Shooting range-oriented data acquisition and analysis method and system and storage medium

Abstract

The invention discloses a target range-oriented data acquisition and analysis method and system and a storage medium. The method comprises the following steps: acquiring action posture information, physiological state information, environment state information and shooting hit information of a trainer; training effect analysis is conducted on the obtained action posture information, physiological state information, environment state information and shooting hit information, and evaluation indexes are generated; and through a large language model technology, a personalized automatic evaluation report is generated based on the evaluation indexes. According to the method, the trainee can obtain accurate feedback in real time in the training process, meanwhile, training effect analysis greatly improves the shooting training efficiency and individuation level, various information data in the shooting training process can be comprehensively collected and analyzed, the trainee is helped to know the training state, progress and defects of the trainee in real time, and the training effect is improved. And a personalized automatic evaluation report is provided according to an analysis result, so that a more efficient shooting training effect is finally realized.

Description

Technical Field

[0001] This application relates to the field of intelligent range management technology, and in particular to a data acquisition and analysis method, system and storage medium for ranges. Background Technology

[0002] With the continuous development of shooting training and technological advancements, traditional shooting range training methods are gradually failing to meet the demands of modern shooting training for accuracy, efficiency, and personalization. Traditional shooting ranges typically rely on manual recording of scores and judgment of shooting actions, making it impossible to monitor trainees' status in real time and comprehensively, and lacking in-depth analysis and feedback on shooting behavior. This not only limits the scientific nature of training but also makes it difficult to provide customized training programs based on individual differences among trainees. Therefore, how to comprehensively evaluate and analyze the training process through intelligent means has become a crucial issue that urgently needs to be addressed in the field of shooting training. Summary of the Invention

[0003] The purpose of this application is to provide a data acquisition and analysis method, system, and storage medium for target ranges, in order to solve one or more technical problems existing in the prior art, or at least provide a beneficial option or create conditions.

[0004] To achieve the above-mentioned objectives, this application employs the following technical solution: This application provides a data acquisition and analysis method for target ranges, including: Acquire information on the trainee's posture, physiological state, environmental conditions, and shooting accuracy; The acquired motion posture information, physiological state information, environmental state information, and shooting hit information are analyzed to assess the training effect and generate evaluation indicators, including the degree of skill improvement, teamwork ability, rationality of training plan, and degree of environmental influence. By using large language model technology, personalized automated evaluation reports are generated based on evaluation indicators.

[0005] Furthermore, the method for generating the assessment indicators of the skill improvement level includes: A multi-index model was constructed based on hit rate, ring count, heart rate variability, impact point dispersion, and posture accuracy, and weights were assigned to calculate the comprehensive skill index. ; in, It is a comprehensive skills index. It is the standardized hit rate. The degree of dispersion of the point of impact, It's about posture accuracy. It is the set weight.

[0006] Furthermore, the method for generating the evaluation indicators of the team collaboration ability includes: Calculate the team's efficiency when performing tasks: ; in, It's about task execution efficiency. The total number of tasks completed. That is the total time spent.

[0007] Furthermore, the method for generating evaluation metrics for the rationality of the training plan includes: Calculate the target achievement rate by comparing the number of completed targets with the total number of targets: ; in, For the target achievement rate, This is the number of targets that have been achieved. It is the total target quantity; Analyze the fit between training load and actual performance, determine the relationship between the two through regression analysis, and evaluate the rationality of the training load; The training plan's rationality score is calculated by considering the overall goal achievement rate, adaptability indicators, and performance volatility. ; in, It is a score for the rationality of the training plan. The standard deviation of the score These are the weighting coefficients. It is a load adaptability indicator.

[0008] Furthermore, the method for generating assessment indicators of the environmental impact level includes: The light intensity influence factor is calculated using the following formula: ; in, The light intensity influencing factor, The actual light intensity is used. When the light intensity is between 300-500 lx, the influence factor is 0. When the light intensity exceeds this range, the deviation value is calculated and standardized to a value of [0, 1]. Wind speed influence factor, formula as follows: ; in, W represents the wind speed influence factor. When the wind speed is less than or equal to 3 m / s, the influence factor is 0. When the wind speed exceeds 3 m / s, the influence factor increases proportionally, with a maximum of 1. Formula for comprehensive calculation of environmental impact: The influence factors of sunlight and wind speed are calculated together according to their weights: ; in, To represent the degree of environmental impact, α and β are weighting coefficients used to balance the effects of sunlight and wind speed, with α = 0.4 and β = 0.6.

