A shooter psychological state assessment method, system and storage medium based on deep learning

Abstract

The present application discloses a method, system and storage medium for evaluating the psychological state of a shooter based on deep learning, including obtaining the physiological data and shooting technical data of a shooter; preprocessing and extracting features from the acquired physiological data and shooting technical data; training a constructed deep learning model using a convolutional neural network and a long short-term memory network based on the extracted features; evaluating the psychological state of the shooter through the trained deep learning model, and outputting psychological stability and growth evaluation indicators. By introducing deep learning technology and combining physiological data with shooting technical data, the present application can more comprehensively and accurately monitor and analyze the psychological stability and growth of shooters, which will provide a scientific basis for shooting training.

Description

Technical Field

[0001] The present application relates to the technical field of smart shooting range management, and in particular to a shooter psychological state assessment method, system and storage medium based on deep learning. Background Art

[0002] In modern shooting training, the comprehensive quality of the shooter, especially the psychological quality, is considered to be the key factor affecting the shooting effect and mission success. With the development of science and technology, how to scientifically monitor and evaluate the shooter's psychological state has become an important topic in shooting training.

[0003] The main challenges in assessing the shooter's mental state currently include:

[0004] Limitations of traditional assessment methods:

[0005] The current methods for assessing shooter psychological states mainly rely on traditional psychological assessment tools, such as regular psychological assessment questionnaires and interviews. These methods are often subjective, time-consuming, and rely on the reliability of self-reports. In shooting and training, shooters may not have the opportunity or willingness to conduct such assessments, and they cannot reflect the changes in instantaneous emotions and psychological states in a timely manner. In a shooting environment that requires agile responses, this type of assessment is not flexible enough.

[0006] Initial application of physiological monitoring technology:

[0007] With the popularity of smart devices, shooting training institutions or organizations have begun to use these technologies to monitor the shooter's physiological state, such as heart rate, heart rate variability (HRV), skin galvanic response, etc. Although these parameters can provide some indication of the shooter's psychological state, there are currently only a few systems that can effectively combine these physiological data with the shooter's shooting performance (such as shooting accuracy, reaction time, etc.) for comprehensive analysis. This makes it impossible to evaluate the shooter's overall quality from a more comprehensive perspective.

[0008] Application limitations of smart shooting range management system:

[0009] Smart range systems are increasingly used in shooting training, with the goal of improving shooters' training efficiency and shooting ability through automated monitoring and analysis. However, existing smart range management systems usually only focus on shooters' technical indicators, such as hit rate, firing time, accuracy, etc., and do not effectively evaluate and feedback their psychological state. The complex relationship between psychological stability and shooting performance has not been deeply studied and verified, so there is a lack of a comprehensive evaluation method.

[0010] Psychological intervention mechanisms that are urgently needed:

[0011] In actual shooting training, psychological intervention has been proven to be effective in improving shooters' shooting ability and maintaining a good mentality. If the shooter's psychological state can be combined with the assessment, psychological fluctuations can be discovered in time, and necessary psychological counseling and training can be given, psychological pressure can be effectively reduced and the effectiveness of shooting and training can be improved. Summary of the invention

[0012] The purpose of this application is to provide a shooter psychological state assessment method, system and storage medium based on deep learning to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.

[0013] This application adopts the following technical solutions to achieve the above invention objectives:

[0014] This application provides a shooter psychological state assessment method based on deep learning, including:

[0015] Acquire the physiological data and shooting technical data of the shooter, wherein the physiological data of the shooter includes heart rate data and heart rate variability data, and the calculation formula of the heart rate variability data is as follows:

[0016] ;

[0017] In the formula, is the interval between consecutive heartbeats, is the average of all consecutive heartbeat intervals, is the total number of observations, is the heart rate variability data, is the number of observations;

[0018] The shooting technical data includes hit rate data, shooting accuracy data and shooting time data. The calculation formula of the hit rate data is as follows:

[0019] ;

[0020] In the formula, is the hit rate, is the number of hits, indicating the number of targets actually hit by the shooter. is the total number of shots;

[0021] The calculation formula of the shooting accuracy data is as follows:

