CN113080848A

CN113080848A - Near-infrared skin detection device and method

Info

Publication number: CN113080848A
Application number: CN202110315329.7A
Authority: CN
Inventors: 金波; 段羽
Original assignee: Shenzhen Deji Medical Devices Co ltd
Current assignee: Shenzhen DJ Innovation Industry Co Ltd
Priority date: 2021-03-24
Filing date: 2021-03-24
Publication date: 2021-07-09
Anticipated expiration: 2041-03-24
Also published as: CN113080848B

Abstract

The invention relates to a near-infrared skin detection device, comprising: at least two light emitting units with different spectral bands for emitting near infrared light of different wavelengths; at least two driving units, wherein different driving units are used for driving different light-emitting units to emit light; a photosensor; and the control module is connected with each driving unit and the light receiving unit and used for controlling each driving unit to drive each light emitting unit to emit light in a time-sharing manner, receiving an electric signal converted from an optical signal which is acquired by the photoelectric sensor in a time-sharing manner and is reflected by the target to be detected and emitted by each light emitting unit, and obtaining the skin state of the target to be detected according to the light intensity of the optical signal. The invention can effectively acquire the infrared absorption condition information of different wave bands of the facial skin at lower cost and accurately detect the skin state.

Description

Near-infrared skin detection device and method

Technical Field

The invention relates to skin detection, in particular to a near-infrared skin detection device and a near-infrared skin detection method.

Background

The quality of facial skin affects the working and living conditions of people, and the main method for diagnosing facial skin is still a method for observing and hearing the facial skin, namely, an experienced physician observes the texture and color of the skin, asks about the relevant conditions of patients and diagnoses the disease condition of the skin according to the accumulated expert experience. Some facial skin diagnostic devices have been developed in recent years, but some of these devices are not able to detect features of the inner layers of the skin and some are expensive.

Disclosure of Invention

Accordingly, there is a need for a low-cost near-infrared skin detection device that can detect the condition of the inner layer of the skin.

A near-infrared light skin detection device, comprising: at least two light emitting units with different spectral bands for emitting near infrared light of different wavelengths; at least two driving units, wherein different driving units are used for driving different light-emitting units to emit light; a photosensor; and the control module is connected with each driving unit and the light receiving unit and used for controlling each driving unit to drive each light emitting unit to emit light in a time-sharing manner, receiving an electric signal converted from an optical signal which is acquired by the photoelectric sensor in a time-sharing manner and is reflected by the target to be detected and emitted by each light emitting unit, and obtaining the skin state of the target to be detected according to the light intensity of the optical signal.

In one embodiment, the control module obtains the skin state of the target to be detected, and the method comprises the step of obtaining the sebum content state and the moisture state through a sebum quantity classifier and a moisture classifier by using a support vector machine.

In one embodiment, each of the light emitting units includes a near-infrared light emitting diode.

In one embodiment, the photosensor comprises an indium gallium arsenide photodiode.

In one embodiment, the control module controls each driving unit by outputting a pulse width modulation signal, and the near-infrared light skin detection device further includes alternating pulse to direct current circuits equal in number to the driving units, each alternating pulse to direct current voltage circuit being configured to convert the pulse width modulation signal output by the control module into a direct current signal.

In one embodiment, the number of the light emitting units and the number of the driving units are 4, and the light emitting wavelengths of the 4 light emitting units are 1050nm,1200nm,1300nm and 1550nm respectively.

In one embodiment, the optical module further comprises a transmitting-receiving coaxial optical path module, and the transmitting-receiving coaxial optical path module sequentially comprises, on an optical path from each light-emitting unit to the photoelectric sensor: a light-combining lens; the light combining diaphragm is positioned on the focus of the light combining lens; a first condenser lens; a half mirror; the second light is transmitted.

