Traditional Chinese medicine constitution non-contact identification method and system based on psychological and emotional characteristics

Abstract

The embodiment of the invention discloses a traditional Chinese medicine constitution non-contact identification method and system based on psychological and emotional characteristics, and the method comprises the steps: collecting N face videos, where N is an integer greater than 10; the face video is processed through a remote photoelectric volume pulse wave tracing method, and psychological and emotional characteristics corresponding to the face video are obtained; constructing an identification model through an Euclidean distance method, and inputting psychological and emotional features corresponding to the face video into the identification model for training to obtain an optimal identification model; and inputting to-be-identified psychological and emotional characteristics into the optimal identification model to obtain a traditional Chinese medicine constitution identification result corresponding to the to-be-identified psychological and emotional characteristics. The problem that the traditional Chinese medicine constitution cannot be scientifically, accurately and conveniently predicted in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence, and particularly to a non-contact identification method and system for traditional Chinese medicine constitution based on psychological and emotional characteristics. Background Art

[0002] With the increasing emphasis on health management and personalized medicine, the classification and determination of traditional Chinese medicine constitution have important application values in the fields of medical treatment, health preservation, etc. At present, in the prediction application of the classification and determination of traditional Chinese medicine constitution, it mainly relies on the nine constitution scales issued by the Chinese Association of Chinese Medicine.

[0003] This constitution scale covers nine types, namely, peaceful constitution, qi-deficiency constitution, yang-deficiency constitution, yin-deficiency constitution, phlegm-dampness constitution, damp-heat constitution, blood stasis constitution, qi stagnation constitution, and special endowment constitution. Its determination method is that the tested person answers a series of questions in the scale, and then converts according to the answer scores to determine the constitution type.

[0004] However, this traditional determination method has significant problems. On the one hand, the final result of the scale completely depends on the answers to each question. In practical applications, although the design of the five-level answers aims to more precisely quantify individual situations, since the degrees represented by adjacent scores (especially 2 points, 3 points, and 4 points) are very close, this makes the boundaries unclear. For example, for a question describing a certain constitution characteristic, it is very difficult for the tested person to accurately distinguish whether their situation is closer to 3 points or 4 points, and this ambiguity easily leads to large deviations when calculating the total score and converting the results.

[0005] On the other hand, the entire determination process has a strong subjectivity. Different tested persons may have differences in understanding the questions, which will affect the accuracy of the answers. Moreover, the operators may also have different judgments due to factors such as personal experience when interpreting and processing the results. In addition, this scale-based determination method is cumbersome and complex to operate. From the tested person filling out the questionnaire to the final result conversion and analysis, it requires a lot of time and energy, and professional personnel are needed for guidance and interpretation, which limits its rapid and convenient application in a large-scale population.

[0006] In summary, the existing methods for the classification and determination of traditional Chinese medicine constitution based on the nine constitution scales have problems such as low accuracy, strong subjectivity, and complex operation, which seriously affect the effect and popularization of the classification of traditional Chinese medicine constitution in practical applications. There is an urgent need for a more scientific, accurate, and convenient method for predicting traditional Chinese medicine constitution. Summary of the Invention

[0007] The purpose of the embodiments of the present invention is to provide a non-contact identification method and system for traditional Chinese medicine constitution based on psychological and emotional characteristics, so as to solve the problem that the prior art cannot scientifically, accurately, and conveniently predict traditional Chinese medicine constitution.

[0008] To achieve the above object, an embodiment of the present invention provides a non-contact identification method for traditional Chinese medicine constitution based on psychological and emotional characteristics. The non-contact identification method for traditional Chinese medicine constitution based on psychological and emotional characteristics specifically includes:

[0009] Collect N face videos, where N is an integer greater than 10;

[0010] Process the face video through remote photoplethysmography to obtain the psychological and emotional characteristics corresponding to the face video;

[0011] Construct an identification model through the Euclidean distance method, input the psychological and emotional characteristics corresponding to the face video into the identification model for training to obtain an optimal identification model;

[0012] Input the psychological and emotional characteristics to be identified into the optimal identification model to obtain the identification result of traditional Chinese medicine constitution corresponding to the psychological and emotional characteristics to be identified.

[0013] Based on the above technical solution, the present invention can also be improved as follows:

[0014] Further, the process of processing the face video through remote photoplethysmography to obtain the psychological and emotional characteristics corresponding to the face video includes:

[0015] Perform light intensity analysis on each frame image in the face video to obtain a signal of the change of light intensity over time, and extract the original rPPG signal based on the signal;

[0016] Preprocess the original rPPG signal to obtain a pure rPPG signal.