[0022] ;

[0023] In the formula, It is the shooting accuracy, which indicates the distance between the shooter's shooting point and the center of the bull's eye. Indicates the horizontal coordinate of the shooter's actual hit point, The vertical coordinate of the shooter's actual hit point, represents the horizontal coordinate of the bull's eye, Indicates the vertical coordinate of the bull's eye;

[0024] Preprocess and extract features from the acquired physiological data and shooting technical data;

[0025] Based on the extracted features, the constructed deep learning model is trained using convolutional neural networks and long short-term memory networks;

[0026] The shooter's psychological state is evaluated through the trained deep learning model, and psychological stability and growth evaluation indicators are output.

[0027] Furthermore, the acquired physiological data and shooting technical data are preprocessed, including:

[0028] The collected physiological data are cleaned and denoised. The wavelet transform formula for denoising the physiological data is as follows:

[0029] ;

[0030] In the formula, Represents the reconstructed signal, and the wavelet basis function is , is the original signal, is a time index, indicating a specific time point during the measurement record. is the total number of signal sample points;

[0031] Interpolation method was used to complete the missing physiological data and shooting technique data;

[0032] The collected shooting technical data is standardized using Z-score, and the formula is:

[0033] ;

[0034] In the formula, is the standard score, is the original data, is the data mean, is the standard deviation;

[0035] Time series analysis is performed on physiological data and shooting technical data, and the timing features within each time window are extracted using the sliding window method.

[0036] Furthermore, feature extraction is performed on the acquired physiological data and shooting technical data, including:

[0037] Extract heart rate fluctuation and heart rate variability features from physiological data;

[0038] Combining the parameters in the shooting technique data, a feature vector containing information on psychological stability and growth is constructed.

[0039] Furthermore, based on the extracted features, a method of using a convolutional neural network and a long short-term memory network to train the constructed deep learning model includes:

[0040] The local features of shooting technical data and physiological data are extracted through convolutional neural networks, and the input data is processed through convolutional layers and pooling layers to generate feature representations;

[0041] Processing time series data of features through long short-term memory networks to capture long-term dependencies in the data;

[0042] Using physiological data and shooting technique data extracted and annotated from historical datasets, the mean square error is used as the loss function, and the goal is to minimize the difference between the output and the true value;

[0043] The Adam optimization algorithm is used to improve learning efficiency, and the learning rate is adjusted in combination with adaptive momentum estimation to adapt to the gradient size of different features;

[0044] The K-fold cross-validation method was used to evaluate the accuracy and generalization ability of the model.

[0045] Furthermore, the psychological state of the shooter is evaluated through the trained deep learning model, including:

[0046] Psychological stability assessment: A psychological stability index is generated for each shooter through the output of the deep learning model, with a value ranging from 0 to 1. The closer to 1, the more stable the psychological state;

[0047] Growth assessment: Based on deep learning model predictions and historical data, a growth score is constructed to reflect the improvement of the mapper's technical and psychological qualities.

[0048] This application provides a shooter psychological state assessment system based on deep learning, including:

[0049] The acquisition unit is used to acquire the physiological data and shooting technical data of the shooter, wherein the physiological data of the shooter includes heart rate data and heart rate variability data, and the calculation formula of the heart rate variability data is as follows:

[0050] ;

[0051] In the formula, is the interval between consecutive heartbeats, is the average of all consecutive heartbeat intervals, is the total number of observations, is the heart rate variability data, is the number of observations;

[0052] The shooting technical data includes hit rate data, shooting accuracy data and shooting time data. The calculation formula of the hit rate data is as follows:

[0053] ;

[0054] In the formula, is the hit rate, is the number of hits, indicating the number of targets actually hit by the shooter. is the total number of shots;

[0055] The calculation formula of the shooting accuracy data is as follows:

[0056] ;

[0057] In the formula, It is the shooting accuracy, which indicates the distance between the shooter's shooting point and the center of the bull's eye. Indicates the horizontal coordinate of the shooter's actual hit point, The vertical coordinate of the shooter's actual hit point, represents the horizontal coordinate of the bull's eye, Indicates the vertical coordinate of the bull's eye;

[0058] A preprocessing and feature extraction unit, used for preprocessing and feature extraction of acquired physiological data and shooting technical data;

[0059] A training unit, for training the constructed deep learning model using a convolutional neural network and a long short-term memory network based on the extracted features;

[0060] The output unit is used to evaluate the shooter's psychological state through the trained deep learning model and output psychological stability and growth evaluation indicators.

