TWI868041B - Skin multidimensional sensitive status data assessment system and assessment method thereof - Google Patents

Abstract

A skin multi-dimensional sensitive status data assessment system that assesses skin within a predetermined time after irritation, including: an image photography module is configured to capture an image of the skin to generate a captured image; a filtering module is configured to filter the captured image to generate a filtered image; an image cutting calculation module is configured to divide the filtered image into at least 16 sub-pictures and calculate the average value of each sub-pictures; a data recording and graphing module is configured to record the average value of each sub-pictures and draw a plurality of corresponding waveform graphs; an algorithm module is configured to execute a skin sensitive status assessment parameter algorithm base on the date of the at least 16 sub-pictures to obtain a discrete parameter; and an assessment module is configured to confirm that the captured image corresponds to sensitive skin or stable skin according to the relationship between the discrete parameter and a threshold. Thus, the present invention can be used to assess skin sensitivity and solve the problem of inaccurate assessment results and discomfort detection process. In addition, a multi-dimensional sensitive status data assessment method is also provided.

Description

Skin multi-dimensional sensitive state data evaluation system and evaluation method

The present invention relates to a skin state evaluation system, in particular to a skin multi-dimensional sensitive state data evaluation system and an evaluation method thereof.

The skin is the largest organ in the human body. It can block foreign invasion to protect various tissues and organs in the body from physical, chemical and pathogenic microorganisms. The skin can also retain moisture and provide warmth, barrier and sensory functions. The elasticity of the skin affects the skin's protective function, and the skin's sensitivity also has an important impact on the skin's protective function.

For example, if your skin becomes itchy or even develops rashes due to changes in temperature or humidity, or you feel uncomfortable after using skin care products, you may have sensitive skin. If people with sensitive skin do not take good care of their skin, they may have damaged their skin layer without realizing it.

Therefore, it is necessary to assess the sensitivity of the skin so that appropriate treatment and care can be given based on the assessment results.

In the prior art, the assessment of skin sensitivity usually relies on subjective assessment methods (e.g., visual observation) or invasive testing methods. However, these methods usually lead to inaccurate assessment results or uncomfortable testing processes.

In order to solve the above problems, the purpose of the present invention is to provide a skin multi-dimensional sensitivity state data evaluation system, which is to evaluate the skin within a predetermined time after being stimulated. The skin multi-dimensional sensitivity state data evaluation system includes: an image photography module, which is configured to capture the image of the skin to generate a captured image; a filtering module, which is connected to the image photography module and is configured to filter the captured image to generate a filtered image; an image cutting calculation module, which is connected to the filtering module and is configured to divide the filtered image into at least 16 sub-pictures, and and calculates the average value of each sub-screen; a data recording and drawing module is connected to the image cutting calculation module and is configured to record the average value of each sub-screen and draw it into a plurality of corresponding waveform graphs; a calculation module is connected to the image cutting calculation module and the data recording and drawing module and is configured to execute a skin sensitivity state evaluation parameter algorithm based on the data of at least 16 sub-screens to obtain discrete parameters; and an evaluation module is connected to the calculation module and is configured to confirm whether the captured image corresponds to sensitive skin or stable skin according to the relationship between the discrete parameters and the threshold value.

In some embodiments, a processing module is further included, which is connected to the image cutting calculation module, the data recording and mapping module, and the calculation module, and the processing module is configured to smooth the data of each sub-screen.

In some embodiments, the filtering process filters optical signals in the wavelength range of 400nm to 1000nm.

In some embodiments, the skin sensitivity status assessment parameter algorithm is: in, is the discrete parameter, 𝑆 is the total signal of photoplethysmography (PPG), is the peak distance of a specific PPG signal, n is the number of other PPG signals closest to the specific PPG signal, is the peak distance that is closest to a specific PPG signal.

In some embodiments, when the dispersion parameter is greater than or equal to a threshold, the captured image is confirmed to correspond to sensitive skin; and when the dispersion parameter is less than the threshold, the captured image is confirmed to correspond to stable skin.