The near-infrared skin detection device uses the plurality of single-waveband near-infrared light emitting units with different wavelengths to emit near-infrared light waves in a time-sharing manner so as to realize the light emitting detection of different wavelengths in the near-infrared waveband, and the cost is lower compared with other realization schemes (such as an NIR long-wave spectrometer) of multi-wavelength light emitting detection. As the infrared absorption wavelengths and intensities of different group molecules (including methyl, methylene and benzene rings) are different, the multi-wavelength near infrared light is suitable for detecting organic substances of hydrocarbon groups, and the light intensity of the near infrared light absorbed by each wave band is acquired by the photoelectric sensor in a time-sharing manner, so that the infrared absorption condition information of different wave bands of facial skin can be effectively acquired, and the skin state can be accurately detected.

It is also necessary to provide a near-infrared skin detection method, comprising: near infrared light with different spectral bands is emitted in a time-sharing manner through at least two light-emitting units with different light-emitting bands; collecting near infrared light reflected light of each light-emitting unit reflected by a target to be detected in a time-sharing manner; and obtaining the skin state of the target to be detected according to the light intensity of the near infrared reflected light acquired in a time-sharing manner.

In one embodiment, the step of obtaining the skin state of the target to be measured according to the light intensity of the near-infrared reflected light collected in a time-sharing manner includes: differentiating the measured value of the light intensity of each current waveband with the reference value of the corresponding waveband to obtain differential data to form a feature vector; and inputting the characteristic vectors into a sebum quantity classifier and a moisture classifier by using a support vector machine mode to obtain a sebum content state and a moisture state.

In one embodiment, the step of emitting the near-infrared light of different spectral bands in a time-sharing manner by at least two light-emitting units with different light-emitting bands and the step of collecting the near-infrared light reflected light of each light-emitting unit reflected by the target to be detected in a time-sharing manner are performed after the light emission and reflected light collection of one light-emitting unit is completed, and then the light emission and reflected light collection of the next light-emitting unit is performed.

It is also necessary to provide a computer readable storage medium having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method of any of the above embodiments.

It is also necessary to provide a computer device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of the method according to any of the preceding embodiments when executing the computer program.

Drawings

For a better understanding of the description and/or illustration of embodiments and/or examples of those inventions disclosed herein, reference may be made to one or more of the drawings. The additional details or examples used to describe the figures should not be considered as limiting the scope of any of the disclosed inventions, the presently described embodiments and/or examples, and the presently understood best modes of these inventions.

FIG. 1 is a schematic diagram of a near-infrared skin detection device according to an embodiment;

FIG. 2 is a flow diagram of an interrupt service routine for an internal timer of a microprocessor in one embodiment;

FIG. 3 is a flowchart of a master process for an internal timer of a microprocessor in one embodiment;

fig. 4a is a schematic circuit diagram of the core digital circuit portion of the control module 111, and fig. 4b is a schematic circuit diagram of 4 PWM signal output isolation circuits and 4 driving units;

FIG. 5 is a schematic optical path diagram of a transmit receive coax optical path module according to an embodiment;

FIG. 6 is a flowchart of near-infrared skin detection device calibration and measurement in one embodiment;

FIG. 7 is a schematic diagram of an SVM sebum quantity classifier according to an embodiment;

FIG. 8 is a schematic diagram of an SVM moisture classifier in one embodiment;

FIG. 9 is a flowchart illustrating a method for near-infrared skin detection according to an embodiment.

Detailed Description

To facilitate an understanding of the invention, the invention will now be described more fully with reference to the accompanying drawings. Preferred embodiments of the present invention are shown in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.

It will be understood that when an element or layer is referred to as being "on," "adjacent to," "connected to," or "coupled to" other elements or layers, it can be directly on, adjacent to, connected or coupled to the other elements or layers or intervening elements or layers may be present. In contrast, when an element is referred to as being "directly on," "directly adjacent to," "directly connected to" or "directly coupled to" other elements or layers, there are no intervening elements or layers present. It will be understood that, although the terms first, second, third, etc. may be used to describe various elements, components, regions, layers and/or sections, these elements, components, regions, layers and/or sections should not be limited by these terms. These terms are only used to distinguish one element, component, region, layer or section from another element, component, region, layer or section. Thus, a first element, component, region, layer or section discussed below could be termed a second element, component, region, layer or section without departing from the teachings of the present invention.