[0009] Furthermore, methods for generating personalized, automated evaluation reports based on evaluation metrics through large language model technology include: Based on evaluation indicators such as skill improvement, teamwork ability, training plan rationality, and environmental impact, the overall training effect is analyzed, and personalized automated evaluation reports are generated based on the analysis results. In the process of generating automated evaluation reports, the natural language generation capabilities of large language model technology are utilized to achieve automated interpretation and writing of analysis results through Prompt engineering methods, retrieval enhancement generation methods, and template filling methods.

[0010] This application provides a data acquisition and analysis system for target ranges, including: The acquisition unit is used to acquire the trainee's movement posture information, physiological state information, environmental state information, and shooting hit information; The training effect analysis unit is used to analyze the acquired action posture information, physiological state information, environmental state information, and shooting hit information to generate evaluation indicators. The generation unit is used to generate personalized, automated evaluation reports based on evaluation metrics using large language model technology.

[0011] This application provides a data acquisition and analysis system for a target range, including a memory and a processor; The memory is used to store instructions; The processor is configured to operate according to the instructions to perform the steps according to the method described above.

[0012] This application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0013] The beneficial effects of this application are as follows: This application enables trainees to receive precise feedback in real time during training, accurately understanding their performance in shooting movements, physiological state, and environmental adaptability. Simultaneously, training effectiveness analysis will significantly improve the efficiency and personalization of shooting training, comprehensively collecting and analyzing various information and data during the training process. This helps trainees understand their training status, progress, and shortcomings in real time, and provides personalized, automated evaluation reports based on the analysis results, ultimately achieving more efficient shooting training results. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating a target range-oriented data acquisition and analysis method according to an embodiment of this application. Figure 2 This is a schematic block diagram of a system in a target range-oriented data acquisition and analysis method according to an embodiment of this application; Figure 3 This is a schematic block diagram of the shooting hit information collection and score statistics module in a range-oriented data collection and analysis method provided in an embodiment of this application; Figure 4 This is a visualization interface diagram of a target range-oriented data acquisition and analysis method provided according to an embodiment of this application. Detailed Implementation

[0015] Reference Figures 1-4 The purpose of this application is to provide a data acquisition and analysis method for target ranges, including: S100 acquires the trainee's movement posture information, physiological state information, environmental state information, and shooting hit information; S200 analyzes the training effect based on the acquired action posture information, physiological state information, environmental state information, and shooting hit information, and generates evaluation indicators, including the degree of skill improvement, teamwork ability, rationality of training plan, and degree of environmental influence. S300 uses large language model technology to generate personalized, automated evaluation reports based on evaluation metrics.

[0016] In S100, information is collected through a data acquisition module, specifically including: Motion and Posture Acquisition Module: This module captures the trainee's body posture and shooting movements in real time using a camera. Image processing technology is employed to analyze the shooting movements, ensuring the standardization of the shooting posture and helping to assess the trainee's motion accuracy and stability. This module can accurately identify motion deviations through high-precision visual data, thereby providing targeted improvement suggestions. The core function of this module is to capture the trainee's body posture and shooting movements in real time using a camera, and then employ image processing technology combined with a deep learning model to accurately analyze the posture and movements. This aims to improve the efficiency and effectiveness of shooting training. The specific implementation method is as follows: Data acquisition unit: Multiple high-definition cameras are installed near the designers to collect data on the trainees' shooting movements and postures via real-time video streaming. The cameras capture images from different angles to ensure coverage of the trainees from multiple directions, capturing details of their movements.

[0017] Data processing unit: Uses image processing algorithms to process the acquired video data, extracts the key points of the trainee, and analyzes them through a deep learning model to determine whether the trainee's movements and postures conform to the standard specifications.

[0018] To capture trainees' movements and poses in real time and efficiently, the core algorithm uses YOLO (You Only Look Once) object detection, which is an efficient end-to-end convolutional neural network (CNN) that can perform object classification and localization tasks in images.

[0019] The YOLO model boasts a fast detection speed, completing target classification and bounding box prediction in a single forward propagation, significantly reducing computational latency. For scenarios with extremely high real-time requirements, such as shooting training, YOLO's performance is particularly well-suited.