In order to solve the above problems, the purpose of the present invention is to provide a method for evaluating the multi-dimensional sensitivity state data of the skin, which is to evaluate the skin within a predetermined time after being stimulated. The method for evaluating the multi-dimensional sensitivity state data of the skin includes the following steps: capturing an image of the skin through an image photography module to generate a captured image; filtering the captured image through a filtering module to generate a filtered image; and segmenting the filtered image through an image cutting calculation module. The image is divided into at least 16 sub-screens, and an average value of each sub-screen is calculated; the average value of each sub-screen is recorded through a data recording and drawing module, and is plotted into a plurality of corresponding waveform graphs; a skin sensitivity state evaluation parameter algorithm is executed through a calculation module based on data of at least 16 sub-screens to obtain discrete parameters; and a captured image is confirmed to correspond to sensitive skin or stable skin according to a relationship between the discrete parameters and a threshold value through an evaluation module.

In summary, the skin multi-dimensional sensitivity state data assessment system and assessment method of the present invention improves the quality of image recognition, analysis, and processing by capturing skin image data and filtering the image data, thereby improving the accuracy and reliability of skin sensitivity state assessment, thereby solving the problem of inaccurate assessment results and uncomfortable detection process in known technologies.

Please refer to FIG. 1, which is a block diagram showing a skin multi-dimensional sensitivity state data evaluation system according to an embodiment of the present invention. The skin multi-dimensional sensitivity state data evaluation system 100 is for evaluating the skin within a predetermined time after being stimulated. The skin multi-dimensional sensitivity state data evaluation system 100 includes: an image photography module 10, a filtering module 12, an image cutting calculation module 14, a data recording and mapping module 16, a calculation module 18, and an evaluation module 20.

The image photography module 10 is configured to capture the image of the skin to generate a captured image, and the captured image may include image data of the skin before and/or after the skin is stimulated. The image data may be two-dimensional and/or three-dimensional image data. The aforementioned skin refers to the skin of a fixed area (e.g., the back of the hand, face, legs, chest, etc.), and the image photography module 10 is operated to capture the image of the skin in the fixed area. The aforementioned capturing method may be through recording or photographing. For example, the image of the skin within a predetermined time is captured by recording, or the image of the skin within a predetermined time is captured by photographing. The aforementioned predetermined time may be designed to different time values according to different sources of stimulation (e.g., temperature, humidity, skin care products, etc.). In this way, it is possible to observe and identify whether the skin has a change in sensitivity due to the stimulation of temperature, humidity, or skin care products within a predetermined time. For example, when the stimulus source is temperature, the predetermined time may be 3 minutes to 5 minutes; when the stimulus source is humidity, the predetermined time may be 10 minutes to 16 minutes; when the stimulus source is skin care products, the predetermined time may be 30 minutes to 2 hours. The time value of the predetermined time may be determined based on the experimental results of big data. In addition, the way the image camera module 10 captures the image of the skin is a non-invasive detection method. Therefore, the discomfort of the embodiment of the present invention is very low compared to the conventional technology, or even no discomfort.

The filter module 12 is connected to the image capturing module 10. The filter module 12 is configured to perform filtering processing on the captured image to generate a filtered image. For example, unnecessary noise and ambient light signals (e.g., light signals with a wavelength range of 400nm to 1000nm) in the captured image are filtered out to improve the image quality corresponding to the skin, thereby improving the accuracy of subsequent image data processing.

The image cutting calculation module 14 is connected to the filtering module. The image cutting calculation module 14 is configured to divide the filtered image into at least 16 sub-frames and calculate the average value of each sub-frame. For example, the image cutting calculation module 14 divides the filtered image into a plurality of sub-frames according to a predetermined window, calculates the average value of each sub-frame, and records the average value of each frame rate (Frame Per Second, FPS) of at least 16 sub-frames.

The data recording and drawing module 16 is connected to the image cutting calculation module 14. The data recording and drawing module 16 is configured to record the average value of each sub-screen and draw a plurality of corresponding waveform graphs. For example, the waveform graph 161 is drawn in time units. For example, seconds and minutes are used as time units. The data recording and drawing module 16 can be composed of, for example, a memory and a graphics processing unit (GPU), a central processing unit (CPU), a digital signal processor (DSP), a general-purpose or special-purpose microprocessor (Microprocessor), a field programmable gate array (FPGA), an application-specific integrated circuit (ASIC), and a neural network accelerator.