[0017] Further, the process of processing the face video through remote photoplethysmography to obtain the psychological and emotional characteristics corresponding to the face video further includes:

[0018] Extract a number of original feature parameters based on the pure rPPG signal;

[0019] Calculate the correlation coefficients of a number of the original feature parameters corresponding to the identification of traditional Chinese medicine constitution;

[0020] Extract the original feature parameters whose correlation coefficients exceed the preset standard range value as the final feature parameters.

[0021] Further, the extraction of a number of original HRV features based on the pure rPPG signal includes:

[0022] Perform cubic spline interpolation on the rPPG signal and collect 17 original feature parameters through the heart rate variability feature calculation method; among them, the original feature parameters include 9 time-domain feature parameters, 4 frequency-domain feature parameters, and 4 non-linear feature parameters;

[0023] Among them, the time-domain feature parameters include the average heart rate of the samples during the acquisition time, the standard deviation of the RR interval, the percentage of the number of RR intervals greater than 50 ms in the total number, the percentage of the number of RR intervals greater than 20 ms in the total number, the square root of the average of the sum of the squares of the differences between adjacent RR intervals, the standard deviation of the differences between adjacent RR intervals, the coefficient of variation of the continuous differences, the coefficient of variation, and the standard deviation of the heart rate;

[0024] The frequency-domain feature parameters include the low-frequency band power, the high-frequency band power, the total signal power, and the low-to-high frequency power ratio;

[0025] The non-linear feature parameters include the standard deviation of the Poincaré plot on the line perpendicular to the marker line, the standard deviation along the marker line, the ratio of the standard deviation of the Poincaré plot on the line perpendicular to the marker line, and the sample entropy.

[0026] Further, extracting the original feature variables whose correlation coefficients exceed the preset standard range values as the final feature variables includes:

[0027] Performing feature screening on the 17 original feature parameters extracted based on the sample data with pressure labels to obtain 13 final feature variables. Among them, the final feature variables include the average heart rate of the samples during the acquisition time, the standard deviation of the RR interval, the percentage of the number of RR intervals greater than 50 ms in the total number, the percentage of the number of RR intervals greater than 20 ms in the total number, the square root of the average of the sum of the squares of the differences between adjacent RR intervals, the standard deviation of the differences between adjacent RR intervals, the coefficient of variation of the continuous differences, the coefficient of variation, the low-frequency band power, the high-frequency band power, the total signal power, the low-to-high frequency power ratio, and the sample entropy.

[0028] Further, inputting the psychological and emotional characteristics to be identified into the optimal identification model to obtain the traditional Chinese medicine constitution identification result corresponding to the psychological and emotional characteristics to be identified, including:

[0029] Calculating the distance between samples through formula (1):

[0030] S = D / (D T1 + D T2 ) Formula (1);

[0031] In the formula, D T1 is the threshold of a certain type of psychological and emotional characteristics, D T2 is the threshold of the first type of traditional Chinese medicine constitution, and D is the distance between the psychological and emotional characteristics and the traditional Chinese medicine constitution;

[0032] When S < 1, the two types of discriminant samples intersect;

[0033] When S = 1, the two types of discriminant samples are tangent;

[0034] When S > 1, the two types of discriminant samples are separated.

[0035] Furthermore, the step of inputting the psychological and emotional characteristics to be identified into the optimal identification model to obtain the traditional Chinese medicine constitution identification result corresponding to the psychological and emotional characteristics to be identified further includes:

[0036] Formulating a conversion table based on the relevant importance of psychological and emotional characteristics and traditional Chinese medicine constitutions, querying the scores of nine traditional Chinese medicine constitutions based on the conversion table, and obtaining the traditional Chinese medicine constitution identification result based on the scores of the nine traditional Chinese medicine constitutions.

[0037] A non-contact identification system for traditional Chinese medicine constitution based on psychological and emotional characteristics, comprising:

[0038] An acquisition module, configured to acquire N face videos, where N is an integer greater than 10;

[0039] A preprocessing module, configured to process the face video by remote photoplethysmography to obtain the psychological and emotional characteristics corresponding to the face video;

[0040] A construction module, configured to construct an identification model by the Euclidean distance method, input the psychological and emotional characteristics corresponding to the face video into the identification model for training, and obtain an optimal identification model;

[0041] An identification module, configured to input the psychological and emotional characteristics to be identified into the optimal identification model to obtain the traditional Chinese medicine constitution identification result corresponding to the psychological and emotional characteristics to be identified.