[0020] Furthermore, the YOLO model is lightweight and can be deployed on edge devices with limited computing resources. At the same time, its structure is easily optimized through transfer learning and hyperparameter tuning, allowing it to quickly adapt to the specific needs of shooting training scenarios.

[0021] The YOLO network uses manually labeled pose datasets (including trainee images from various angles and poses) as training data. The annotation process requires precise work by professionals to ensure the accuracy of each bounding box. Each image is labeled with the position of each joint, the pose angle, and whether it conforms to a standard pose.

[0022] To improve the model's robustness and generalization ability, we used data augmentation techniques. Specifically, data augmentation can simulate various training environments, including but not limited to: 1. Adjust image brightness Contrast and saturation This generates training samples under diverse lighting conditions.

[0023] 2. By using image overlay technology, the trainee's pose is embedded into different backgrounds to form composite samples. The background replacement formula can be expressed as:

[0024] in, It is a binary mask. and These are the background image and the foreground image, respectively. For the new image.

[0025] 3. Geometric Transformations: Perform rotation, scaling, translation, and random cropping operations on images.

[0026] 4. Noise Addition: Gaussian noise is added to enhance the noise immunity of the model. The noise model is defined as follows:

[0027] in, With a mean of 0 and a variance of Gaussian noise, For the original image, The image after noise has been added.

[0028] Dynamic fuzziness: By incorporating the robustness of dynamic fuzziness addition and subtraction models, dynamic fuzziness is defined as:

[0029] in, This represents the blurred image. Indicates the initial clear image. This represents a dynamic fuzzy kernel. Indicates the degree of dynamic blur. Represents spatial coordinates, Indicates time.

[0030] YOLO model optimization: To adapt to different training environments (such as different backgrounds and lighting conditions), the model needs to be trained and optimized multiple times to improve its detection accuracy in diverse environments. During training, cross-validation is used to adjust hyperparameters to ensure that the model finds the optimal balance between accuracy and speed.

[0031] As data accumulates, the YOLO model will be continuously optimized, improving its accuracy and response speed. Regular manual annotation and the addition of new data will further enhance its adaptability and accuracy, ensuring stable operation under different training environments and conditions.

[0032] Feedback Unit: Based on the analysis results, it provides real-time training feedback to help trainees adjust their movements and postures, ensuring the accuracy and stability of their movements.

[0033] In this way, the motion and posture acquisition module can capture the trainee's shooting movements in real time and accurately, and provide targeted feedback to help the trainee improve the accuracy and stability of the shooting movements.

[0034] Physiological state acquisition module: By wearing a wristband, the system monitors the trainee's physiological parameters in real time, such as heart rate and respiratory rate, comprehensively reflecting their physiological state. Using data collected by the wristband's sensors, it can determine whether the trainee is in optimal shooting condition, thus providing real-time feedback on physiological adaptability and helping them maintain peak physiological performance for shooting. The smart bracelet worn by the shooter collects heart rate data in real time and transmits it back to the backend via Wiify. Heart rate data is a key indicator for assessing psychological stability. The bracelet uses optical sensors (such as PPG sensors) to monitor blood flow fluctuations and calculate heart rate. To ensure data accuracy, heart rate data is acquired through high-frequency sampling and timestamps are recorded. Additionally, the shooter's heart rate variability (HRV) is monitored. HRV reflects the regulatory capacity of the autonomic nervous system and is an effective physiological parameter for assessing psychological stability. The formula for calculating HRV is as follows:

[0035] in, The interval between consecutive heartbeats. The average of these intervals, The total number of observations. For heart rate variability data, The number of observations.

[0036] The collected data will be saved to the database and displayed to the shooter in real time for reference and adjustment.

[0037] Environmental Condition Acquisition Module: This module monitors changes in the shooting environment in real time using a meteorological instrument, including key parameters such as temperature, humidity, light intensity, and wind speed. It analyzes the potential impact of these environmental factors on shooting performance, providing accurate environmental data support for a scientific evaluation of trainees' performance under different environmental conditions. The main function of this module is to monitor changes in the shooting environment in real time using a meteorological instrument, including key environmental parameters such as temperature, humidity, light intensity, and wind speed, and analyze the impact of these factors on shooting performance. By providing accurate environmental data support, it helps trainees conduct scientific assessments under different environmental conditions, thereby optimizing the training process and shooting performance. The specific implementation method of this module is as follows: 1. Deploy various meteorological monitoring instruments (temperature and humidity sensors, light sensors, anemometers, etc.) in the training ground to monitor and collect environmental data in real time. These sensors should have high accuracy and real-time response capabilities, and be able to provide dynamic data at different stages of the shooting process.

