A special environment worker psychological decompression system and device

Abstract

The application provides a special environment worker psychological decompression system and device, relates to the technical field of psychological decompression, and mainly comprises a data acquisition module, a data processing module and a virtual reality generation module arranged in sequence; the data acquisition module comprises a SAM unit, a virtual scene unit, an electrocardio unit and an electroencephalogram unit, and is respectively used for collecting SAM scale data, virtual scene information, electrocardio data and electroencephalogram data of a test subject and sending the data to the data processing module; the data processing module comprises a database, a first classification unit, a second classification unit and a prediction model unit; the virtual reality generation module comprises a visual generation unit, generates a virtual reality image based on a second alternative material through a NeRF neural network model. According to the scheme, two-dimensional materials can be selected in a targeted manner, three-dimensional virtual reality images are generated in real time through a NeRF neural network model, so that the sense of immersion and the sense of dynamics are effectively improved, and then the psychological decompression efficiency and effect are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of psychological stress reduction, and in particular to a special environment worker psychological stress reduction system and device. BACKGROUND

[0002] In the field of special environment work, workers are long-term in high-intensity, high-risk work environment, facing complex and changing emergency tasks and continuous psychological stress challenges. Studies have shown that the anxiety index of special environment workers is generally 3-5 times higher than that of workers in conventional work environment, and the cumulative effect of psychological stress is significant. If there is no timely and effective psychological intervention measures, not only the work efficiency will be affected, but also the occurrence of post-traumatic stress disorder (PTSD), anxiety and other serious psychological problems is more likely to occur.

[0003] However, the existing psychological stress reduction system using traditional two-dimensional images as emotional stimulus sources in China and international emotional material library has the technical bottlenecks of insufficient immersion, lack of dynamic feedback, and low individualized adaptation, which makes it difficult to break through the critical value of 40% in actual stress reduction effect. Therefore, it has become an urgent need to develop a virtual reality psychological stress reduction system based on multi-dimensional physiological data feedback and scale feedback, with high immersion and dynamic adaptability, to improve the mental health level of special environment workers. SUMMARY

[0004] The purpose of the present application is to provide a special environment worker psychological stress reduction system and device to solve at least one of the above technical problems in the prior art.

[0005] In a first aspect, to solve the above technical problems, the present application provides a special environment worker psychological stress reduction system, comprising a data acquisition module, a data processing module and a virtual reality generation module arranged in sequence:

[0006] The data acquisition module comprises a SAM unit, a virtual scene unit, an electrocardiogram unit and an electroencephalogram unit, respectively used for collecting SAM scale data, virtual scene information, electrocardiogram data and electroencephalogram data of the subject and sending them to the data processing module;

[0007] The data processing module comprises a database, a first classification unit, a second classification unit and a prediction model unit:

[0008] The database stores two-dimensional picture materials and corresponding sound effect materials;

[0009] The first classification unit filters the first candidate materials based on the virtual scene information;

[0010] The second classification unit calculates a decompression index based on the SAM scale data, the electrocardiogram data, and the electroencephalogram data, to evaluate the decompression effect of the current material on the current subject in the database and to mark the current material.

[0011] The prediction model unit stores a prediction model to predict the decompression effect of the unmarked material on the current subject in the database, wherein the decompression effect includes positive decompression, zero decompression, and negative decompression, and sends the first alternative material with positive decompression as the second alternative material to the virtual reality generation module.

[0012] The virtual reality generation module includes a visual generation unit that generates a virtual reality image based on the second alternative material through a NeRF neural network model, and the virtual reality image is used for image-assisted positive decompression for the current subject.

[0013] Through the above system, the scale data, virtual scene data, and physiological data of the special environment worker subject can be classified and the appropriate two-dimensional material can be selected. Then, through the NeRF neural network model, a three-dimensional virtual reality image can be generated in real time, thereby effectively improving the immersion, dynamic sense, and pertinence of the material, and further improving the psychological decompression efficiency and effect.

[0014] In a feasible implementation, the SAM scale data includes dimensions such as pleasantness, arousal, and control.

[0015] In a feasible implementation, the virtual scene information includes natural scenery, natural animals, and warm gathering scenes.

[0016] In a feasible implementation, the specific method of obtaining the first alternative material in the first classification unit includes:

[0017] Through the pre-trained image classification model, the patterns of the materials in the database are identified and classified: materials without elements such as scorching heat, summer, and island are added to the scorching heat environment decompression set; materials without elements such as ice and snow, and grass are added to the severe cold environment decompression set; materials without elements such as darkness, narrowness, tunnels, and crowding are added to the narrow space decompression set.

[0018] Then, based on the virtual scene information of the subject, the materials in the corresponding decompression set are used as the first alternative material.

[0019] In this way, the materials can be automatically classified, avoiding the special environment workers from being exposed to pattern materials that can easily cause them to be highly nervous, thereby ensuring that they start psychological decompression in a calm state of mind.

[0020] Of course, when the number of materials in the database is small, the materials can also be classified by manual means.