[0042] An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method are implemented.

[0043] A non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method are implemented.

[0044] The embodiments of the present invention have the following advantages:

[0045] In the present invention, a non-contact identification method for traditional Chinese medicine constitution based on psychological and emotional characteristics uses the method of collecting facial videos, without any contact with the human body. Compared with traditional methods for identifying traditional Chinese medicine constitution, this non-contact means greatly improves the convenience of identification, avoids physical intervention on the tested person, reduces discomfort and hygiene problems that may be brought about by contact, and at the same time makes the detection process more natural and fast. Information can be collected without the tested person's awareness, improving acceptance.

[0046] It breaks through the limitations of traditional identification of traditional Chinese medicine constitution that only relies on physical symptoms and signs, and takes psychological and emotional characteristics into consideration. Psychological and emotional states are important reflections of the overall state of the human body, which is in line with the holistic concept emphasized by traditional Chinese medicine. In this way, more potential information related to constitution can be excavated, making the constitution identification more comprehensive and accurate, and helping to discover constitution tendencies hidden at the psychological and emotional levels.

[0047] The Euclidean distance method is used to construct an identification model, and it is trained with the psychological and emotional characteristics corresponding to a large number of facial videos to obtain the optimal identification model. This data-driven model training method can fully explore the complex relationship between psychological and emotional characteristics and traditional Chinese medicine constitution, thereby improving the accuracy of the identification results. By continuously optimizing the model, it can adapt to different populations and individual differences, reducing misjudgments. Brief Description of the Drawings

[0048] In order to more clearly illustrate the implementation manners of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the implementation manners or the prior art. Obviously, the drawings described below are only exemplary. For those of ordinary skill in the art, without creative efforts, other implementation drawings can also be obtained based on the provided drawings.

[0049] The structures, proportions, sizes, etc. illustrated in this specification are only used to cooperate with the content disclosed in the specification for those familiar with this technology to understand and read, and are not used to limit the limiting conditions for the implementation of the present invention. Therefore, they do not have technical substantive significance. Any modification of the structure, change in the proportional relationship, or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope covered by the technical content disclosed in the present invention.

[0050] Figure 1 It is a flow chart of the non-contact identification method for traditional Chinese medicine constitution based on psychological and emotional characteristics of the present invention;

[0051] Figure 2 It is an architecture diagram of the non-contact identification system for traditional Chinese medicine constitution based on psychological and emotional characteristics of the present invention;

[0052] Figure 3 Schematic diagram of the physical structure of the electronic device provided by the present invention.

[0053] The accompanying reference numerals are as follows:

[0054] Acquisition module 10, preprocessing module 20, construction module 30, identification module 40, electronic device 50, processor 501, memory 502, bus 503. Specific implementation manners

[0055] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0056] It should be noted that the terms "first", "second", etc. in the description and drawings of the present invention are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so as to implement the embodiments of the present invention described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0057] It should be noted that, without conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. The embodiments of the present invention will be described in detail below in conjunction with the drawings.

[0058] Figure 1 Flowchart of an embodiment of the non-contact identification method for traditional Chinese medicine constitution based on psychological and emotional characteristics of the present invention. As Figure 1 shown, a non-contact identification method for traditional Chinese medicine constitution based on psychological and emotional characteristics provided by an embodiment of the present invention includes the following steps:

[0059] S101, collect face video with the number N, where N is an integer greater than 10;

[0060] Specifically: Use a high-definition camera as the acquisition device, ensuring that its resolution can clearly capture the subtle changes in the face. For example, the resolution is at least 1920×1080 pixels. Install the camera at a fixed position to ensure that the shooting angle can completely capture the face image. Generally, it is recommended to keep it horizontal to the face of the person being collected, and the distance is between 0.5 - 1.5 meters.

[0061] The acquisition environment should maintain uniform and sufficient light, avoiding direct strong light or shadow interference. The combination of natural light and soft indoor lighting can be used, and the light intensity is controlled within the range of 500 - 1000 lux. At the same time, the background should be simple, avoiding complex patterns or colors to reduce interference with face feature extraction.

[0062] Instruct the person being collected to maintain a natural and relaxed state, keep the head stable, and avoid large-scale shaking or overly exaggerated expressions. For each person being collected, multiple groups of videos are collected, and the duration of each group of videos is 10 - 30 seconds to cover the facial states at different moments. During the collection process, the person being collected is required to perform some simple actions, such as natural blinking, slight eye rotation, smiling, etc. These actions help to obtain more abundant facial information.