[0061] The present application provides a shooter psychological state assessment system based on deep learning, including a memory and a processor;

[0062] The memory is used to store instructions;

[0063] The processor is used to operate according to the instructions to execute the steps according to the above method.

[0064] The present application provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the above method are implemented.

[0065] The beneficial effects of this application are as follows:

[0066] This application introduces deep learning technology and combines physiological data with shooting technical data to more comprehensively and accurately monitor and analyze the shooter's psychological stability and growth, which will provide a scientific basis for shooting training. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is an overall flow chart of a shooter psychological state assessment method based on deep learning provided according to an embodiment of the present application;

[0068] Figure 2 A specific flow chart of a shooter psychological state assessment method based on deep learning provided according to an embodiment of the present application;

[0069] Figure 3 This is a schematic diagram of the fusion of shooter physiological data and shooting technical data in a shooter psychological state assessment method based on deep learning according to an embodiment of the present application. DETAILED DESCRIPTION

[0070] Reference Figure 1-Figure 3 The purpose of this application is to provide a shooter psychological state assessment method based on deep learning, which can integrate the shooter's wristband physiological data, shooting comprehensive data and historical data, and evaluate the psychological stability and growth of the shooter through a deep learning algorithm. The specific steps are as follows:

[0071] S100, obtaining the physiological data and shooting technical data of the shooter, including:

[0072] Data collection is one of the core steps of this application, which can provide accurate and multi-dimensional data input for the deep learning model. The collected data includes the shooter's physiological data (such as heart rate, heart rate variability, etc.) and technical data during the shooting process (such as hit rate, shooting accuracy, time, etc.).

[0073] S101, physiological data collection:

[0074] The smart bracelet device worn by the shooter is used to collect physiological data in real time. Heart rate data is one of the key indicators for evaluating psychological stability. The bracelet monitors blood flow fluctuations through optical sensors (such as PPG sensors) to calculate the heart rate. To ensure the accuracy of the data, the heart rate data will be acquired through high-frequency sampling and the timestamp of the data will be recorded. In addition, the shooter's heart rate variability (HRV) is monitored. HRV reflects the regulatory ability of the autonomic nervous system and is an effective physiological parameter for evaluating psychological stability. The calculation formula of HRV is as follows:

[0075] ;

[0076] In the formula, is the interval between consecutive heartbeats, is the average of all consecutive heartbeat intervals, is the total number of observations, is the heart rate variability data, is the number of observations.

[0077] S102, shooting technical data collection:

[0078] Shooting technical data is obtained through real-time feedback from sensors and shooting equipment in the shooting range. The following data is recorded every time you shoot:

[0079] Hit rate, calculated as:

[0080] ;

[0081] In the formula, is the hit rate, is the number of hits, indicating the number of targets actually hit by the shooter. is the total number of shots;

[0082] Shooting accuracy is calculated as follows:

[0083] ;

[0084] In the formula, It is the shooting accuracy, which indicates the distance between the shooter's shooting point and the center of the bull's eye. Indicates the horizontal coordinate of the shooter's actual hit point, The vertical coordinate of the shooter's actual hit point, represents the horizontal coordinate of the bull's eye, Indicates the vertical coordinate of the bull's eye.

[0085] Shooting time.

[0086] S103, historical data collection and storage:

[0087] According to the results of each shooting training or competition, historical data is automatically generated and stored, including the shooter's physiological parameters (such as heart rate fluctuations, HRV, etc.), shooting results (such as hit rate, shooting accuracy), etc. These historical data will be used for the training of deep learning models to help better identify the psychological changes and growth trends of players.