[0021] In one feasible implementation, the specific method for calculating the decompression index in the second classification unit includes:

[0022] Step a1: Construct evaluation indicators, specifically including SAM index, heart rate coefficient, and EEG index:

[0023] The SAM index is used to evaluate the subject's subjective degree of relaxation. Its construction method includes converting the relaxed PAD value in the PAD emotional scale into a positive value range to obtain a standard relaxed value; collecting the subject's SAM scale data for different materials; and calculating the Euclidean distance d between the SAM scale data and the standard relaxed value as the SAM index (the smaller the value, the more relaxed the subject). The specific formula can be:

[0024] ;

[0025] Where x0, y0, and z0 represent the standard values ​​of ease in the x, y, and z directions, respectively; xd, yd, and zd represent the SAM scale data values ​​in the x, y, and z directions, respectively.

[0026] The heart rate coefficient is used to evaluate the subject's level of mental stability. The more unstable the heart rate, the greater the fluctuation in the heart rate coefficient, indicating that the subject's mental state is more unstable. The specific calculation formula can be:

[0027] ;

[0028] in, Indicates the first The first subject regarding the first Heart rate coefficient of each material; Indicates the first The first subject regarding the first The standard deviation of heart rate for each data point; Indicates the first The first subject regarding the first The average heart rate of each sample;

[0029] The The specific calculation formula can be:

[0030] ;

[0031] in, This indicates the statistical number of subjects; Indicates the first The first subject regarding the first Heart rate of the data; Indicates the first The preset average heart rate of each material;

[0032] The specific formula of the average heart rate coefficient of the

[0033]

[0034] In addition, for the The specific formula of the average heart rate coefficient of the

[0035]

[0036] The electroencephalogram index is used to evaluate the objective emotional state of the subject, and the construction method comprises the following steps: through the electroencephalogram data processing software, the electroencephalogram data is preprocessed to reduce interference noise; through ERP component analysis, the electroencephalogram data is divided according to the emotional type and quantitatively processed into an emotional value, for example, the emotional type is divided into tension, neutral and relaxation, then the emotional value of tension is set to-1, the emotional value of neutral is set to 0, and the emotional value of relaxation is set to 1; according to the measurement time, the electroencephalogram data is evenly divided into several time periods, and then the emotional values in each time period are sorted according to the size, then the average value of the emotional values of the The specific formula of the average value of the emotional values of the

[0037]

[0038] Among them, represents the maximum emotional value; represents the median emotional value; represents the minimum emotional value;

[0039] For each material, the average value of the emotional values of the subject is calculated, and the maximum and minimum normalization method is used to map it to the [0, 1] interval, and the specific formula can be:

[0040]

[0041] Among them, represents the original value of the current emotional value; represents the minimum value in the average emotional value corresponding to each material; represents the maximum value in the average emotional value corresponding to each material; represents the conversion result of the current emotional value;

[0042] Step a2, divide the electroencephalogram index into a positive index, divide the SAM index and the heart rate coefficient into a reverse index, and perform positive processing respectively to obtain a positive matrix : ​​​​​​​​

[0043] The formula for positiveizing positive indicators is:

[0044] ;

[0045] The formula for turning a contrarian indicator into a positive one is:

[0046] ;

[0047] in, Indicates the first The first (row) of materials Indicators in multiple dimensions The value range is 1 to (Total number of materials) The value range is 1 to 3 (i.e., EEG index, SAM index and heart rate coefficient). express Positive values; Indicates the first The minimum specification for each material; Indicates the first The maximum indicator for each material;

[0048] Step a3: Transform the matrix Standardization is performed to obtain the standardized matrix. The formula for calculating the elements of a normalized matrix can be:

[0049] ;

[0050] Step a4: Calculate the normalization matrix The Middle Entropy of each dimension indicator The specific formula can be:

[0051] ;

[0052] in, Indicates the first The first dimension The proportion of a material's indicator value to all its indicator values ​​is calculated using the following formula:

[0053] ;

[0054] The verification coefficient is represented by the following formula:

[0055] ;

[0056] Step a5, calculate the first Difference coefficients in each dimension The specific formula can be:

[0057] ;

[0058] The weight of the first dimension is recalculated The specific formula can be:

[0059] ;

[0060] The comprehensive score of the first material is then calculated The specific formula can be:

[0061] ;

[0062] Step a6, the maximum value of each dimension in the standardized matrix is used to form the optimal decompression vector ; the minimum value of each dimension in the standardized matrix is used to form the worst decompression vector ;

[0063] Step a7, the distance between the first material and the optimal decompression vector and the distance between the first material and the worst decompression vector are defined respectively, and the specific formula can be:

[0064] ;

[0065] ;

[0066] Step a8, the decompression index of the first material is calculated , with a value range of [0, 1]: when is closer to 1, the decompression effect of the corresponding material is better; when is closer to 0, the decompression effect of the corresponding material is worse; the specific formula can be:

[0067] ;

[0068] In this way, the relationship between information entropy and the probability of a random event can be used to measure the degree of dispersion of an indicator, and the greater the degree of dispersion of an indicator, the greater the impact of the indicator on the comprehensive evaluation, and the greater the weight. Then, the decompression index based on data and ideal conditions is obtained through comprehensive evaluation.