[0063] Between collecting videos of different people being collected, clean and calibrate the camera to ensure the consistency of image quality. At the same time, perform real-time preview and preliminary inspection on the collected videos. If it is found that the videos are blurred, blocked, or have other quality problems, re-collect immediately.

[0064] S102, Process the face video through remote photoplethysmography to obtain the psychological and emotional characteristics corresponding to the face video;

[0065] Specifically, from the collected face video, extract image data frame by frame. Use image analysis algorithms to identify the facial skin area, especially the parts rich in blood vessels such as the cheeks and forehead. The skin color changes in these parts are related to the blood volume changes, and photoplethysmography is based on this principle to detect the subtle fluctuations of blood volume changes.

[0066] According to the absorption and reflection characteristics of the skin to light of different wavelengths, select light of appropriate wavelengths for analysis. For example, green light (wavelength about 500 - 570nm) is usually used because the absorption difference of oxyhemoglobin and reduced hemoglobin in the blood by green light is more obvious, which can better reflect the blood volume changes. By analyzing the light intensity changes in the skin area under green light irradiation in each frame of the image, a signal similar to the pulse wave is extracted.

[0067] Analyze the characteristic parameters of the extracted photoplethysmogram (PPG) signals, such as the amplitude, frequency, waveform, etc. of the pulse wave. Different psychological and emotional states will cause changes in these parameters. For example, when a person is in a tense mood, the pulse wave frequency may increase and the amplitude may also increase.

[0068] Combined with facial expression analysis, further determine the psychological and emotional characteristics. Use technologies such as the Facial Action Coding System (FACS) to analyze the movements of facial muscles in the video, such as expressions like frowning and raising the corners of the mouth, and combine this expression information with the characteristics of the PPG signal. For example, a frowning action accompanied by a specific change in the pulse wave frequency may imply an anxious mood. At the same time, observe the changes in pupil size, as the pupils also have different responses under different emotions, such as dilating in excitement or fear.

[0069] Perform light intensity analysis on each frame image in the face video to obtain a signal of the change of light intensity over time, and extract the original rPPG signal based on this signal;

[0070] Preprocess the original rPPG signal to obtain a pure rPPG signal.

[0071] Extract a number of original characteristic parameters based on the pure rPPG signal;

[0072] Calculate the correlation coefficients of a number of the original characteristic parameters corresponding to the identification of traditional Chinese medicine constitutions;

[0073] Extract the original characteristic parameters whose correlation coefficients exceed the preset standard range value as the final characteristic parameters.

[0074] Perform cubic spline interpolation on the rPPG signal, upsample it to 240 Hz, then extract peak points, calculate the time intervals between peak points to obtain the RR interval sequence, remove the outliers in the RR interval sequence and plot the HRV time domain curve, use statistical methods to obtain time domain characteristics, and perform frequency domain analysis on the signal using the Welch power spectrum diagram to extract frequency domain characteristic parameters. Based on the Poincaré scatter plot analysis method, extract non-linear characteristic parameters. Collect 17 original characteristic parameters through the heart rate variability characteristic calculation method; among them, the original characteristic parameters include 9 time domain characteristic parameters, 4 frequency domain characteristic parameters, and 4 non-linear characteristic parameters;

[0075] Among them, the time-domain characteristic parameters are HR (the average heart rate of the samples within the acquisition time), SDNN (the standard deviation of the RR intervals), PNN50 (the percentage of the number of RR intervals greater than 50 ms in the total number), PNN20 (the percentage of the number of RR intervals greater than 20 ms in the total number), RMSSD (the square root of the mean of the sum of the squares of the differences between adjacent RR intervals), SDSD (the standard deviation of the differences between adjacent RRs), CVSD (the coefficient of variation of the continuous differences), CVnn i (the coefficient of variation), and std_hr (the standard deviation of the heart rate);

[0076] The frequency-domain characteristic parameters are LF (low-frequency band power), HF (high-frequency band power), TP (total signal power), and LF / HF (low-to-high frequency power ratio);

[0077] The non-linear characteristic parameters are: SD1 (the standard deviation of the Poincaré plot on the straight line perpendicular to the marker line), SD2 (the standard deviation along the marker line), SD1 / SD2 (the ratio of SD2 to SD1), and SampEn (sample entropy).