[0038] 2. The collected environmental data is transmitted to a central data processing server or cloud platform via communication technologies (such as Wi-Fi, Bluetooth, wired, etc.).

[0039] 3. A dedicated data processing system is used to analyze and process environmental data, identifying key parameters that affect shooting performance, such as wind direction, wind speed, sunlight, temperature, and whether it is raining.

[0040] 4. Provide real-time feedback to trainees on the shooter terminal to help them adjust shooting strategies and training methods, and improve shooting accuracy in specific environments.

[0041] Shooting hit information collection and performance statistics module: accurately records the hit information of each shot, including the hit location and accuracy, and calculates the score after training to provide data support for subsequent performance analysis and training evaluation; The automatic shooting performance analysis module primarily utilizes image processing technology and coordinate mapping algorithms to comprehensively evaluate and analyze shooting performance. This module intelligently integrates target area settings with scoring rules. The data extracted by this module can be used by shooting training participants to understand their performance, and can also provide data support and guidance for subsequent training effectiveness evaluation and trend prediction. The specific process is as follows: 1. Area Setup: A visual canvas is provided using SVG technology, allowing users to directly draw and edit the target area on the webpage. Users can freely draw and adjust the target shape using a mouse or touch device.

[0042] 2. Scoring Rule Settings: Before training, users can specify corresponding scoring rules in the course settings for different types of trainees to evaluate their training rating. Supported scoring indicators include: number of hits, number of hits on the target, number of hits on the ring, hit time, and number of hits on the area.

[0043] 3. Score Extraction: When a shooting event occurs, a score mask is generated based on the target curve and region configuration. This mask is a binary image based on the target region, where the pixel value of each region represents the score for that region.

[0044] Based on the xy coordinate data transmitted by the shooting equipment, the system matches the shooting point with the target area. After the coordinates of the shooting point are mapped to the corresponding area, the score for that area is found, and the score for that shot is calculated in real time.

[0045] 4. Performance Statistics: After training, this module will perform statistical analysis on the performance data, such as calculating the average score, excellent rate, pass rate, hit rate, reaction time, etc., which intuitively reflects the shooter's shooting level.

[0046] In S200, the data analysis module performs in-depth analysis of the collected data, which requires the following collected data: Environmental data: Wind speed: The higher the wind speed, the more significant the impact on the deviation of the bullet trajectory.

[0047] Light intensity: Both excessively strong and insufficient light can affect aiming accuracy.

[0048] Motion posture data: Posture accuracy: The degree to which the actual movement posture matches the standard posture.

[0049] Shooting score data: Ring score: The number of rings scored in each shot, reflecting the accuracy of the shooter.

[0050] Point of impact: The point where each bullet lands, used to analyze actual performance.

[0051] Firing time: The time required for each shot, used for efficiency evaluation.

[0052] Completion time: The actual time spent on each training session.

[0053] Mission objective: The shooting target set, such as hit rate target, ring score, etc.

[0054] Evaluation metrics for training effectiveness include: The marksman's skill improvement level, teamwork ability, the rationality of the training plan, and the degree of environmental influence. The formula for calculating the skill upgrade level of a marksman: A multi-index model is constructed, comprehensively considering hit rate, ring count, HRV (heart rate variability), impact point dispersion, and posture accuracy, and weights are set according to actual conditions to calculate the comprehensive skill index: ; in, It is a comprehensive skills index. It is the standardized hit rate. The degree of dispersion of the point of impact, It's about posture accuracy. It is the set weight.

[0055] Formula for calculating team collaboration ability: Task completion efficiency assessment: Calculate the team's efficiency when performing tasks: ; in, It's about task execution efficiency. The total number of tasks completed. That is the total time spent.

[0056] Formula for calculating the rationality assessment of a training plan: Target achievement rate analysis: Calculate the target achievement rate by comparing the number of completed targets with the total number of targets: ; in, For the target achievement rate, This is the number of targets that have been achieved. It represents the total target quantity.