The calculation module 18 is connected to the image cutting calculation module 14 and the data recording and drawing module 16. The calculation module 18 is configured to execute a skin sensitivity state evaluation parameter algorithm based on the data of the at least 16 sub-frames to obtain discrete parameters. The skin sensitivity state evaluation parameter algorithm is:

in, is the discrete parameter, 𝑆 is the total signal of photoplethysmography (PPG), is the peak distance of a specific PPG signal, n is the number of other PPG signals closest to the specific PPG signal, is the wavefront distance closest to the specific PPG signal. Since sensitive skin is prone to inflammation leading to microvascular dilation or hyperplasia, this will cause sensitive skin to have a tendency to increase in disorder compared to normal skin. Therefore, the discrete parameter value calculated by the calculation module 18 is used to evaluate the sensitivity of the skin.

The evaluation module 20 is connected to the calculation module 18. The evaluation module 20 is configured to confirm whether the captured image corresponds to sensitive skin or stable skin based on the relationship between the discrete parameter and the threshold. For example, the evaluation module 20 can automatically compare the discrete parameter value with the threshold to confirm the relationship between the discrete parameter and the threshold. When the discrete parameter is greater than or equal to the threshold, it is confirmed that the captured image corresponds to sensitive skin. When the discrete parameter is less than the threshold, it is confirmed that the captured image corresponds to stable skin. The threshold can be set by the user or determined by an artificial intelligence model. Thus, the skin multi-dimensional sensitivity state data evaluation system 100 of the embodiment of the present invention adopts filtering processing, image analysis and automatic evaluation to evaluate the skin sensitivity state, which is more accurate than the detection method using human subjectivity (e.g., naked eye observation) in the known technology. In addition, the skin multi-dimensional sensitivity state data evaluation system 100 of the embodiment of the present invention adopts image analysis, which is very low in discomfort, or even no discomfort, compared with the invasive detection method used in the known technology.

Please refer to FIG. 2 , which shows a waveform diagram drawn by the data recording and drawing module 16 according to an embodiment of the present invention. As shown in FIG. 2 , the data recording and drawing module 16 draws a waveform diagram 161 corresponding to the 16 sub-screens based on the average value of each of the 16 sub-screens. The n above the waveform diagram 161 represents the nth sub-screen. For example, n=1 represents the first sub-screen, n=2 represents the second sub-screen, n=3 represents the third sub-screen, and so on. In this way, the changes in blood concentration in different skin areas within the image capture range of the image photography module 10 can be identified. These obvious waveform diagrams 161 correspond to the amounts of capillary rhythm under the skin. It is assumed that the more similar each waveform is to the other, the more consistent the skin blood flow is identified. Therefore, if there is sensitive skin (e.g., inflamed tissue), it can be identified most specifically from these waveforms 161.

Please refer to Figure 1, Figure 2 and Figure 3 together. Figure 3 is a flow chart showing the method for evaluating the multi-dimensional skin sensitivity state data according to the embodiment of the present invention. The method for evaluating the multi-dimensional skin sensitivity state data of the embodiment of the present invention is to evaluate the skin within a predetermined time after being stimulated.

In step S300, the image of the skin is captured by the image camera module 10 to generate a captured image. The skin multi-dimensional sensitivity state data evaluation system 100 of the embodiment of the present invention adopts an image analysis method. Compared with the invasive detection method adopted in the prior art, the discomfort of the embodiment of the present invention is very low, or even no discomfort.

In step S310, the captured image is filtered by the filter module 12 to generate a filtered image. For example, the filter module 12 removes unnecessary noise and ambient light signals (e.g., light signals with a wavelength range of 400nm to 1000nm) in the captured image to improve the image quality corresponding to the skin, thereby improving the accuracy of subsequent image data processing.

In step S320, the filtered image is divided into at least 16 sub-frames by the image cutting calculation module 14, and the average value of each sub-frame is calculated. For example, the filtered image is divided into a plurality of sub-frames according to a predetermined window by the image cutting calculation module 14, and the average value of each sub-frame is calculated, and the average value of each frame rate frame of at least 16 sub-frames is recorded.

In step S330, the average value of each sub-frame is recorded by the data recording and drawing module 16, and a plurality of corresponding waveform graphs 161 are drawn. For example, the waveform graphs 161 are drawn in units of time.