[0078] Feature screening is performed on the 17 extracted original characteristic parameters based on the sample data with pressure labels. SPSS is used for data statistical analysis. Through paired t-tests, 13 characteristic parameters with significant differences are selected as the final characteristic variables. Among them, the three parameters std_hr, SD1 / SD2, and SampEn do not show obvious differences, so these three characteristic parameters are discarded. Among them, the final characteristic variables include the average heart rate of the samples within the acquisition time, the standard deviation of the RR intervals, the percentage of the number of RR intervals greater than 50 ms in the total number, the percentage of the number of RR intervals greater than 20 ms in the total number, the square root of the mean of the sum of the squares of the differences between adjacent RR intervals, the standard deviation of the differences between adjacent RRs, the coefficient of variation of the continuous differences, the coefficient of variation, low-frequency band power, high-frequency band power, total signal power, low-to-high frequency power ratio, and sample entropy.

[0079] When in a state of long-term stress, the autonomic nervous system will undergo corresponding changes, which are then manifested in the characteristic values of HRV. Stress can lead to a decrease in HRV, and this decrease is mainly manifested as a reduction in various time-domain indicators. For example, the time-domain indicator SDNN can be used to evaluate the state of the body's stress resistance. Good stress resistance often enables the body to self-regulate as quickly as possible when under stress, avoiding the generation of excessive negative emotions; RMSSD can be used to measure abnormal lesions or abnormal syndromes of the human heart; when the values of SDNN and RMSSD are both lower than 10 ms, it often indicates an increased incidence of heart disease. For example, the frequency-domain feature TP will show a significant decrease under chronic stress, and the reduction of total energy indicates a decrease in the ability of autonomic nerve regulation; LF / HF represents the overall balance between the sympathetic and parasympathetic nerves, and it will increase when under external stress. Both too high and too low are states of autonomic nerve imbalance. The sympathetic nerve is responsible for regulating sudden events, while the parasympathetic nerve is responsible for controlling various body functions and normal activities. The same applies to other conditions such as depression and anxiety.

[0080] S103, construct an identification model through the Euclidean distance method, input the psychological and emotional characteristics corresponding to the face video into the identification model for training to obtain the optimal identification model;

[0081] Specifically, divide the psychological and emotional characteristics into a training set, a validation set, and a test set;

[0082] Train the identification model based on the training set;

[0083] Evaluate the performance of the trained identification model based on the validation set to obtain an identification model that meets the performance conditions;

[0084] Evaluate the prediction results of the identification model that meets the performance conditions based on the test set to obtain the evaluation index corresponding to the identification model.

[0085] Evaluate the performance of the trained identification model based on the validation set to obtain an identification model that meets the performance conditions; evaluate the similarity calculation results of the identification model that meets the performance conditions based on the test set to obtain the evaluation index corresponding to the identification model. Evaluate the performance of the identification model to obtain a percentage score (that is, the highest score is 100 points and the lowest score is 0 points), and determine the identification model with a score greater than the set value based on the percentage score. For example, the identification model with a score greater than 90 points is the identification model that meets the performance conditions.

[0086] Calculate the evaluation index for the identification model that meets the performance conditions to obtain the evaluation index of the identification model, and calculate the evaluation value corresponding to each evaluation index. The evaluation value is used to represent the ability value of the identification model on the evaluation index.

[0087] Circularly calculate the distance between the psychological and emotional characteristics to be identified and the class centers of various physical constitutions in the training set, and obtain the traditional Chinese medicine physical constitution category corresponding to the nearest class center distance;

[0088] Calculate the distance between samples through formula (1):

[0089] S = D / (D T1 + D T2 ) Formula (1);

[0090] In the formula, D T1 is the threshold of a certain psychological and emotional characteristic, D T2 is the threshold of the first type of traditional Chinese medicine physical constitution, and D is the distance between the psychological and emotional characteristics and the traditional Chinese medicine physical constitution;

[0091] When S < 1, the two types of discriminant samples intersect;

[0092] When S = 1, the two types of discriminant samples are tangent;

[0093] When S > 1, the two types of discriminant samples are separated.

[0094] S104. Input the psychological and emotional characteristics to be identified into the optimal identification model to obtain the traditional Chinese medicine physical constitution identification result corresponding to the psychological and emotional characteristics to be identified;

[0095] Specifically, formulate a conversion table based on the relevant importance of psychological and emotional characteristics and traditional Chinese medicine physical constitutions, query the scores of nine traditional Chinese medicine physical constitutions based on the conversion table, and obtain the traditional Chinese medicine physical constitution identification result based on the scores of nine traditional Chinese medicine physical constitutions.