[0057] Load adaptability analysis: Analyze the fit between training load and actual performance, determine the relationship between the two through regression analysis, and evaluate the rationality of the training load.

[0058] Reasonableness score: The training plan's rationality score is calculated by considering the overall goal achievement rate, adaptability indicators, and performance volatility. ; in, It is a score for the rationality of the training plan. The standard deviation of the score These are the weighting coefficients. It is a load adaptability indicator.

[0059] Formula for calculating the degree of environmental impact: 1. Ideal value range setting: When the light intensity is between 300-500 lx, it has almost no effect on shooting. The greater the wind speed, the greater the impact on bullet trajectory deviation, especially when it exceeds 3 m / s, the effect increases significantly.

[0060]

[0061] 2. Calculation methods for the influence factors of sunlight and wind speed (1) Calculation of light intensity influence factor: ; in, The light intensity influencing factor, The actual light intensity is used. When the light intensity is between 300-500 lx, the influence factor is 0. When the light intensity exceeds this range, the deviation value is calculated and standardized to a value of [0, 1]. (2) Calculation of wind speed influence factor: ; in, W represents the wind speed influence factor. When the wind speed is less than or equal to 3 m / s, the influence factor is 0. When the wind speed exceeds 3 m / s, the influence factor increases proportionally, with a maximum of 1. 3. Comprehensive calculation formula for the degree of environmental impact: The influence factors of sunlight and wind speed are calculated together according to their weights: ; in, To represent the degree of environmental impact, α and β are weighting coefficients (α represents the weight of light intensity on shooting performance. A larger α value means that light intensity has a more significant impact on shooting performance. β represents the weight of wind speed on shooting performance. A larger β value means that wind speed has a more significant impact on shooting performance). These coefficients are used to balance the effects of light and wind speed, and are set to α = 0.4 and β = 0.6.

[0062] Specifically, the S300 includes: By summarizing data from four core dimensions—shooter skill improvement, teamwork ability, training plan rationality, and environmental influence—a comprehensive analysis of training effectiveness is conducted, and a personalized, automated evaluation report is generated based on the analysis results. The report generation process utilizes Large Language Modeling (LLM) technology, employing methods such as Prompt engineering, Retrieval Augmentation (RAG), and template filling to automate report writing, enhancing the intelligence and personalization of report generation and providing precise improvement suggestions for training optimization.

[0063] Data integration: Summarize the assessment results of skill improvement, teamwork ability, training plan rationality and environmental impact, and comprehensively analyze the overall training effect.

[0064] Evaluation Report Generation: During the evaluation report generation process, the natural language generation capabilities of LLM are utilized to automate the interpretation of the analysis results. This process mainly includes the following technical steps: (1) Use Prompt engineering techniques to guide LLM in generating reports. To generate highly personalized reports, a series of Prompt templates adapted to different analysis scenarios were designed using Prompt engineering techniques. These Prompt templates provide guidance based on different data dimensions, enabling LLM to output accurate and professional evaluation content.

[0065] Skill Enhancement Prompt Example Generate a skills improvement assessment report based on the following training data, including trend analysis of indicators such as hit rate, average score, shooting time, and bullet impact dispersion, as well as improvement suggestions.

[0066] Team Collaboration Section Prompt Example Generate a team collaboration assessment report based on the team members' efficiency in performing tasks, including task completion status and suggestions for improvement.

[0067] The Prompt content will be dynamically adjusted to ensure that the LLM can generate highly relevant report content based on different input data, avoiding templated and repetitive feedback. (2) Improve report accuracy using Search Enhancement Generation (RAG) technology.

[0068] The Retrieval-Augmented Generation (RAG) technique is used to retrieve historical records and external knowledge base content related to the current training data in real time when generating evaluation reports, ensuring the accuracy and contextual relevance of the report content.

[0069] Knowledge base construction Historical training data, historical evaluation reports, and other information are stored in a vector database for LLM to retrieve when generating reports.

[0070] RAG generation process Step 1: Data Retrieval → Retrieve historical data related to the current training from the knowledge base, including reports and improvement suggestions for similar training scenarios.

[0071] Step 2: Report Generation → LLM. When generating the report, the retrieved historical data and current training data are used as input to ensure that the output is accurate and personalized. (3) Use template filling technology to generate structured reports

[0072] To improve the consistency of report generation and the standardization of output format, template-based filling technology is adopted. Dynamically populated fields are reserved in the designed report template, which are automatically filled with the generated analysis results and improvement suggestions by LLM.