[0069] ​​​​​​​In an implementable embodiment, the training method of the prediction model comprises:

[0070] Based on the labeled decompression index of the material, a training set and a test set are constructed; based on the training set and the test set, a prediction model (such as an RNN model, an LSTM model, and an attention model, etc.) is trained and tested for predicting the decompression effect of the material.

[0071] In an implementable embodiment, the loss function of the prediction model includes a decompression index, so that the prediction model can be iterated according to the decompression effect.

[0072] In an implementable embodiment, the NeRF neural network model belongs to the prior art, which includes, during model training, establishing a local polar coordinate system at a virtual viewpoint V on the surface of the material, performing a ±30° lateral fan perturbation on the original training line of sight OV perpendicular to the surface of the material, and generating a plurality of random lines of sight OV' in real time through an RRC (random ray casting) strategy, and adding them to the training set to train the NeRF neural network model, so as to improve the display effect of the three-dimensional virtual reality image under the condition of small-angle transformed viewing angle; in this way, the three-dimensional virtual reality image can meet the visual clarity requirement of the subject when observing the material by slightly swinging the head during psychological decompression treatment in a lying position.

[0073] In a second aspect, based on the same inventive concept, the present application also provides a special environment worker psychological decompression method using the above-mentioned system, which specifically comprises the following steps:

[0074] Step b1, collecting the initial SAM scale data and virtual scene information of the subject;

[0075] Step b2, selecting a first candidate material in the database based on the virtual scene information; and then playing a virtual reality image to the subject through the NeRF neural network model;

[0076] Step b3, collecting the current SAM scale data, electrocardiogram data, and electroencephalogram data of the subject;

[0077] Step b4, calculating the decompression index of the current material based on the current SAM scale data, electrocardiogram data, and electroencephalogram data, for evaluating the decompression effect of the current material on the current subject and marking the current material; selecting a second candidate material in the database through the prediction model; and then playing a virtual reality image to the subject through the NeRF neural network model, and iteratively executing step b3 until the iteration end condition is reached.

[0078] Through the above steps, the virtual scene information, the SAM scale data, the electrocardiogram data and the electroencephalogram data of the subject can be used to select the material through the pre-trained prediction model; and the three-dimensional psychological stress relief scene can be generated in real time through the NeRF neural network model, so as to effectively improve the immersion, dynamic and pertinence of the material, and further improve the psychological stress relief effect and efficiency.

[0079] In a feasible implementation, the iteration end condition is that the SAM scale data, the electrocardiogram data and the electroencephalogram data of the subject meet the SAM scale data range, the electrocardiogram data range and the electroencephalogram data range defined by the Chinese emotional material library and / or the international emotional material library for the relaxed psychological state.

[0080] In a third aspect, based on the same inventive concept, the application also provides a special environment worker psychological stress relief device for implementing the above system, comprising a VR head-mounted display, an ECG device, an EEG device and a computing platform which are electrically connected to each other.

[0081] The VR head-mounted display is used to display a VR scene.

[0082] The ECG device refers to an electrocardiogram monitoring device, which is used to monitor the heart rate in real time.

[0083] The EEG device refers to an electroencephalogram monitoring device, which is used to record the brain waves in real time.

[0084] The computing platform comprises a processor, a memory and a bus, the memory stores instructions and data read by the processor, the processor is used to call the instructions and data in the memory, and the bus is connected between each functional component for transmitting information; in this way, the device can be realized through a (wearable) device and a (mobile) application, so as to facilitate the subject to perform self-psychological stress relief treatment at any time.

[0085] In a feasible implementation, the device comprises a shelter, and multiple sets of stress relief seats are arranged in parallel in the shelter, so that multiple special environment worker subjects can simultaneously perform psychological stress relief treatment, which can avoid the feeling of loneliness and the fear in a closed environment when a single subject performs self-stress relief.

[0086] In a feasible implementation, the VR head-mounted display, the ECG device, the EEG device and the computing platform are arranged at each stress relief seat; the computing platforms are electrically connected to each other for data interaction, so that each subject can see each other through the VR head-mounted display, which is conducive to relaxing the subject's state of mind and eliminating the feeling of loneliness and fear.

[0087] In a feasible implementation, the shelter further comprises an air conditioning system electrically connected with the computing platforms; the air conditioning system comprises a plurality of independent air outlets, each air outlet being directed to a decompression seat; in this way, the air supply speed, air supply temperature and air supply humidity can be configured according to the material arranged to be played by each computing platform, which is conducive to increasing the sense of immersion.

[0088] In a feasible implementation, the shelter further comprises a sound system electrically connected with the computing platforms; the sound system comprises a plurality of independent earphones; in this way, music and sound effects can be configured according to the material arranged to be played by each computing platform, which is conducive to increasing the sense of immersion.