Spatial relational terms such as "under," "below," "under," "above," "over," and the like may be used herein for convenience in describing the relationship of one element or feature to another element or feature as illustrated in the figures. It will be understood that the spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures. For example, if the device in the figures is turned over, then elements or features described as "below" or "beneath" other elements or features would then be oriented "above" the other elements or features. Thus, the exemplary terms "under" and "under" can encompass both an orientation of above and below. The device may be otherwise oriented (rotated 90 degrees or at other orientations) and the spatial descriptors used herein interpreted accordingly.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. As used herein, the term "and/or" includes any and all combinations of the associated listed items.

An exemplary facial skin detector uses the principle of optical photography, and irradiates the skin with illumination light from blue band to near infrared, and then uses a camera to capture the image of the skin surface, by which the characteristics of the skin surface, such as surface skin defects like color spots, acne, and inflammation, can be detected. However, the symptoms such as color spots and acne are only the signs of skin diseases, and the treatment is only performed according to the signs, so that the effect is not good and the recurrence probability is high. In terms of physiological structure, the skin is divided into an epidermal layer, a dermal layer and a subcutaneous layer from outside to inside, the sebum content of the inner skin of the dermal layer and the subcutaneous layer and the quantity and distribution of skin moisture have important influence on skin inflammation, and the characteristics of the inner skin layer cannot be detected by using the current imaging skin detector. Some research and development institutions and individuals develop a skin infrared detector with a single-band wavelength of 650-1050nm, which still can only detect information of skin superficial layer, has limited penetrating power and cannot detect information of skin sebum and water content.

The infrared ray with the wavelength band of 900-2500nm belongs to a NIR (Near infrared) -spectrum long-wave interval, and infrared absorption wavelengths and intensities of different group molecules (including methyl, methylene and benzene rings) are different in the interval, so that the wavelength is suitable for detecting organic substances with hydrocarbon radicals, and the method can be used for detecting the amount of grease and moisture in the inner layer of the skin (including below the surface layer of the skin). The NIR long-wave spectrometer can effectively detect the spectral absorption rate of each wavelength of the waveband by using an optical light splitting principle. However, the NIR long-wave spectrometer is expensive in cost, an external light source is required, data is transmitted to a computer by using an external bus interface after measurement is completed, and then is analyzed by using special software, so that the operation is complex, and the test efficiency is low, thereby limiting the popularization and application of the NIR long-wave spectrometer.

The application provides a near-infrared skin detection device, which comprises a control module, a photoelectric sensor, a plurality of light emitting units and a plurality of driving units. The light-emitting units comprise at least two types, and the light-emitting units of different types have different light-emitting wave bands which are near-infrared wave bands. Different driving units are used for driving different light emitting units to emit light, and the driving units can correspond to the light emitting units one by one (the number of the driving units is equal). The control module is connected with each driving unit and each light receiving unit. When the near-infrared skin detection device works, the control module controls each driving unit to drive each light-emitting unit to emit light in a time-sharing mode, a target to be detected (such as human face skin) reflects light waves emitted by each light-emitting unit, the control module controls the photoelectric sensor to receive reflected light signals in a time-sharing mode, the photoelectric sensor converts the light signals into electric signals corresponding to light fronts, and the control module obtains the skin state of the target to be detected according to the electric signals.

The near-infrared skin detection device uses the plurality of single-waveband near-infrared light emitting units with different wavelengths to emit near-infrared light waves in a time-sharing manner so as to realize the light emitting detection of different wavelengths in the near-infrared waveband, and the cost is lower compared with other realization schemes (such as an NIR long-wave spectrometer) of multi-wavelength light emitting detection. As the infrared absorption wavelengths and the intensities of different group molecules (including methyl, methylene and benzene rings) are different, the multi-wavelength near infrared light is suitable for detecting organic substances of hydrocarbon groups, and the light intensity of the near infrared light absorbed by each wave band is collected by the photoelectric sensor in a time-sharing manner, so that the infrared absorption condition information of different wave bands of facial skin can be effectively obtained, the skin state can be accurately detected, and the expensive image sensor device of the near infrared wave band is avoided.