[0073] Template-filling technology ensures that the generated reports have a clear structure and fluent language, making it easy for users to intuitively understand the training results. (4) Example of assessment report content

[0074] Skills Improvement Trend Analysis Over the past five training sessions, your overall skill index has shown a continuous upward trend, increasing from 0.78 to 0.92. You have performed particularly well in accuracy and bullet impact dispersion. However, your shooting time has fluctuated slightly; it is recommended that you strengthen your shooting rhythm training while improving accuracy.

[0075] Teamwork rating The team collaboration score was 85, an 8% improvement over previous training sessions. While the team's collaboration ability is high, there is still room for improvement. By refining task allocation and workflow, the team can significantly improve collaboration efficiency, thereby enhancing the overall quality and speed of task completion.

[0076] Training plan rationality assessment Your training plan received a rationality score of 78, indicating a high goal achievement rate, but a low load adaptability score. It is recommended that you adjust the load appropriately in future training to avoid overtraining that could lead to inconsistent performance.

[0077] Environmental Impact Assessment Recommendations Your environmental impact score of 70% indicates that the environmental conditions during training were somewhat challenging, primarily due to unstable lighting and varying wind speeds. However, you still achieved relatively stable results under these adverse conditions, demonstrating strong adaptability. It is recommended that future training focus on improving adaptability to different environmental conditions to further enhance performance stability in complex environments.

[0078] Finally, the collected data is analyzed in depth using the visualization module. The collected data required includes: The core task of the visualization module is to transform complex data analysis results into intuitive charts and graphs, enabling trainees to clearly understand and analyze their training status, effects, and progress. Through highly interactive and easy-to-understand charts, trainees can quickly obtain the information they need and make targeted adjustments. This module primarily uses ECharts (an open-source front-end charting library) for data visualization. ECharts provides various chart types and supports rich interactive features, effectively enhancing the user experience and helping trainees make informed decisions at different training stages.

[0079] This module acquires processed data from other modules (such as the data acquisition module and training effect evaluation module). This data typically includes the trainee's performance data, movement stability, physiological state, environmental parameters, etc.

[0080] Based on ECharts' chart generation engine, select the appropriate chart type (such as line chart, bar chart, pie chart, radar chart) to display different types of data.

[0081] With the interactive features provided by ECharts, trainees can interact with charts as needed (e.g., zoom, drag, switch data dimensions, etc.) to gain a deeper understanding of the trends and patterns behind the data.

[0082] In the visualization module, to enhance user experience and provide more precise analysis, a search box allows users to set a query time range. Users can select a specific time period in the search box, and only relevant data within that time period will be displayed. This feature not only allows trainees to focus on the training results of a particular stage but also helps users better identify data trends and changes.

[0083] This application provides a data acquisition and analysis system for target ranges, including: The acquisition unit is used to acquire the trainee's movement posture information, physiological state information, environmental state information, and shooting hit information; The training effect analysis unit is used to analyze the acquired action posture information, physiological state information, environmental state information, and shooting hit information to generate evaluation indicators. The generation unit is used to generate personalized, automated evaluation reports based on evaluation metrics using large language model technology.

[0084] The data acquisition and analysis system for a target range provided in this application may further include a memory and a processor; the memory is used to store instructions. The processor is used to operate according to the instructions to execute the steps of the aforementioned target range-oriented data acquisition and analysis method.

[0085] This 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 target range-oriented data acquisition and analysis method.

[0086] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0087] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0088] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0089] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0090] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A target range oriented data acquisition analysis method, characterized by, The method comprises the following steps: acquiring action posture information, physiological state information, environmental state information, and shooting hit information of a trainee; performing training effect analysis on the acquired action posture information, physiological state information, environmental state information, and shooting hit information, and generating evaluation indexes, wherein the evaluation indexes include skill improvement degree, team cooperation ability, training plan rationality, and environmental impact degree; generating an individualized automatic evaluation report based on the evaluation indexes through a large language model technology.

2. The data acquisition and analysis method for a target range according to claim 1, wherein, The method for generating the evaluation index of the skill improvement degree comprises the following steps: constructing a multi-index model based on hit rate, ring number, heart rate variability, dispersion degree of impact point, and posture accuracy, setting weights, and calculating a comprehensive skill index: ； wherein, is a comprehensive skill index, is a normalized hit rate, is a dispersion degree of a point of impact, is a posture accuracy rate, is a set weight.