[0089] In a feasible implementation, the inner wall of the shelter is provided with sound insulation material, which is conducive to excluding external interference and increasing the sense of immersion of the subjects.

[0090] With the above technical solution, the present application has the following beneficial effects:

[0091] The special environment worker psychological decompression system and device provided by the present application can select appropriate two-dimensional materials through targeted classification based on the scale data, virtual scene data and physiological data of the special environment worker subjects; then, the three-dimensional virtual reality image is generated in real time through the NeRF neural network model, so that the sense of immersion, dynamic sense and pertinence of the material can be effectively improved, and the psychological decompression efficiency and effect are improved; at the same time, the virtual reality scene of the special environment worker multi-person companion decompression atmosphere and various sensory experiences can reduce the psychological pressure brought by the psychological decompression treatment behavior of the subjects themselves. BRIEF DESCRIPTION OF DRAWINGS

[0092] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the following specific embodiment or prior art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0093] Figure 1 A special environment worker psychological decompression system diagram is provided for the embodiments of the present application;

[0094] Figure 2 A specific method flowchart for calculating the decompression index is provided for the embodiments of the present application;

[0095] Figure 3 A special environment worker psychological decompression method flowchart is provided for the embodiments of the present application;

[0096] Figure 4A shelter internal layout schematic diagram provided by the embodiment of the present application;

[0097] Figure 5 An airbag layout schematic diagram provided by the embodiment of the present application.

[0098] Reference signs:

[0099] 1 - decompression seat; 11 - lower airbag; 12 - upper airbag; 2 - ECG device; 3 - EEG device; 4 - air outlet; 5 - VR head-mounted display. DETAILED DESCRIPTION

[0100] The technical solutions of the present application will be described below in conjunction with the drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0101] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second", "third" are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0102] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0103] The present application will be further explained in conjunction with the specific embodiments.

[0104] It should also be noted that the following specific embodiments or specific embodiments are a series of optimized setting methods listed by the present application to further explain the specific invention content, and these setting methods can be used in combination or in association with each other.

[0105] Embodiment one:

[0106] For example, Figure 1As shown, the embodiment provides a special environment worker psychological decompression system, comprising a data acquisition module, a data processing module and a virtual reality generation module arranged in sequence:

[0107] The data acquisition module comprises a SAM unit, a virtual scene unit, an electrocardio unit and an electroencephalogram unit, which are respectively used to collect SAM scale data, virtual scene information (i.e. information about special working environment), electrocardio data and electroencephalogram data of the subject and send them to the data processing module.

[0108] The data processing module comprises a database, a first classification unit, a second classification unit and a prediction model unit.

[0109] The database can store two-dimensional picture materials and corresponding sound effect materials in the Chinese emotion material library and / or the international emotion material library with valence above 6.5, arousal below 3.5 and dominance between 5-7.

[0110] The first classification unit filters the first candidate materials based on the virtual scene information.

[0111] The second classification unit calculates a decompression index based on the SAM scale data, the electrocardio data and the electroencephalogram data, which is used to evaluate the decompression effect of the current material in the database on the current subject and mark the current material.

[0112] The prediction model unit stores a prediction model, which is used to predict the decompression effect of the unmarked material in the database on the current subject, and the decompression effect includes positive decompression, zero decompression and negative decompression, and the first candidate material with positive decompression is sent to the virtual reality generation module as the second candidate material.

[0113] The virtual reality generation module comprises a visual generation unit, which generates a virtual reality image based on the second candidate material through a NeRF neural network model; the virtual reality image is used for image-assisted positive decompression for the current subject.

[0114] Through the above system, the scale data, virtual scene data and physiological data of the special environment worker subject can be classified and targeted, and appropriate two-dimensional materials can be selected; then through the NeRF neural network model, a three-dimensional virtual reality image can be generated in real time, so that the immersion, dynamic and targeting of the materials can be effectively improved, and the psychological decompression efficiency and effect can be improved.

[0115] Further, the SAM scale data includes valence, arousal and control sense.

[0116] Further, the virtual scene information includes natural scenery, natural animals and warm gathering scenes.

[0117] Furthermore, the specific method for obtaining the first candidate material in the first classification unit includes:

[0118] Using a pre-trained image classification model, the images in the database are identified and classified: images without elements of heat, summer, or islands are added to the heat environment stress relief set; images without elements of ice and snow or weeds are added to the cold environment stress relief set; and images without elements of darkness, narrowness, tunnels, or crowds are added to the narrow space stress relief set.

[0119] Then, based on the virtual scene information of the subjects, the materials in the corresponding stress-relief sets were selected as the first candidate materials;

[0120] In this way, materials can be automatically categorized, preventing workers in special environments from coming into contact with graphic materials that could easily cause them high levels of stress, thus ensuring that they can begin stress reduction in a peaceful state of mind.

[0121] Of course, when the number of materials in the database is small, the materials can also be classified manually as described above.