Fig. 1 is a schematic structural diagram of a near-infrared skin detection device in an embodiment, in the embodiment, each of the light emitting units (not shown in fig. 1) and the driving units is 4 (the first driving unit 103, the second driving unit 104, the third driving unit 105, and the fourth driving unit 106), and each of the driving units is electrically connected to one of the light emitting units. In one embodiment of the present application, a near infrared Light Emitting Diode (LED) is used as the light emitting unit, and light emission wavelengths of the 4 light emitting units are 1050nm,1200nm,1300nm, and 1550nm, respectively.

In one embodiment of the present application, the photo sensor 101 employs an indium gallium arsenide photodiode in order to detect the near infrared 900-1700nm wavelength light reflected from the skin surface to be measured.

In one embodiment of the present application, a signal conditioning circuit is further connected between the photosensor 101 and the control module 111. The signal conditioning circuit is used for converting the weak photocurrent to a voltage of 0-3V and outputting the converted weak photocurrent to the control module 111.

In one embodiment of the present application, the control module 111 is a 32bit industrial Microprocessor (MCU). The microprocessor can use an ARM core controller with the model number of STM32F103, and the microprocessor has various resources including a timer, a serial port transceiver and other digital IO input and output ports. The microprocessor converts the voltage output by the signal conditioning circuit into analog voltage values related to the photocurrent through an internal AD converter.

In an embodiment of the application, the near-infrared skin detection device further includes an upper computer communication interface and a human-computer interaction interface, and the upper computer communication interface is connected with the human-computer interaction interface. The user can send out corresponding commands such as detection, calibration, OK, startup/shutdown and setting through the man-machine interaction interface, and the microprocessor carries out corresponding processing after receiving the commands.

In one embodiment of the present application, the control module 111 controls each driving unit by outputting a Pulse Width Modulation (PWM) signal to adjust the light emitting intensity of each light emitting unit. The near-infrared skin detection device further comprises PWM signal output isolation circuits with the same number as the driving units, and the PWM signal output isolation circuits are connected between the driving units and the control module 111. The PWM signal output isolation circuit may include an alternating pulse to direct current circuit for converting the PWM signal output by the control module 111 into a direct current signal.

Specifically, the control module 111 may output a PWM signal with a duty ratio varying according to a sine wave for a variable time length under the control of an external command (through a human-computer interaction interface), and each driving unit controls the light emitting intensity of the corresponding light emitting unit according to the duty ratio of the PWM signal. The microprocessor can use a high-precision timer to generate four paths of independent PWM signals, and the current of the light source is controlled by changing the pulse time width of each pulse period, so that the stepless continuous control of the light source is realized.

In one embodiment of the present application, the PWM signal is generated by using an internal timer TIM1 of a microprocessor, and using an auto-reload register (ARR) and a prescaler register PSC of the microprocessor to configure a frequency fc of the PWM, the frequency value of fc is represented by formula (1):

where f is the system clock frequency of the microprocessor, ARR is the register value of the auto-reload register, and PSC is the register value of the pre-divide register.

The internal timer TIM1 of the microprocessor has four positive output IO ports defined on the PA8, PA9, PA10 and PA11 pins of the processor, respectively, each of which outputs a PWM signal with a duty cycle controlled by a capture-compare register (CCR). After the PWM output function is started, the TIM1 starts counting up or down, when the counted number reaches the CCR value, the IO port outputs a low level, otherwise, the IO port outputs a high level, and when the TIM1 counts up to the preset ARR value, the timer count returns to zero again. The TIM1 interrupt service function of the microprocessor is turned on before outputting the PWM signal, and the number of PWM waves is counted in the interrupt service routine. The master control program of the TIM1 is shown in fig. 2 and 3, respectively, and operates as follows:

1) the sel variable is used to mark which channel is enabled and the FIN1, FIN2, FIN3, FIN4 are used to mark which channel PWM signal output ends. FIN1 is equal to 1, FIN2 is equal to 0, FIN3 is equal to 0, FIN4 is equal to 0 and ends at channel 1 output, FIN1 is equal to 0, FIN2 is equal to 1, FIN3 is equal to 0, FIN4 is equal to 0 and ends at channel 2 output, FIN1 is equal to 0, FIN2 is equal to 0, FIN3 is equal to 1, FIN4 is equal to 0 and ends at channel 3 output, FIN1 is equal to 0, FIN2 is equal to 0, FIN3 is equal to 0, and FIN4 is equal to 1 and ends at channel 4 output.