[0096] When the psychological pressure is too high, usually there is peaceful constitution > qi stagnation constitution > qi deficiency constitution > yang deficiency constitution > phlegm-dampness constitution > yin deficiency constitution > damp-heat constitution > blood stasis constitution > special endowment constitution; when depression occurs, usually there is qi stagnation constitution > blood stasis constitution > qi deficiency constitution > special endowment constitution > phlegm-dampness constitution > damp-heat constitution > yin deficiency constitution > yang deficiency constitution > peaceful constitution; when anxiety occurs, usually there is damp-heat constitution > qi stagnation constitution > special endowment constitution > blood stasis constitution > yang deficiency constitution > qi deficiency constitution > phlegm-dampness constitution > yin deficiency constitution, etc.

[0097] The model calculates according to the internal algorithm and the parameters obtained through training based on the input psychological and emotional characteristics to be identified. Inside the model, by calculating the Euclidean distance (or other relevant calculation methods, depending on the model structure) between the characteristics to be identified and the characteristics of the training samples, the known physical constitution type sample most similar to the sample to be identified is determined.

[0098] Once the model has completed the calculation, the corresponding traditional Chinese medicine (TCM) constitution identification result is output. The result can be presented in text form, clearly indicating which type of TCM constitution the individual to be identified belongs to (such as yang deficiency constitution, yin deficiency constitution, phlegm-dampness constitution, etc.). At the same time, the confidence information of the result can be attached. The confidence level represents the degree of certainty of the model for this identification result. For example, a confidence level of 80% means that the model has a relatively high degree of confidence that the result is accurate, but there is still a certain degree of uncertainty. According to the confidence level, it can be further decided whether it is necessary to conduct manual review of the result or adopt other auxiliary diagnosis methods.

[0099] Figure 2 This is the architecture diagram of the embodiment of the non-contact identification system for traditional Chinese medicine constitution based on psychological and emotional characteristics of the present invention; as Figure 2 shown, a non-contact identification system for traditional Chinese medicine constitution based on psychological and emotional characteristics provided by the embodiment of the present invention includes the following steps:

[0100] The acquisition module 10 is used to acquire N face videos, where N is an integer greater than 10;

[0101] The preprocessing module 20 is used to process the face video through remote photoplethysmography to obtain the psychological and emotional characteristics corresponding to the face video;

[0102] The preprocessing module 20 is further used for:

[0103] Performing light intensity analysis on each frame image in the face video to obtain a signal of the change of light intensity over time, and extracting an original rPPG signal based on the signal;

[0104] Preprocessing the original rPPG signal to obtain a pure rPPG signal.

[0105] Extracting a number of original feature parameters based on the pure rPPG signal;

[0106] Calculating the correlation coefficients of a number of the original feature parameters corresponding to the identification of traditional Chinese medicine constitution;

[0107] Extracting the original feature parameters whose correlation coefficients exceed the preset standard range value as the final feature parameters.

[0108] Performing cubic spline interpolation on the rPPG signal, and collecting 17 original feature parameters through the heart rate variability feature calculation method; wherein, the original feature parameters include 9 time-domain feature parameters, 4 frequency-domain feature parameters, and 4 non-linear feature parameters;

[0109] Among them, the time-domain characteristic parameters include the average heart rate of the samples within the acquisition time, the standard deviation of the RR intervals, the percentage of the number of RR intervals greater than 50 ms in the total number, the percentage of the number of RR intervals greater than 20 ms in the total number, the square root of the average of the sum of the squares of the differences between adjacent RR intervals, the standard deviation of the differences between adjacent RRs, the coefficient of variation of the continuous differences, the coefficient of variation, and the standard deviation of the heart rate;

[0110] The frequency-domain characteristic parameters include the low-frequency band power, the high-frequency band power, the total signal power, and the low-to-high frequency power ratio;

[0111] The non-linear characteristic parameters include the standard deviation of the Poincaré plot on the straight line perpendicular to the marker line, the standard deviation along the marker line, the ratio of the standard deviation of the Poincaré plot on the straight line perpendicular to the marker line, and the sample entropy.

[0112] Based on the sample data with pressure labels, feature screening is performed on the 17 extracted original characteristic parameters to obtain 13 final characteristic variables. Among them, the final characteristic variables include the average heart rate of the samples within the acquisition time, the standard deviation of the RR intervals, the percentage of the number of RR intervals greater than 50 ms in the total number, the percentage of the number of RR intervals greater than 20 ms in the total number, the square root of the average of the sum of the squares of the differences between adjacent RR intervals, the standard deviation of the differences between adjacent RRs, the coefficient of variation of the continuous differences, the coefficient of variation, the low-frequency band power, the high-frequency band power, the total signal power, the low-to-high frequency power ratio, and the sample entropy.