[0122] Furthermore, such as Figure 2 As shown, the specific method for calculating the decompression index in the second classification unit includes:

[0123] Step a1: Construct evaluation indicators, specifically including SAM index, heart rate coefficient, and EEG index:

[0124] The SAM index is used to evaluate the subject's subjective degree of relaxation. Its construction method includes converting the relaxed PAD value in the PAD emotional scale into a positive value range to obtain the relaxed standard value (x0; y0; z0); collecting the subject's SAM scale data (xd; yd; zd) for different materials; and calculating the Euclidean distance d between the SAM scale data and the relaxed standard value as the SAM index (the smaller the value, the more relaxed the subject). The specific formula can be:

[0125] ;

[0126] The heart rate coefficient is used to evaluate the subject's level of mental stability. The more unstable the heart rate, the greater the fluctuation in the heart rate coefficient, indicating that the subject's mental state is more unstable. The specific calculation formula can be:

[0127] ;

[0128] in, Indicates the first The first subject regarding the first Heart rate coefficient of each material; Indicates the first The first subject regarding the first The standard deviation of heart rate for each data point; Indicates the first The first subject regarding the first The average heart rate of each sample;

[0129] The The specific calculation formula can be:

[0130] ;

[0131] in, This indicates the statistical number of subjects; Indicates the first The first subject regarding the first Heart rate of the data; Indicates the first The preset average heart rate of each material;

[0132] The The specific calculation formula can be:

[0133] ;

[0134] Furthermore, for the first Average heart rate coefficient of each material The specific formula can be:

[0135] ;

[0136] The EEG index is used to evaluate the objective emotional state of the subject. Its construction method includes: preprocessing the EEG data using EEG data processing software to reduce interference noise; classifying and quantifying the EEG data into emotional values ​​according to emotion type using ERP component analysis (e.g., classifying emotions into tension, neutrality, and relaxation, setting the emotional value for tension to -1, neutrality to 0, and relaxation to 1); dividing the EEG data into several time periods based on the measurement time, and sorting them according to the magnitude of the emotional value in each time period, then calculating the... The first subject regarding the first The average sentiment of each piece of material The specific formula is:

[0137] ;

[0138] in, This represents the highest emotional value; This represents the median sentiment value. Represents the minimum emotional value;

[0139] For each data set, the average emotional value of the participants is calculated and mapped to the [0,1] interval using an extremum normalization method. The specific formula is as follows:

[0140] ;

[0141] in, This represents the original value of the current sentiment score. This represents the minimum average sentiment value among all the materials. This represents the maximum value among the average sentiment values ​​corresponding to each piece of material. This indicates the current sentiment value conversion result;

[0142] Step a2: Classify the EEG index as a positive indicator and the SAM index and heart rate coefficient as negative indicators, and perform positive transformation on each to obtain a positive transformation matrix. :

[0143] The formula for positiveizing positive indicators is:

[0144] ;

[0145] The formula for turning a contrarian indicator into a positive one is:

[0146] ;

[0147] in, Indicates the first The first (row) of materials Indicators in multiple dimensions The value range is 1 to (Total number of materials) The value range is 1 to 3 (i.e., EEG index, SAM index and heart rate coefficient). express Positive values; Indicates the first The minimum specification for each material; Indicates the first The maximum indicator for each material;

[0148] Step a3: Transform the matrix Standardization is performed to obtain the standardized matrix. The formula for calculating the elements of a normalized matrix can be:

[0149] ;

[0150] Step a4: Calculate the normalization matrix The Middle Entropy of each dimension indicator The specific formula can be:

[0151] ;

[0152] in, Indicates the first The first dimension The proportion of a material's indicator value to all its indicator values ​​is calculated using the following formula:

[0153] ;

[0154] The verification coefficient is represented by the following formula:

[0155] ;

[0156] Step a5, calculate the first Difference coefficients in each dimension The specific formula can be:

[0157] ;

[0158] Calculate the next Weights of each dimension The specific formula can be:

[0159] ;

[0160] Then calculate the first... Overall score of each material The specific formula can be:

[0161] ;

[0162] Step a6: Standardize the matrix Maximum value of each dimension of the index , forming the optimal decompression vector ; standardize the matrix Minimum value of each dimension of the index This forms the worst-case decompression vector. ;

[0163] Step a7: Define the first Distance between each material and the optimal decompression vector Passing the exam Distance between each material and the worst decompression vector The specific formula can be:

[0164] ;

[0165] ;

[0166] Step a8, calculate the decompression index of the material, whose value range is [0, 1]: when closer to 1, the decompression effect of the corresponding material is better; when closer to 0, the decompression effect of the corresponding material is worse; the specific formula can be:

[0167] ;

[0168] In this way, the relationship between information entropy and the probability of a random event can be used to measure the dispersion degree of an index. The greater the dispersion degree of the index, the greater the influence of the index on the comprehensive evaluation, and the greater the weight. Then, the decompression index based on data and ideal conditions is obtained by comprehensive evaluation.