2) As shown in the flow of fig. 3, in the main control program, first, parameters such as a time length, a pulse frequency, and a duty ratio of PWM output are set, then, the channel 1 is opened, a PWM signal is output from an IO pin addressing the PA8, then, a variable sel is set to 1, the end of channel 1 output is waited, and FIN1 is used for detecting the end of channel 1 output; the TIM1 interrupt service routine shown in fig. 2 detects that the variable sel is 1, i.e., counts the channel 1 output pulses, and when the count reaches a set number, turns off the PA8 output, setting FIN1 to 1, so that the master detects that the channel 1 output is over. After detecting that the output of the channel 1 is finished, the main control program starts the channel 2 after a necessary delay, that is, the channel 2 is output from the IO pin of the PA9, sel is set to 2 after the channel 2 is started, the interrupt service program counts the PWM signal of the channel 2 after detecting that the variable sel is 2, FIN2 is set to 1, FIN1 is set to 0, FIN3 is set to 0, and FIN4 is set to 0 after the output of the channel 2 is finished, so as to notify the main control program that the output is finished. The PWM output mechanism of the

subsequent channels

3 and 4 is similar to that of the

channels

1 and 2.

3) And the main control program also starts the ADC for sampling before selecting different channels for output, and when the PWM signal output is finished, the main control program closes the ADC for sampling and records the sampling data into different channel storage variables.

Fig. 4a is a circuit schematic diagram of the core digital circuit part of the control module 111, and fig. 4b is a circuit schematic diagram of 4 PWM signal output isolation circuits and 4 driving units. The 4-way PWM signal is converted into an analog voltage using the circuit shown in fig. 4b to control the current flowing through the near-infrared LED devices (D1, D2, D3, and D4). In fig. 4a, each output pin (PA8, PA9, PA10 and PA11) of the PWM signal is connected to the analog signal conversion circuit of fig. 4b, each analog signal conversion circuit uses a second-order rc filter circuit to convert the PWM digital pulse into a dc voltage, the operational amplifier LM324 compares the dc voltage with the feedback voltage of the near-infrared LED (the voltage of LED cathodes b1, b2, b2 and b 4), and adjusts the conduction degree of the transistors Q1, Q2, Q3 and Q4 according to the comparison structure, thereby controlling the current of the near-infrared LED (D1, D2, D3 and D4). In one embodiment of the present application, the near infrared LED is rated at 20 milliamps, with feedback resistors Rs1, Rs2, Rs3 Rs4 set at 75 ohms. The current of the LED can be adjusted by adjusting the duty ratio of the output PWM signal by using the microprocessor, so that the luminous intensity of the LED is adjusted.

In an embodiment of the application, the near-infrared light skin detection device further comprises a transmitting and receiving coaxial optical path module, so that integrated infrared light projection and receiving are realized. Referring to fig. 5, the transmitting-receiving coaxial optical path module sequentially includes, on an optical path from the light emitting unit to the photosensor: a light combining lens 502, a light combining aperture 507, a first condenser lens 503, a half mirror 504, and a second condenser lens 506. In one embodiment of the present application, the first condenser lens 503 is a collimating focus lens.