[0088] S200, preprocessing and feature extraction of the acquired physiological data and shooting technical data, specifically including:

[0089] After the data is collected, it needs to be preprocessed to ensure data quality and provide effective input for the training and evaluation of deep learning models. The process of data preprocessing usually includes the following steps:

[0090] S201, data cleaning and denoising:

[0091] The purpose of data cleaning is to remove invalid or abnormal data. Since heart rate data is affected by external environment (such as unstable sports equipment, sensor error, etc.), some noise data may appear. Therefore, the first step of data cleaning is to remove obvious outliers. For denoising heart rate data, wavelet transform can be used as follows:

[0092] ;

[0093] In the formula, Represents the reconstructed signal, and the wavelet basis function is , is the original signal, is a time index, indicating a specific time point during the measurement record. is the total number of signal sample points;

[0094] S202, missing value processing:

[0095] In actual applications, some data may be lost due to equipment failure or transmission problems. To solve this problem, we use interpolation to complete the missing heart rate data and shooting technique data. Interpolation can predict the missing part based on the existing data to avoid data deviation.

[0096] S203, feature standardization and normalization:

[0097] For different data features (such as heart rate, shooting accuracy, etc.), their dimensions and numerical ranges may vary greatly. In order to improve the efficiency and effect of model training, all input data will be standardized or normalized. Z-score standardization is used, and its formula is:

[0098] ;

[0099] In the formula, is the standard score, is the original data, is the data mean, is the standard deviation;

[0100] S204, Time Series Analysis:

[0101] Since shooting training and competition is a continuous process, heart rate and shooting technique data have time series characteristics. Therefore, a key task in data preprocessing is to perform time series analysis on the data. The time series features in each time window are extracted by the sliding window method. These features include the average, maximum, minimum, and fluctuation amplitude of the heart rate.

[0102] S205, extracting the characteristics of heart rate fluctuation and heart rate variability from the physiological data, combining them with the parameters in the shooting technical data, and constructing a characteristic vector containing psychological stability and growth information.

[0103] S300, based on the extracted features, uses convolutional neural networks and long short-term memory networks to train the constructed deep learning model, including:

[0104] The application of deep learning technology in this application is centered on building a comprehensive and automatically learning model to explore the relationship between the physiological data and shooting technical data of shooters, thereby evaluating the psychological stability and growth of the shooters.

[0105] S301, model architecture design:

[0106] The deep learning model of this application is based on a multi-layer neural network, combining a convolutional neural network (CNN) and a long short-term memory network (LSTM). The main reason is that CNN is suitable for extracting local features from images or one-dimensional sequence data, while LSTM is good at processing time series data, especially capable of capturing long-term dependencies in the data.

[0107] Convolutional Neural Network (CNN): Mainly used to extract local features of shooting technical data and physiological data (such as heart rate fluctuations), and process the input data through convolutional layers and pooling layers to generate meaningful feature representations.

[0108] Long Short-Term Memory (LSTM): Specially designed to process time series data of these features, especially the long-term dependencies of heart rate data and shooting technique data, so that the model can not only analyze the current state, but also understand the impact of historical context on the psychological state. LSTM controls the flow of information through input gates, forget gates, and output gates to effectively capture state changes.

[0109] By alternately stacking CNN and LSTM layers, a complex neural network structure is formed. The input layer of the model receives physiological data (heart rate, HRV, etc.) and shooting technical data (hit rate, accuracy, shooting time, etc.), and after multiple layers of convolution, pooling and LSTM layers, it finally outputs the psychological stability and growth evaluation results.

[0110] S302, model training:

[0111] In order to train deep learning models, we need a large amount of labeled data. During model training, historical data sets are used, including shooter physiological data, shooting technique data, and historical evaluation results. The training goal is to minimize the loss function, usually the mean square error loss function.

[0112] Optimization algorithm: During training, Adam Optimizer (adaptive momentum estimation) is used, which combines the advantages of AdaGrad and RMSProp and performs well in processing sparse gradients and large-scale data.

[0113] Batch processing and training: In order to improve training efficiency, data is divided into multiple batches for processing. Each batch of data is forward propagated and backward propagated through the neural network to update the network weights. Batch training can improve the convergence speed of the model.