In step S340, the calculation module 18 executes a skin sensitivity state evaluation parameter algorithm based on the data of the at least 16 sub-frames to obtain discrete parameters. The skin sensitivity state evaluation parameter algorithm is:

in, is the discrete parameter, 𝑆 is the total signal of photoplethysmography (PPG), is the peak distance of a specific PPG signal, n is the number of other PPG signals closest to the specific PPG signal, is the wavefront distance closest to the specific PPG signal. Since sensitive skin is prone to inflammation leading to microvascular dilation or hyperplasia, this will cause sensitive skin to have a tendency to increase in disorder compared to normal skin. Therefore, the discrete parameter value calculated by the calculation module 18 is used to evaluate the sensitivity of the skin.

In step S350, the evaluation module 20 is used to confirm whether the captured image corresponds to sensitive skin or stable skin according to the relationship between the discrete parameter and the threshold. For example, the evaluation module 20 automatically compares the discrete parameter value with the threshold to confirm the relationship between the discrete parameter and the threshold. When the discrete parameter is greater than or equal to the threshold, it is confirmed that the captured image corresponds to sensitive skin. When the discrete parameter is less than the threshold, it is confirmed that the captured image corresponds to stable skin.

Please refer to FIG4, which is a block diagram showing a skin multi-dimensional sensitivity state data evaluation system according to another embodiment of the present invention. The embodiment of FIG4 is different from the embodiment of FIG1 in that the skin multi-dimensional sensitivity state data evaluation system 110 further includes a processing module 22, and the remaining modules are the same or similar to the embodiment of FIG1, and will not be described in detail below.

The processing module 22 is connected to the image cutting calculation module 14, the data recording and mapping module 16, and the calculation module 18. The processing module 22 is configured to smooth the data of each sub-screen. The processing module 22 can be composed of, for example, a CPU, a DSP, a general-purpose or special-purpose microprocessor, an FPGA, an application-specific integrated circuit ASIC, and a neural network accelerator. Since image data usually has noise, the processing module 22 can smooth the data of each sub-screen to reduce the noise of the image data (i.e., the data of each sub-screen), thereby improving the accuracy and reliability of the skin multidimensional sensitivity state data assessment system 110. In other embodiments, the processing module 22 may also perform edge detection, adaptive thresholding, erosion, and expansion on the data of each sub-frame to reduce the noise of the image data.

Please refer to FIG. 4 and FIG. 5 together. FIG. 5 is a flow chart showing a method for evaluating skin multi-dimensional sensitivity state data according to another embodiment of the present invention. FIG. 5 differs from the embodiment of FIG. 3 in that: after step S320, step S322 is further included. The remaining steps are the same or similar to the embodiment of FIG. 3 and will not be described in detail below.

In step S322, the data of each sub-screen is smoothed by the processing module 22. Since the image data usually has noise, the noise of the image data (i.e., the data of each sub-screen) is reduced by smoothing the data of each sub-screen by the processing module 22, thereby improving the accuracy and reliability of the skin multi-dimensional sensitivity state data evaluation system 110. In other embodiments, the processing module 22 can also perform edge detection, adaptive thresholding, erosion and expansion processing techniques on the data of each sub-screen to reduce the noise of the image data.

In summary, the skin multi-dimensional sensitivity state data assessment system and assessment method of the present invention improves the quality of image recognition, analysis, and processing by capturing skin image data and filtering the image data, thereby improving the accuracy and reliability of skin sensitivity state assessment, thereby solving the problem of inaccurate assessment results and uncomfortable detection process in known technologies.

10: Image photography module 12: Filtering module 14: Image cutting calculation module 16: Data recording and mapping module 18: Calculation module 20: Evaluation module 22: Processing module 100: Skin multi-dimensional sensitivity state data evaluation system 110: Skin multi-dimensional sensitivity state data evaluation system 161: Waveform diagram S300, S310, S320, S322, S330, S340, S350: Steps

FIG. 1 is a block diagram showing a skin multidimensional sensitivity state data evaluation system according to an embodiment of the present invention. FIG. 2 is a waveform diagram drawn by a data recording and drawing module according to an embodiment of the present invention. FIG. 3 is a flow chart showing a skin multidimensional sensitivity state data evaluation method according to an embodiment of the present invention. FIG. 4 is a block diagram showing a skin multidimensional sensitivity state data evaluation system according to another embodiment of the present invention. FIG. 5 is a flow chart showing a skin multidimensional sensitivity state data evaluation method according to another embodiment of the present invention.

