter data.

The skin beautifying detection system and method thereof provided by the present invention, when compared with other conventional techniques, have the following advantages:

1. The present invention provides a system and a method thereof, which can train multiple sets of facial feature model data through deep learning, and then establish the subject through the color image data of the subject's face and the depth data of the face image. 3D facial model data, and because the response speed will not be affected by network conditions and the number of required devices when performing skin quality detection and judgment, the models are distributed on the edge (device side), so that the device It can judge, detect the target and carry out radiation therapy by itself, which will greatly increase the usability of the device. 2. In this case, by continuously collecting data and treatment response, after removing identifiable data, data integration and analysis can be carried out to find out the predisposing age and personal life background of certain diseases, which can be used as a risk prediction for future patients. 3. Through the system and method of this case, the 3D camera and the use of edge computing are used in each diagnosis and treatment process, so that the initial cloud-based model for personal skin quality detection and treatment methods can achieve dynamic optimization (Dynamic optimization). Optimization) purpose.

The present invention has been disclosed above through the above-mentioned embodiments, but it is not intended to limit the present invention. Anyone familiar with this technical field with ordinary knowledge can understand the aforementioned technical features and embodiments of the present invention without departing from the present invention. Within the spirit and scope, some changes and modifications can be made, so the scope of patent protection of the present invention shall be determined by the claims attached to this specification.

1: Remote Servo Device 11: Target Feature Builder 12: Irradiation parameter storage 13: 3D Model Builder 2: Photography equipment 3: Skin beauty testing equipment 4: Skin beauty irradiation equipment 41: Irradiation data adjuster 42: LED illumination body 421: Irradiation unit 4211: LED light 4212: Displacement mechanism 4213: Infrared Receiver 422: Control Unit 5: The subject 51: Eczema 52: Age spots A1: Coordinates A2: Coordinates A3: Coordinates A4: Coordinates B1: Coordinates B2: Coordinates B3: Coordinates B4: Coordinates

[Fig. 1] is a schematic diagram of the overall structure of the skin beautifying detection system and the method thereof of the present invention. [Fig. 2] is a schematic diagram of the structure of the remote servo device of the skin beautifying detection system and the method thereof of the present invention. [Fig. 3A] is a schematic diagram of the structure of the skin beautifying irradiation equipment of the skin beautifying detection system and the method thereof of the present invention. [FIG. 3B] is a schematic diagram of the structure of the irradiation unit of the skin beautifying detection system and the method thereof of the present invention. [FIG. 3C] It is a simple structural schematic diagram of the irradiation unit of the skin beautifying detection system and the method thereof of the present invention. [FIG. 4] It is a schematic diagram of the rectangular array layout of a plurality of irradiation units of the skin beautifying detection system and the method thereof of the present invention. [Fig. 5] It is a schematic diagram of the implementation of face modeling and skin condition feature marking of the skin beautifying detection system and method of the present invention. [Fig. 6] is a schematic flow chart of the skin beautifying detection system and the method thereof of the present invention.

1: Remote Servo Device

2: Photography equipment

3: Skin beauty testing equipment

4: Skin beauty irradiation equipment

Claims (18)