[0113] The construction module 30 is used to construct an identification model by the Euclidean distance method, and input the psychological and emotional characteristics corresponding to the face video into the identification model for training to obtain an optimal identification model;

[0114] The construction module 30 is further used for:

[0115] Calculate the distance between samples through formula (1):

[0116] S = D / (D T1 + D T2 ) Formula (1);

[0117] In the formula, D T1 is the threshold of a certain type of psychological and emotional characteristics, D T2 is the threshold of the first type of traditional Chinese medicine constitution, and D is the distance between the psychological and emotional characteristics and the traditional Chinese medicine constitution;

[0118] When S < 1, the two types of discriminant samples intersect;

[0119] When S = 1, the two types of discriminant samples are tangent;

[0120] When S > 1, the two types of discriminant samples are separated.

[0121] An identification module 40, configured to input the psychological and emotional characteristics to be identified into the optimal identification model, and obtain a traditional Chinese medicine constitution identification result corresponding to the psychological and emotional characteristics to be identified.

[0122] The identification module 40 is further configured to:

[0123] Formulate a conversion table based on the relevant importance of psychological and emotional characteristics and traditional Chinese medicine constitutions, query the scores of nine traditional Chinese medicine constitutions based on the conversion table, and obtain a traditional Chinese medicine constitution identification result based on the scores of the nine traditional Chinese medicine constitutions.

[0124] Figure 3 The following is a schematic physical structure diagram of an electronic device provided by an embodiment of the present invention. As Figure 3 shown, the electronic device 50 includes: a processor 501 (processor), a memory 502 (memory), and a bus 503;

[0125] Wherein, the processor 501 and the memory 502 communicate with each other through the bus 503;

[0126] The processor 501 is configured to call program instructions in the memory 502 to execute the methods provided by the above method embodiments, for example, including: collecting N face videos, where N is an integer greater than 10; processing the face videos through remote photoplethysmography to obtain psychological and emotional characteristics corresponding to the face videos; constructing an identification model through the Euclidean distance method, inputting the psychological and emotional characteristics corresponding to the face videos into the identification model for training to obtain an optimal identification model; inputting the psychological and emotional characteristics to be identified into the optimal identification model to obtain a traditional Chinese medicine constitution identification result corresponding to the psychological and emotional characteristics to be identified.

[0127] This embodiment provides a non-transitory computer-readable storage medium, and the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions cause the computer to execute the methods provided by the above method embodiments, for example, including: collecting N face videos, where N is an integer greater than 10; processing the face videos through remote photoplethysmography to obtain psychological and emotional characteristics corresponding to the face videos; constructing an identification model through the Euclidean distance method, inputting the psychological and emotional characteristics corresponding to the face videos into the identification model for training to obtain an optimal identification model; inputting the psychological and emotional characteristics to be identified into the optimal identification model to obtain a traditional Chinese medicine constitution identification result corresponding to the psychological and emotional characteristics to be identified.

[0128] Those of ordinary skill in the art can understand that all or part of the steps to implement the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps including the above method embodiments; and the foregoing storage medium includes various storage media such as ROM, RAM, magnetic disks, or optical discs that can store program codes.

[0129] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0130] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disks, optical discs, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiments.

[0131] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A non-contact identification method of TCM constitution based on psychological and emotional characteristics, characterized in that: The non-contact identification method of TCM constitution based on psychological and emotional characteristics specifically includes: Collecting N number of human face videos, where N is an integer greater than 10; Processing the human face video by remote photoplethysmography to obtain psychological and emotional characteristics corresponding to the human face video; Constructing a recognition model by using the Euclidean distance method, inputting the psychological and emotional features corresponding to the face video into the recognition model for training, and obtaining an optimal recognition model; The psychological and emotional characteristics to be identified are input into the optimal identification model to obtain the TCM constitution identification results corresponding to the psychological and emotional characteristics to be identified.

2. The non-contact identification method of TCM constitution based on psychological and emotional characteristics according to claim 1 is characterized in that: The processing of the face video by remote photoplethysmography to obtain the psychological and emotional characteristics corresponding to the face video includes: Performing light intensity analysis on each frame of the face video to obtain a signal of light intensity variation over time, and extracting an original rPPG signal based on the signal; The original rPPG signal is preprocessed to obtain a pure rPPG signal.