[0169] Further, the training method of the prediction model comprises:

[0170] Based on the labeled decompression index of the material, a training set and a test set are constructed. Based on the training set and the test set, the prediction model (such as an RNN model, an LSTM model, and an attention model) is trained and tested for predicting the decompression effect of the material.

[0171] Further, the loss function of the prediction model includes the decompression index, so that the prediction model can be iterated according to the decompression effect.

[0172] Further, the NeRF neural network model belongs to the prior art, which includes, during model training, establishing a local polar coordinate system at a virtual viewpoint V on the surface of the material, performing ±30° transverse fan perturbation on the original training line of sight OV perpendicular to the surface of the material, and generating a plurality of random lines of sight OV' in real time through the RRC (random ray casting) strategy, and adding them to the training set to train the NeRF neural network model, so as to improve the display effect of the three-dimensional virtual reality image under the condition of small-angle transformation of the viewing angle. In this way, the three-dimensional virtual reality image can meet the visual clarity requirement of the subject when observing the material by slightly swinging the head during psychological decompression treatment in a lying position.

[0173] Embodiment Two:

[0174] As shown in the figure, the embodiment provides a special environment worker psychological decompression method using the above system, which specifically comprises the following steps: Figure 3

[0175] Step b1, collecting the initial SAM scale data and virtual scene information of the subject;

[0176] ​​Step b2, based on the virtual scene information, selecting a first candidate material in the database; and then constructing a virtual reality image through the NeRF neural network model and playing it to the subject;

[0177] Step b3, collecting the current SAM scale data, ECG data and EEG data of the subject;

[0178] Step b4, based on the current SAM scale data, ECG data and EEG data, calculating the stress reduction index of the current material, which is used to evaluate the stress reduction effect of the current material on the current subject and mark the current material; selecting a second candidate material in the database through the prediction model; and then reconstructing a virtual reality image through the NeRF neural network model and playing it to the subject, and iteratively executing step b3 until the iteration end condition is reached.

[0179] Through the above steps, the virtual scene information, SAM scale data, psychological data and EEG data of the subject can be used to select materials through the pre-trained prediction model; and then a three-dimensional psychological stress reduction scene can be generated in real time through the NeRF neural network model, so as to effectively improve the immersion, dynamic and pertinence of the materials, and further improve the psychological stress reduction effect and efficiency.

[0180] Further, the iteration end condition is that the SAM scale data, ECG data and EEG data of the subject meet the SAM scale data range, ECG data range and EEG data range defined in the Chinese and international emotional material library for relaxed psychological state.

[0181] Embodiment three:

[0182] The embodiment provides a special environment worker psychological stress reduction device for implementing the above system, which comprises a VR head-mounted display, an ECG device, an EEG device and a computing platform which are electrically connected to each other.

[0183] The VR head-mounted display is used for displaying a VR scene, and a conventional device can be used.

[0184] The ECG device refers to an electrocardiogram monitoring device, which is attached to the chest of the subject, and is used for real-time monitoring of heart rate, and a conventional device can be used.

[0185] The EEG device refers to a brain wave monitoring device, which is worn on the head of the subject, and is used for real-time recording of brain waves, and a conventional device can be used.

[0186] The computing platform includes a processor, a memory, and a bus. The memory stores instructions and data read by the processor, and the processor is used to call the instructions and data in the memory. The bus connects the various functional components for information transmission. This can be achieved through (wearable) devices and (mobile) applications, thereby facilitating subjects to perform self-psychological stress reduction therapy at any time.

[0187] Example 4:

[0188] like Figure 4 As shown in the figure, based on the third embodiment, this embodiment provides a cabin, in which three sets of decompression seats 1 are arranged in parallel so that three subjects working in special environments can simultaneously undergo psychological decompression treatment. This can avoid the feelings of loneliness that a single subject may experience when decompressing on their own and the feelings of fear that may arise in a confined environment.

[0189] Furthermore, a VR headset 5, an ECG device 2, an EEG device 3, and a computing platform are installed at the head of each decompression seat 1; the computing platforms are electrically connected to each other for data interaction, so that the three subjects can see each other through the VR headset 5, which helps the subjects relax and eliminate feelings of loneliness and fear.

[0190] Furthermore, the cabin also includes an air conditioning system, which is electrically connected to the computing platform. The air conditioning system includes three independent air outlets 4, each of which is located above the head of a decompression seat 1 and points towards the decompression seat 1. This allows the airflow speed, airflow temperature, and airflow humidity to be configured according to the materials played on each computing platform, which helps to increase the sense of immersion.

[0191] Furthermore, the cabin also includes an audio system electrically connected to the computing platform; the audio system includes several independent headphones; this allows music and sound effects to be configured according to the materials played on each computing platform, which helps to increase the sense of immersion.

[0192] Furthermore, the inner walls of the cabin are equipped with soundproofing materials, which helps to eliminate external interference and increase the immersion of the test subjects.