[0114] Cross-validation: K-fold cross-validation is used to verify the accuracy and generalization ability of the model. After each training, the model is evaluated with the validation set to ensure that it performs well on unseen data.

[0115] S400 uses a trained deep learning model to evaluate the shooter’s psychological state and outputs psychological stability and growth evaluation indicators, including:

[0116] The fully trained deep learning model will evaluate the shooter's psychological stability and growth in real time. The output will be presented in the form of a visual chart, allowing users to understand the current status at a glance.

[0117] S401, Psychological stability assessment: A psychological stability index is generated for each player through the output of the deep learning model, with a value ranging from 0 to 1. The closer to 1, the more stable the psychological state.

[0118] S402, Growth Assessment: Similarly, based on the deep learning model prediction and historical data, a growth score is constructed to reflect the improvement of the player's technical and psychological quality. For example, if the psychological stability has gradually improved over the past three months, a higher growth score will be fed back.

[0119] S403, Visual Analysis: Use radar charts to display all aspects of the shooter's situation, helping users to understand the shooter's status and growth trend more intuitively.

[0120] This application provides a shooter psychological state assessment system based on deep learning, including:

[0121] The acquisition unit is used to acquire the physiological data and shooting technical data of the shooter, wherein the physiological data of the shooter includes heart rate data and heart rate variability data, and the calculation formula of the heart rate variability data is as follows:

[0122] ;

[0123] In the formula, is the interval between consecutive heartbeats, is the average of all consecutive heartbeat intervals, is the total number of observations, is the heart rate variability data, is the number of observations;

[0124] The shooting technical data includes hit rate data, shooting accuracy data and shooting time data. The calculation formula of the hit rate data is as follows:

[0125] ;

[0126] In the formula, is the hit rate, is the number of hits, indicating the number of targets actually hit by the shooter. is the total number of shots;

[0127] The calculation formula of the shooting accuracy data is as follows:

[0128] ;

[0129] In the formula, It is the shooting accuracy, which indicates the distance between the shooter's shooting point and the center of the bull's eye. Indicates the horizontal coordinate of the shooter's actual hit point, The vertical coordinate of the shooter's actual hit point, represents the horizontal coordinate of the bull's eye, Indicates the vertical coordinate of the bull's eye;

[0130] A preprocessing and feature extraction unit, used for preprocessing and feature extraction of acquired physiological data and shooting technical data;

[0131] A training unit, for training the constructed deep learning model using a convolutional neural network and a long short-term memory network based on the extracted features;

[0132] The output unit is used to evaluate the shooter's psychological state through the trained deep learning model and output psychological stability and growth evaluation indicators.

[0133] The shooter psychological state assessment system based on deep learning provided by the present application may also include: a memory and a processor; the memory is used to store instructions;

[0134] The processor is used to operate according to the instructions to execute the steps of the aforementioned shooter psychological state assessment method based on deep learning.

[0135] The present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the aforementioned deep learning-based shooter psychological state assessment method.

[0136] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0137] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0138] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0139] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0140] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A shooter psychological state assessment method based on deep learning, characterized in that: include: Acquire the physiological data and shooting technical data of the shooter, wherein the physiological data of the shooter includes heart rate data and heart rate variability data, and the calculation formula of the heart rate variability data is as follows: ； In the formula, is the interval between consecutive heartbeats, is the average of all consecutive heartbeat intervals, is the total number of observations, is the heart rate variability data, is the number of observations; The shooting technical data includes hit rate data, shooting accuracy data and shooting time data; The calculation formula of the shooting accuracy data is as follows: ； In the formula, It is the shooting accuracy, which indicates the distance between the shooter's shooting point and the center of the bull's eye. Indicates the horizontal coordinate of the shooter's actual hit point, The vertical coordinate of the shooter's actual hit point, represents the horizontal coordinate of the bull's eye, Indicates the vertical coordinate of the bull's eye; Preprocess and extract features from the acquired physiological data and shooting technical data; Based on the extracted features, the constructed deep learning model is trained using convolutional neural networks and long short-term memory networks; The shooter's psychological state is evaluated through the trained deep learning model, and psychological stability and growth evaluation indicators are output.