A skin beautifying detection system, comprising: A remote servo device, including: A target feature builder, which is based on a plurality of photos with human faces, and performs target classification for different skin state feature images, so as to establish a plurality of sets of facial feature model data; an irradiation parameter storage, which stores multiple sets of irradiation reference files according to the characteristic patterns of different skin conditions; a 3D model builder for creating 3D face model data of a subject according to the color image data of the subject's face and the depth data of the face image; one or more photographing equipment for photographing to obtain color image data and depth data of the subject's face image, and transmit them to the remote servo equipment; One or more skin beautifying detection devices are connected to the remote servo device for receiving multiple sets of facial feature model data and the subject's 3D facial model data, and comparing them to indicate The characteristic pattern of the skin condition on one or more coordinates of the subject's face; One or more skin beautifying irradiation devices are connected with the remote servo device and the skin beautifying detection device, and the skin beautifying irradiation device includes: an irradiation data adjuster for setting an irradiation parameter data according to a plurality of sets of irradiation reference files for a skin condition characteristic pattern on one or more coordinates of the subject's face; and An LED illuminating body has a plurality of illuminating units and a control unit connected with the illuminating data regulator, each illuminating unit at least includes an LED light, wherein the control unit can control different LEDs according to the illuminating parameter data The lamp illuminates one or a plurality of coordinates on the subject's face. The skin beautifying detection system according to claim 1, wherein the skin condition characteristic pattern can be an image of skin texture or/and skin defects. The skin beautifying detection system according to claim 2, wherein the image of the skin type is an image of normal skin type, oily skin type, dry skin type or/and combination skin type. The skin beauty detection system according to claim 2, wherein the image of the skin defect is an image of spots, wrinkles, acne, acne scars, keratinization or/and moles. The skin beautifying detection system according to claim 1, wherein the target feature builder can use a deep learning algorithm to classify targets according to different skin state feature images, so as to create multiple sets of facial feature model data. The skin beautifying detection system according to claim 1, wherein the skin beautifying detection device is capable of judging the skin condition feature patterns on one or more coordinates of the subject's face according to multiple sets of facial feature model data. category. The skin beautifying detection system according to claim 1, wherein the operations of skin condition detection and judgment can be dispersedly deployed on the skin beautifying detection device and the skin beautifying irradiation device in an edge operation mode, so that the skin beautifying detection device and the skin beautifying irradiation device The skin-beautifying irradiation device judges and detects the characteristic pattern of the skin state by itself. The skin beautifying detection system according to claim 1, wherein after the skin beautifying irradiation device is irradiated, new face color image data and face image depth data can be captured by the photographing device, and sent to the remote servo device , so as to update a new 3D face model data, and the skin beauty detection device can compare the 3D face model data of the subject before and after irradiation, and reset a new one through the irradiation data adjuster. Irradiation parameter data. The skin beautifying detection system according to claim 1, wherein the LED light for the LED illuminating the body is a variable spectrum LED light. The skin beautifying detection system according to claim 1, wherein the irradiation unit is further provided with one or more infrared receivers for assisting in judging the relationship between the LED light and one or more coordinates on the subject's face distance. The skin beauty detection system according to claim 1, wherein the control unit can change the irradiation depth, irradiation angle or/and irradiation position of the LED light. The skin beauty detection system according to claim 1, wherein the control unit can change the irradiation wavelength or/and irradiation time of different LED lamps. The skin beautifying detection system according to claim 1, wherein each irradiation unit can further change the position of the LED light through a displacement mechanism, and the displacement mechanism can be an active adjustment mechanism or a passive adjustment mechanism. A skin beautifying detection method, comprising: on a remote servo device, classifying objects according to a plurality of photos with human faces, so as to establish a plurality of sets of facial feature model data; through one or more photographing devices, Shooting to obtain color image data and depth data of a subject's face, and creating 3D face model data of the subject through the remote servo device; combining multiple sets of facial feature model data and The 3D face model data of the subject is sent to a skin beautifying detection device for comparison by the skin beautifying detection device and marking the skin state on one or more coordinates of the subject's face feature pattern. The skin beautifying detection method according to claim 14, wherein the skin condition characteristic pattern can be an image of skin skin quality or/and skin defects. The skin beauty detection method as claimed in claim 15, wherein the image of the skin texture is Image for normal, oily, dry or/and combination skin. The skin beauty detection method according to claim 15, wherein the image of the skin defect is an image of spots, wrinkles, acne, acne scars, keratinization or/and moles. The skin beautifying detection method according to claim 14, wherein the remote servo device can use a deep learning algorithm to classify objects according to different skin state characteristic images, so as to establish multiple sets of facial characteristic model data.

Images