3. The data acquisition and analysis method for a target range according to claim 1, wherein, The method for generating the evaluation index of the team cooperation ability comprises the following steps: calculating the efficiency of the team when performing a task: ； wherein, is the efficiency of performing tasks, is the total number of tasks completed, is the total time spent.

4. The data acquisition and analysis method for a target range according to claim 1, wherein, The method for generating the evaluation index of the training plan rationality comprises the following steps: counting the number of completed goals and the total number of goals, and calculating a goal achievement rate: ； wherein, the target achievement rate, is the number of completed targets, is the total number of targets; analyzing the adaptation degree of training load and actual performance, determining the relationship between the two through regression analysis, and evaluating the rationality of the training load; comprehensively calculating the training plan rationality score based on the goal achievement rate, adaptability index, and performance volatility: ； wherein, is the training plan rationality score, denotes the standard deviation of the performance, is the weight coefficient, is the load adaptability indicator.

5. The method of data collection and analysis for a target range of claim 1, wherein, The method for generating the evaluation index of the environmental impact degree comprises the following steps: illumination intensity influence factor, formula as follows: ； wherein, is a light intensity influence factor, is the actual light intensity, the influence factor is 0 when the light intensity is in the range of 300-500 lx, and when the light intensity is outside this range, the deviation is calculated and normalized to a value of [0, 1]; wind speed influence factor, formula as follows: ； wherein, W is the actual wind speed, when the wind speed is less than or equal to 3 m / s, the influence factor is 0, when the wind speed exceeds 3 m / s, the influence factor increases in proportion, and the maximum is 1; comprehensive calculation formula of environmental impact degree: comprehensively calculating the illumination and wind speed influence factors according to weights: ； wherein, is the degree of environmental impact, and a and β are weight coefficients for balancing the effects of light and wind speed, with a = 0.4 and β = 0.

6.

6. The method of data collection and analysis for a target range of claim 1, wherein, The method for generating an individualized automatic evaluation report based on the evaluation indexes through a large language model technology comprises the following steps: based on the evaluation indexes of skill improvement degree, team cooperation ability, training plan rationality, and environmental impact degree, analyzing the overall training effect, and generating an individualized automatic evaluation report based on the analysis result; in the automatic evaluation report generation process, using the natural language generation capability of the large language model technology, through the Prompt engineering method, the retrieval enhancement generation method, and the template filling method, realizing the automatic interpretation and writing of the analysis result.

7. A target range oriented data acquisition and analysis system, characterized by, The method comprises the following steps: an acquisition unit configured to acquire action posture information, physiological state information, environmental state information, and shooting hit information of a trainee; a training effect analysis unit configured to perform training effect analysis on the acquired action posture information, physiological state information, environmental state information, and shooting hit information, and generate evaluation indexes; a generation unit configured to generate an individualized automatic evaluation report based on the evaluation indexes through a large language model technology.

8. A target range oriented data acquisition and analysis system, characterized by, The method comprises the following steps: an acquisition unit configured to acquire action posture information, physiological state information, environmental state information, and shooting hit information of a trainee; a training effect analysis unit configured to perform training effect analysis on the acquired action posture information, physiological state information, environmental state information, and shooting hit information, and generate evaluation indexes; 9. A computer readable storage medium having stored thereon a computer program, characterized in that, a generation unit configured to generate an individualized automatic evaluation report based on the evaluation indexes through a large language model technology. The method comprises the following steps: an acquisition unit configured to acquire action posture information, physiological state information, environmental state information, and shooting hit information of a trainee; a training effect analysis unit configured to perform training effect analysis on the acquired action posture information, physiological state information, environmental state information, and shooting hit information, and generate evaluation indexes; a generation unit configured to generate an individualized automatic evaluation report based on the evaluation indexes through a large language model technology. The method comprises the following steps: an acquisition unit configured to acquire action posture information, physiological state information, environmental state information, and shooting hit information of a trainee; a training effect analysis unit configured to perform training effect analysis on the acquired action posture information, physiological state information, environmental state information, and shooting hit information, and generate evaluation indexes; a generation unit configured to generate an individualized automatic evaluation report based on the evaluation indexes through a large language model technology.

Images