In the embodiment shown in fig. 5, the near-infrared skin detection device includes a multiband LED light source board 501, and the multiband LED light source board 501 includes infrared LED lamp beads with 4 wavelengths, which are 1050nm,1200nm,1300nm, and 1550nm, respectively. Further, the LED lamp beads are arranged on a white circuit board in a 2x2 matrix mode. The infrared light emitted by the LED is projected onto the skin of the human face through the optical path of the light combining lens 502, the light combining diaphragm 507, the first condenser lens 503 and the half mirror 504. The back focal length of the light combining lens 502 is 16mm, and the light combining diaphragm 507 is located at the back focal point of the light combining lens 502. The first condenser lens 503 adopts a long focal length 25mm lens, and the first condenser lens 503 is placed at a position fifty millimeters away from the light combining diaphragm 507. The half mirror 504 projects the near infrared light focused by the first condenser lens 503. After the near infrared light is projected to the skin, the absorption coefficient of the near infrared light is influenced by the skin sebum and the moisture content, and when the sebum content is higher or the moisture in the skin is more, the infrared light is absorbed more and the reflection intensity is low. Near infrared light reflected by skin passes through the half-reflecting mirror 504, is converged by the second condenser lens 506 and then irradiates on a receiving target surface of the photoelectric sensor 101, the illumination intensity is in direct proportion to the photocurrent of the photoelectric effect in the photoelectric sensor 101, the photocurrent is converted into an analog voltage value of 0-3V through the signal conditioning circuit, and then the analog voltage is input into an ADC (analog to digital converter) of the microprocessor and converted into digital information.

The emitting and receiving of the near infrared light adopt a coaxial light path structure, and the device has the advantages of compact structure, high space utilization rate and convenient installation and debugging of optical components.

In one embodiment of the present application, the second condenser lens 506 is a convex lens with a focal length of 13mm, and is spaced 25mm from the center of the half mirror 504. The rear end of the second condenser lens 506 is 17mm from the light-sensing surface of the photosensor 101. For sensing light waves in the infrared 900-1700nm band, the photosensor 101 may use near-infrared photodiode devices of the InGaAs type.

In one embodiment of the present application, the control module 111 uses a support vector machine to obtain the sebum content status and the moisture status through a sebum amount classifier and a moisture classifier. Specifically, the operation process of the near-infrared skin detection device comprises two procedures of calibration and measurement. FIG. 6 is a flowchart illustrating calibration and measurement of a near-infrared skin detection device according to an embodiment. In the process of calibration, near infrared ray emission, sampling and filtering processing are carried out, and then the received near infrared ray intensity is written into a database to be used as a reference value for comparison. During the 'measuring' operation, infrared emission, sampling, filtering processing, reading out the reference value of comparison are performed, and then the actual value and the reference value of the current measurement are differentiated. The actual measured values of the infrared light waves of 4 wave bands of 1050nm,1200nm,1300nm and 1550nm are differentiated from the reference values of the actual measured values, 4 differential data form a characteristic vector, the characteristic vector is judged by machine learning, and the state of the skin is output.

In one embodiment of the present application, the microprocessor of the control module 111 uses a low power 32-bit processor, so the classification algorithm uses a Support Vector Machine (SVM) with a small amount of computation. The input to the sebum classifier and the moisture classifier was normalized using the four-way spectral sensor response values a1, a2, a3, a4 (i.e., the microprocessor ADC converter time-division digitized output values mentioned above). The normalization formula can be represented by equation (2):

in the formula a_nN is 1,2,3, 4; a is the normalized coefficient, i.e., the difference between the maximum and minimum values of each input (a1-a 4).

The sebum amount classifier is shown in FIG. 7. For detecting sebum in inner layers of skin, a 3-layer SVM (support vector machine) structure is used, and each support vector machine adopts a two-classification structure. The first layer is classified as "extra sebum" and "less sebum", i.e., { extra sebum, less sebum }, and the second layer is classified as "much sebum", "more sebum", and "less sebum" for "more sebum", and "less sebum". The third layer refines the "more sebum" category into "more sebum" and "medium sebum". In summary, the final classifications for sebum on the inner skin are { "too much sebum", "medium sebum", "less sebum", "too little sebum" }, each layer classification is encoded as {1, -1}, as shown in fig. 7, with the left side of the branch being encoded as 1 and the right side as-1.