10: Image photography module

12: Filter module

14: Image cutting calculation module

16: Data recording and mapping module

18: Calculation module

20: Evaluation module

100: Skin sensitivity status assessment system

Claims (10)

A skin multi-dimensional sensitivity state data evaluation system is used to evaluate the skin within a predetermined time after being stimulated. The skin multi-dimensional sensitivity state data evaluation system includes: An image photography module, configured to capture the image of the skin to generate a captured image; A filtering module, connected to the image photography module, configured to filter the captured image to generate a filtered image; An image cutting and calculation module, connected to the filtering module, configured to divide the filtered image into at least 16 sub-screens and calculate the average value of each sub-screen; A data recording and drawing module, connected to the image cutting and calculation module, configured to record the average value of each sub-screen and draw a plurality of corresponding waveform graphs; A calculation module, connected to the image cutting calculation module and the data recording and mapping module, is configured to execute a skin sensitivity state evaluation parameter algorithm based on the data of the at least 16 sub-screens to obtain a discrete parameter; and an evaluation module, connected to the calculation module, is configured to confirm whether the captured image corresponds to sensitive skin or stable skin based on the relationship between the discrete parameter and the threshold. The skin multi-dimensional sensitivity state data assessment system as described in claim 1 further includes a processing module connected to the image cutting calculation module, the data recording and mapping module, and the calculation module, and the processing module is configured to smooth the data of each sub-screen. A skin multi-dimensional sensitivity state data assessment system as described in claim 1, wherein the filtering wavelength range of the filtering processing is a light signal of 400nm~1000nm. The multi-dimensional skin sensitivity state data assessment system as described in claim 1, wherein the skin sensitivity state assessment parameter algorithm is: in, is the discrete parameter, 𝑆 is the total signal of photoplethysmography (PPG), is the peak distance of a specific PPG signal, n is the number of other PPG signals closest to the specific PPG signal, is one of the wave peak intervals closest to the specific PPG signal. A skin multi-dimensional sensitivity state data assessment system as described in claim 1, wherein when the discrete parameter is greater than or equal to the threshold value, it is confirmed that the captured image corresponds to the sensitive skin; and when the discrete parameter is less than the threshold value, it is confirmed that the captured image corresponds to the stable skin. A method for evaluating skin multi-dimensional sensitivity state data is to evaluate the skin within a predetermined time after being stimulated. The method comprises the following steps: Capturing the image of the skin through an image photography module to generate a captured image; Filtering the captured image through a filtering module to generate a filtered image; Segmenting the filtered image into at least 16 sub-screens through an image cutting calculation module, and calculating the average value of each sub-screen; Recording the average value of each sub-screen through a data recording and drawing module, and drawing multiple corresponding waveform graphs; Executing a skin sensitivity state evaluation parameter algorithm based on the data of the at least 16 sub-frames through a calculation module to obtain a discrete parameter; and Confirming through an evaluation module whether the captured image corresponds to sensitive skin or stable skin according to the relationship between the discrete parameter and the threshold. The method for evaluating skin multidimensional sensitivity state data as described in claim 6, wherein after the filtered image is divided into at least 16 sub-frames by an image cutting calculation module and the average value of each sub-frame is calculated, it further includes the step of smoothing the data of each sub-frame by a processing module. A method for evaluating skin multidimensional sensitivity state data as described in claim 6, wherein the filtering wavelength range of the filtering processing is a light signal of 400nm~1000nm. The method for evaluating skin multidimensional sensitivity state data as described in claim 6, wherein the skin sensitivity state evaluation parameter algorithm is: in, is the discrete parameter, 𝑆 is the total amount of PPG signal, is the peak distance of a specific PPG signal, n is the number of other PPG signals closest to the specific PPG signal, is one of the wave peak intervals closest to the specific PPG signal. A method for evaluating skin multidimensional sensitivity state data as described in claim 6, wherein when the discrete parameter is greater than or equal to the threshold value, it is confirmed that the captured image corresponds to the sensitive skin; and when the discrete parameter is less than the threshold value, it is confirmed that the captured image corresponds to the stable skin.