[0193] Example 5:

[0194] like Figure 5 As shown, based on Embodiment 4, a lower airbag 11 and an upper airbag 12 are also provided on the decompression seat 1, which are used to wrap the subject's torso in an upper and lower combination manner, and can independently adjust the air pressure according to the control of the computing platform.

[0195] The lower air bag 11 is laid on the upper surface of the backrest of the decompression chair 1; one end of the upper air bag 12 is connected to the backrest side wall of the decompression chair 1 through a hinge, which can be opened and closed in alignment with the lower air bag 11 and locked; in this way, when the upper air bag 12 is opened, the subject can lean on the lower air bag 11, and when the upper air bag 12 is folded, positioned and locked, the subject's body can be wrapped.

[0196] Further, the upper air bag 12 is arranged in a semi-cylindrical cover, the upper end of the cover is provided with a first opening for the neck of the subject to pass through, and the lower end of the cover is provided with a second opening for the torso of the subject to pass through.

[0197] Further, the lower air bag 11 and the upper air bag 12 each include an air pump and a gas pressure sensor, which are electrically connected to the computing platform respectively, for program-controlled air pressure of each air bag; in this way, the bondage and pressure on the body of the special environment worker wearing different special environment work clothes (such as fire-fighting clothes, anti-explosion clothes, mine clothes, etc.) can be accurately simulated for psychological decompression treatment, and the air pressure of each air bag is controlled by program, which is combined with virtual reality images, audio and other psychological decompression means to assist in psychological decompression.

[0198] Further, the lower air bag 11 and the upper air bag 12 each include a plurality of parallel and independent cylindrical air bags, each of which includes an air pump and a gas pressure sensor, so as to more evenly apply wrapping force to the subject, and the wearing feeling of different special environment work clothes is simulated by adjusting the air pressure of different cylindrical air bags.

[0199] For example, at the initial stage of psychological decompression, the air pressure of each air bag is increased to make the subject feel as if wearing heavy special environment work clothes; as the psychological decompression proceeds, the air pressure of each air bag is gradually reduced to make the subject feel relieved, thereby helping to improve the effect of psychological decompression. Of course, the air pressure can also be adjusted in combination with the materials in the virtual reality images to enhance the sense of immersion and further improve the effect of psychological decompression. It should be noted that the air bag should be used separately from the ECG device 2 to avoid damage or failure of the electrocardiogram monitoring patch (electrode) attached to the subject's chest.

[0200] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not limited thereto; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions described in the foregoing embodiments can still be modified, or some or all of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A psychological stress reduction system for workers in special environments, characterized in that, This includes a data acquisition module, a data processing module, and a virtual reality generation module, arranged sequentially: The data acquisition module includes a SAM unit, a virtual scene unit, an electrocardiogram (ECG) unit, and an electroencephalogram (EEG) unit, which are used to collect the subject's SAM scale data, virtual scene information, ECG data, and EEG data, respectively, and send them to the data processing module. The data processing module includes a database, a first classification unit, a second classification unit, and a prediction model unit. The database stores two-dimensional image materials and corresponding sound effect materials; The first classification unit filters first candidate materials based on virtual scene information; The second classification unit calculates a stress reduction index based on SAM scale data, electrocardiogram data, and electroencephalogram data. This index is used to evaluate the stress reduction effect of the current material in the database on the current subject and to label the current material. The prediction model unit stores a prediction model for predicting the stress reduction effect of unlabeled materials in the database on the current subject. The stress reduction effect includes positive stress reduction, zero stress reduction, and negative stress reduction. The first candidate material for positive stress reduction is sent to the virtual reality generation module as the second candidate material. The virtual reality generation module includes a visual generation unit that generates virtual reality images based on a second alternative material using a NeRF neural network model; the virtual reality images are used for image-assisted positive stress reduction for the current subject. The system includes a psychological stress relief device for workers in special environments, comprising a VR headset, an ECG device, an EEG device, a computing platform, and a modular cabin that are electrically connected to each other. The VR headset is used to display VR scenes; The ECG device is used to monitor heart rate in real time; The EEG device is used to record brain waves in real time; Multiple decompression seats are arranged in parallel inside the cabin; each decompression seat is equipped with a VR headset, ECG equipment, EEG equipment and a computing platform; the computing platforms are electrically connected to each other for data interaction. The decompression seat is equipped with a lower airbag and an upper airbag, which are used to wrap the subject's torso in an upper and lower combination, and the air pressure can be independently adjusted according to the calculation platform. The lower airbag is laid on the upper surface of the backrest of the decompression seat; one end of the upper airbag is connected to the side wall of the backrest of the decompression seat by a hinge, so as to be able to open and close and lock with the lower airbag. The upper airbag is set in a semi-cylindrical cover, with a first opening at the upper end of the cover for the subject's neck to pass through, and a second opening at the lower end of the cover for the subject's torso to pass through. Both the lower airbag and the upper airbag contain several parallel and independent cylindrical airbags, each of which includes an air pump and a pressure sensor; the air pump and the pressure sensor are electrically connected to the computing platform.