The moisture classifier is used for detecting the dryness and wetness of skin and predicting a moisture content index, and adopts a 3-layer SVM (support vector machine) structure as shown in FIG. 8, and each SVM adopts a two-classification structure. The first layer is classified as "high moisture" and "low moisture", i.e., { high moisture, low moisture }, and the second layer is classified as "high moisture" and "high moisture" for "high moisture", and "low moisture" for "low moisture", and "low moisture". The third layer refines the "moisture high" category into "moisture high" and "moisture medium". In summary, the skin inner layer moisture is finally classified as { "much moisture", "medium moisture", "little moisture" }, each layer classification is encoded as {1, -1}, and as shown in fig. 8, the branches are encoded as 1 on the left and-1 on the right.

Through the use of the SVM classifier, the skin state is classified, and guidance is provided for subsequent treatment. The diagnosis and treatment effect is better when the diagnosis and treatment instrument is used together with an image normal facial skin diagnosis instrument.

Through the SVM classifier, the skin class code is as follows, the first bit is the detection item code, 0 is sebum, and 1 bit is moisture.

Sebum coding:

{0, 1, 1, 1} sebum is abundant

{0, 1, 0, 1} sebum is abundant

{0, 1, 0, 0} sebum, etc

{0, 0, 1, 0} sebum-reducing

{0, 0, 0, 0} sebum is very little

Skin dryness category code:

{1, 1, 1, 1} moisture is high

{1, 1, 0, 1} moisture content is high

{1, 1, 0, 0} moisture, etc

{1, 0, 1, 0} moisture is less

{1, 0, 0, 0} moisture is very little

The near-infrared skin detection device uses four-bit coding values to store the category attribute of the skin, and can transmit the category attribute to an upper computer by using a serial port.

The technical scheme is that a machine learning method is used for processing the skin characteristic classification problem, a support vector machine classifier of two indexes of skin fat and water is designed, a binary tree method is used for multi-classification of the support vector machine, a machine learning support vector machine classification algorithm is used for processing multi-channel infrared ray acquisition data, classification attributes after skin classification are coded by using a binary system, a classifier model can be quickly established, consumed hardware resources are few, and classification judgment can be economically carried out on the skin state.

The application correspondingly provides a near-infrared skin detection method. FIG. 9 is a flowchart of a near-infrared skin detection method according to an embodiment, including the following steps:

s910, emitting near infrared light of different spectral bands in a time-sharing manner through at least two light-emitting units with different light-emitting bands.

In one embodiment of the present application, the number of the light emitting units is 4, and the light emitting wavelengths of the 4 light emitting units are 1050nm,1200nm,1300nm and 1550nm, respectively.

S920, collecting the near infrared light reflected light of each light-emitting unit reflected by the target to be detected in a time-sharing manner.

In one embodiment of the present application, an indium gallium arsenide photodiode is used to collect the reflected light.

And S930, obtaining the skin state of the target to be detected according to the light intensity of the near infrared reflected light collected in a time-sharing manner.

The skin condition obtained by the near-infrared skin detection method can be used as an intermediate result to assist a doctor in diagnosing skin inflammation.

In one embodiment of the present application, step S930 includes: and differentiating the measured value of the light intensity of each current waveband with the reference value of the corresponding waveband to obtain differential data to form a characteristic vector. And inputting the characteristic vectors into a sebum quantity classifier and a moisture classifier by using a support vector machine mode to obtain a sebum content state and a moisture state.

In an embodiment of the present application, steps S910 and S920 are performed after completing the collection of the light emitting and reflected light of one light emitting unit, and then performing the collection of the light emitting and reflected light of the next light emitting unit. Reference may be made in particular to fig. 6.

In one embodiment of the present application, a near-infrared skin detection method includes a calibration procedure and a measurement procedure. The calibration process includes steps S910 and S920, and then the electric signals corresponding to the collected reflected light are amplified, filtered and subjected to analog-to-digital conversion, and the obtained light intensity is written into a database as a reference value for comparison. The measurement process also includes steps S910 and S920, and then the electric signals corresponding to the collected reflected light are amplified, filtered and subjected to analog-to-digital conversion, and the obtained current light intensity value and the reference value are differentiated to form a feature vector from the obtained difference data.