2. A shooter psychological state assessment method based on deep learning according to claim 1, characterized in that: The calculation formula of the hit rate data is as follows: ； In the formula, is the hit rate, is the number of hits, indicating the number of targets actually hit by the shooter. is the total number of shots.

3. A shooter psychological state assessment method based on deep learning according to claim 1, characterized in that: Preprocess the acquired physiological data and shooting technical data, including: The collected physiological data are cleaned and denoised. The wavelet transform formula for denoising the physiological data is as follows: ； In the formula, Represents the reconstructed signal, and the wavelet basis function is , is the original signal, is a time index, indicating a specific time point during the measurement record. is the total number of signal sample points; Interpolation method was used to complete the missing physiological data and shooting technique data; The collected shooting technical data is standardized using Z-score, and the formula is: ； In the formula, is the standard score, is the original data, is the data mean, is the standard deviation; Time series analysis is performed on physiological data and shooting technical data, and the timing features within each time window are extracted using the sliding window method.

4. The method for assessing a shooter's psychological state based on deep learning according to claim 1, characterized in that: Feature extraction is performed on the acquired physiological data and shooting technical data, including: Extract heart rate fluctuation and heart rate variability features from physiological data; Combining the parameters in the shooting technique data, a feature vector containing information on psychological stability and growth is constructed.

5. The method for assessing a shooter's psychological state based on deep learning according to claim 1, characterized in that: Based on the extracted features, the method of using convolutional neural network and long short-term memory network to train the constructed deep learning model includes: The local features of shooting technical data and physiological data are extracted through convolutional neural networks, and the input data is processed through convolutional layers and pooling layers to generate feature representations; Processing time series data of features through long short-term memory networks to capture long-term dependencies in the data; Using physiological data and shooting technique data extracted and annotated from historical datasets, the mean square error is used as the loss function, and the goal is to minimize the difference between the output and the true value; The Adam optimization algorithm is used to improve learning efficiency, and the learning rate is adjusted in combination with adaptive momentum estimation to adapt to the gradient size of different features; The K-fold cross-validation method was used to evaluate the accuracy and generalization ability of the model.

6. The method for assessing a shooter's psychological state based on deep learning according to claim 1, characterized in that: The trained deep learning model is used to assess the shooter’s psychological state, including: Psychological stability assessment: A psychological stability index is generated for each shooter through the output of the deep learning model, with a value ranging from 0 to 1. The closer to 1, the more stable the psychological state; Growth assessment: Based on deep learning model predictions and historical data, a growth score is constructed to reflect the improvement of the mapper's technical and psychological qualities.

7. A shooter psychological state assessment system based on deep learning, characterized in that: include: The acquisition unit is used to acquire the physiological data and shooting technical data of the shooter, wherein the physiological data of the shooter includes heart rate data and heart rate variability data, and the calculation formula of the heart rate variability data is as follows: ； In the formula, is the interval between consecutive heartbeats, is the average of all consecutive heartbeat intervals, is the total number of observations, is the heart rate variability data, is the number of observations; The shooting technical data includes hit rate data, shooting accuracy data and shooting time data; The calculation formula of the shooting accuracy data is as follows: ； In the formula, It is the shooting accuracy, which indicates the distance between the shooter's shooting point and the center of the bull's eye. Indicates the horizontal coordinate of the shooter's actual hit point, The vertical coordinate of the shooter's actual hit point, represents the horizontal coordinate of the bull's eye, Indicates the vertical coordinate of the bull's eye; A preprocessing and feature extraction unit, used for preprocessing and feature extraction of acquired physiological data and shooting technical data; A training unit, for training the constructed deep learning model using a convolutional neural network and a long short-term memory network based on the extracted features; The output unit is used to evaluate the shooter's psychological state through the trained deep learning model and output psychological stability and growth evaluation indicators.

8. A shooter psychological state assessment system based on deep learning, characterized in that: including memory and processor; The memory is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images