3. The non-contact identification method of TCM constitution based on psychological and emotional characteristics according to claim 2 is characterized in that: The method of processing the face video of a person by remote photoplethysmography to obtain the psychological and emotional characteristics corresponding to the face video of the person also includes: Extracting a plurality of original characteristic parameters based on the pure rPPG signal; Calculating the correlation coefficients of the original characteristic parameters corresponding to TCM constitution identification; The original feature parameters whose correlation coefficients exceed the preset standard range are extracted as the final feature parameters.

4. The non-contact identification method of TCM constitution based on psychological and emotional characteristics according to claim 3 is characterized in that: The extracting of several original HRV features based on the pure rPPG signal includes: The rPPG signal is interpolated by cubic spline, and 17 original characteristic parameters are collected by the heart rate variability characteristic calculation method; wherein the original characteristic parameters include 9 time domain characteristic parameters, 4 frequency domain characteristic parameters and 4 nonlinear characteristic parameters; The time domain characteristic parameters include the heart rate mean of the samples within the acquisition time, the standard deviation of the RR interval, the percentage of the number of RR intervals greater than 50ms, the percentage of the number of RR intervals greater than 20ms, the square root of the mean of the sum of squares of the differences between adjacent RR intervals, the standard deviation of the differences between adjacent RRs, the coefficient of variation of the continuous differences, the coefficient of variation and the heart rate standard deviation; The frequency domain characteristic parameters include low frequency band power, high frequency band power, total signal power and low to high frequency power ratio; The nonlinear characteristic parameters include a standard deviation of the Poincare map on a straight line perpendicular to the marking line, a ratio of a standard deviation along the marking line to a standard deviation of the Poincare map on a straight line perpendicular to the marking line, and a sample entropy.

5. The non-contact identification method of TCM constitution based on psychological and emotional characteristics according to claim 3 is characterized in that: The extracting of the original characteristic variables whose correlation coefficients exceed the preset standard range as the final characteristic variables includes: Based on the sample data with pressure labels, the 17 extracted original feature parameters were screened to obtain 13 final feature variables, including the mean heart rate of the samples within the acquisition time, the standard deviation of the RR interval, the percentage of the RR interval greater than 50 ms, the percentage of the RR interval greater than 20 ms, the square root of the mean of the sum of squares of the differences between adjacent RR intervals, the standard deviation of the differences between adjacent RRs, the coefficient of variation of continuous differences, the coefficient of variation, low-frequency band power, high-frequency band power, total signal power, low-high frequency power ratio and sample entropy.

6. The non-contact identification method of TCM constitution based on psychological and emotional characteristics according to claim 1 is characterized in that: The step of inputting the psychological and emotional characteristics to be identified into the optimal identification model to obtain the TCM constitution identification result corresponding to the psychological and emotional characteristics to be identified includes: The distance between samples is calculated by formula (1): S=D / (D T1 +D T2 ) formula (1); Where D T1 is the threshold of a certain type of psychological and emotional characteristics, D T2 is the threshold of the first type of TCM constitution, and D is the distance between psychological and emotional characteristics and TCM constitution; When S < 1, the two types of discriminant samples intersect; When S = 1, the two types of discriminant samples are tangent; When S>1, the two types of discriminant samples are separated.

7. The non-contact identification method of TCM constitution based on psychological and emotional characteristics according to claim 6 is characterized in that: The step of inputting the psychological and emotional characteristics to be identified into the optimal identification model to obtain the TCM constitution identification result corresponding to the psychological and emotional characteristics to be identified further includes: A conversion table is formulated based on the relative importance of psychological and emotional characteristics and TCM constitution, nine TCM constitution scores are queried based on the conversion table, and TCM constitution identification results are obtained based on the nine TCM constitution scores.

8. A non-contact identification system of TCM constitution based on psychological and emotional characteristics, characterized in that: include: A collection module, used for collecting N number of human face videos, where N is an integer greater than 10; A preprocessing module, used for processing the human face video by remote photoplethysmography to obtain the psychological and emotional characteristics corresponding to the human face video; A construction module is used to construct a recognition model by using the Euclidean distance method, and input the psychological and emotional features corresponding to the face video into the recognition model for training to obtain an optimal recognition model; The identification module is used to input the psychological and emotional characteristics to be identified into the optimal identification model to obtain the TCM constitution identification results corresponding to the psychological and emotional characteristics to be identified.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A non-transitory computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images