In one embodiment of the present application, the method further comprises the step of normalizing the reference value and the current light intensity value. The normalization process may be performed using equation (2). The construction of the sebum classifier and the moisture classifier can be referred to fig. 7 and 8.

It should be understood that, although the steps in the flowcharts of the present application are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flow chart of the present application may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the steps or the stages is not necessarily performed in sequence, but may be performed alternately or alternately with other steps or at least a part of the steps or the stages in other steps.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.

The present application also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method according to any of the embodiments described above.

The present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method according to any of the foregoing embodiments when executing the computer program.

In the description herein, references to the description of "some embodiments," "other embodiments," "desired embodiments," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, a schematic description of the above terminology may not necessarily refer to the same embodiment or example.

The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A near-infrared skin detection device, comprising:

at least two light emitting units with different spectral bands for emitting near infrared light of different wavelengths;

at least two driving units, wherein different driving units are used for driving different light-emitting units to emit light;

a photosensor;

and the control module is connected with each driving unit and the light receiving unit and used for controlling each driving unit to drive each light emitting unit to emit light in a time-sharing manner, receiving an electric signal converted from an optical signal which is acquired by the photoelectric sensor in a time-sharing manner and is reflected by the target to be detected and emitted by each light emitting unit, and obtaining the skin state of the target to be detected according to the light intensity of the optical signal.

2. The near-infrared skin detection device according to claim 1, wherein the control module obtains the skin state of the target to be detected, and comprises obtaining the sebum content state and the moisture state by a sebum quantity classifier and a moisture classifier by using a support vector machine.

3. The near-infrared skin detection device of claim 1, wherein each of the light emitting units comprises a near-infrared light emitting diode.

4. The near-infrared light skin detection device of claim 1, wherein the photosensor comprises an indium gallium arsenide photodiode.

5. The near-infrared light skin detection device according to claim 1, wherein the control module controls each of the driving units by outputting a pulse width modulation signal, the near-infrared light skin detection device further comprises alternating pulse to direct current (AC to DC) circuits equal in number to the driving units, each of the AC to DC voltage circuits is configured to convert the pulse width modulation signal output by the control module into a direct current signal.

6. The near-infrared skin detection device according to claim 1, wherein the number of the light emitting units and the number of the driving units are 4, and the light emitting wavelengths of the 4 light emitting units are 1050nm,1200nm,1300nm and 1550nm, respectively.

7. The near-infrared skin detection device according to claim 1, further comprising a transmitting-receiving coaxial optical path module, the transmitting-receiving coaxial optical path module comprising in order on an optical path from each of the light emitting units to the photosensor:

a light-combining lens;

the light combining diaphragm is positioned on the focus of the light combining lens;

a first condenser lens;

a half mirror;

the second light is transmitted.

8. A near-infrared skin detection method, comprising:

near infrared light with different spectral bands is emitted in a time-sharing manner through at least two light-emitting units with different light-emitting bands;

collecting near infrared light reflected light of each light-emitting unit reflected by a target to be detected in a time-sharing manner;

and obtaining the skin state of the target to be detected according to the light intensity of the near infrared reflected light acquired in a time-sharing manner.

9. The near-infrared skin detection method according to claim 8, wherein the step of obtaining the skin state of the target to be detected based on the light intensity of the near-infrared reflected light collected in a time-sharing manner comprises:

differentiating the measured value of the light intensity of each current waveband with the reference value of the corresponding waveband to obtain differential data to form a feature vector;

and inputting the characteristic vectors into a sebum quantity classifier and a moisture classifier by using a support vector machine mode to obtain a sebum content state and a moisture state.

10. The near-infrared skin detection method of claim 8, wherein the step of emitting the near-infrared light of different spectral bands in a time-sharing manner by at least two light-emitting units with different light-emitting bands and the step of collecting the near-infrared light reflected light of each light-emitting unit reflected by the target to be detected in a time-sharing manner are performed after the light emission and the reflected light collection of one light-emitting unit are completed, and then the light emission and the reflected light collection of the next light-emitting unit are performed.