2. The system according to claim 1, characterized in that, The specific method for obtaining the first candidate material in the first classification unit includes: Using a pre-trained image classification model, the images in the database are identified and classified: images without elements of heat, summer, or islands are added to the heat environment stress relief set; images without elements of ice and snow or weeds are added to the cold environment stress relief set; and images without elements of darkness, narrowness, tunnels, or crowds are added to the narrow space stress relief set. Then, based on the virtual scene information of the subjects, the materials in the corresponding stress-relief sets were selected as the first alternative materials.

3. The system according to claim 1, characterized in that, The specific methods for calculating the decompression index in the second classification unit include: Step a1: Construct evaluation indicators, specifically including SAM index, heart rate coefficient, and EEG index: The SAM index is used to evaluate the subject's subjective degree of relaxation. Its construction method includes converting the relaxed PAD value from the PAD emotional scale into a positive value range to obtain a standard relaxed value; collecting the subject's SAM scale data for different materials; and calculating the Euclidean distance d between the SAM scale data and the standard relaxed value as the SAM index. The specific formula is as follows: ； Where x0, y0, and z0 represent the standard ease values ​​in the x, y, and z directions, respectively; xd, yd, and zd represent the SAM scale data values ​​in the x, y, and z directions, respectively. The heart rate coefficient is used to evaluate the subject's mental stability; the more unstable the heart rate, the greater the fluctuation in the heart rate coefficient. The specific calculation formula is as follows: ； in, Indicates the first The first subject regarding the first Heart rate coefficient of each material; Indicates the first The first subject regarding the first The standard deviation of heart rate for each data point; Indicates the first The first subject regarding the first The average heart rate of each sample; The The specific calculation formula is as follows: ； in, This indicates the statistical number of subjects; Indicates the first The first subject regarding the first Heart rate of the data; Indicates the first The preset average heart rate of each material; The The specific calculation formula is as follows: ； Furthermore, for the first Average heart rate coefficient of each material The specific formula is: ； The EEG index is used to evaluate the objective emotional state of the subject. Its construction method includes: preprocessing EEG data using EEG data processing software; classifying and quantifying the EEG data into emotional values ​​according to emotional type using ERP component analysis; dividing the EEG data into several time periods based on the measurement time, and sorting them according to the magnitude of the emotional value in each time period, then calculating the... The first subject regarding the first The average sentiment of each piece of material The specific formula is: ； in, This represents the highest emotional value; This represents the median sentiment value. Represents the minimum emotional value; For each data set, the average emotional value of the participants was calculated and mapped to the [0,1] interval using an extremum normalization method. The specific formula is as follows: ; in, This represents the original value of the current sentiment score. This represents the minimum average sentiment value among all the materials. This represents the maximum value among the average sentiment values ​​corresponding to each piece of material. This indicates the current sentiment value conversion result; Step a2: Classify the EEG index as a positive indicator and the SAM index and heart rate coefficient as negative indicators, and perform positive transformation on each to obtain a positive transformation matrix. : The formula for positiveizing positive indicators is: ； The formula for turning a contrarian indicator into a positive one is: ； in, Indicates the first The first material Indicators in each dimension The value range is 1 to , This indicates the total number of materials. The value range is from 1 to 3; express Positive values; Indicates the first The minimum specification for each material; Indicates the first The maximum indicator for each material; Step a3: Transform the matrix Standardization is performed to obtain the standardized matrix. The formula for calculating the elements of a standardized matrix is: ; Step a4: Calculate the normalization matrix The Middle Entropy of each dimension indicator The specific formula is as follows: ; in, Indicates the first The first dimension The proportion of a material's indicator value to all its indicator values ​​is calculated using the following formula: ； The verification coefficient is represented by the following formula: ； Step a5, calculate the first Difference coefficients in each dimension The specific formula is as follows: ； Calculate the next Weights of each dimension The specific formula is: ; Then calculate the first... Overall score of each material The specific formula is as follows: ； Step a6: Standardize the matrix Maximum value of each dimension of the index , forming the optimal decompression vector ; standardize the matrix Minimum value of each dimension of the index This forms the worst-case decompression vector. ; Step a7: Define the first Distance between each material and the optimal decompression vector Passing the exam Distance between each material and the worst decompression vector The specific formula is as follows: ； ； Step a8, calculate the first The stress-relief index of each material Its range is [0,1]; the specific formula is: 。 4. The system according to claim 3, characterized in that, The training method of the prediction model includes: constructing a training set and a test set based on the labeled stress reduction index; training and testing the prediction model based on the training set and the test set to predict the stress reduction effect of the material; the loss function of the prediction model includes the stress reduction index.

5. The system according to claim 3, characterized in that, During model training, the NeRF neural network model establishes a local polar coordinate system at the virtual viewpoint V on the material surface, performs a ±30° lateral fan-shaped perturbation on the original training line of sight OV perpendicular to the material surface, and generates several random line of sight OV' in real time through the RRC strategy, which are then added to the training set to train the NeRF neural network model.

Images