A smart shooting range data management method and system

Abstract

The present invention relates to the field of data management technology, and in particular to a smart shooting range data management method and system, including: collecting first video data from a first data source at a preset frequency; collecting second video data from a second data source at the same time, and transmitting it to the cloud for storage. Download the second video data from the cloud, and together with the previously stored first video data, form a sample data set and save it locally, and divide the first video data into a first video data group and a second video data group. Then, construct feature data set 1, feature data set 2, and feature data set 3, establish an association between feature data set 1 and feature data set 2 in the time dimension, and perform a similar association operation between feature data set 1 and feature data set 3. Finally, construct a data chain, through which shooting behavior patterns are deeply analyzed, performance trends of weapons and equipment are predicted, and more scientific and reasonable shooting strategy recommendations are proposed accordingly.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a smart shooting range data management method and system. Background Art

[0002] A smart range refers to a range that uses modern information technology to achieve efficient management and use of range resources. In a smart range, there are many types of data involved. How to effectively manage and use this data is the key to improving the operating efficiency of a smart range.

[0003] Smart shooting range data mainly includes: shooting performance data, shooting weapon status data, and shooting video surveillance data. How to use the Internet of Things technology to achieve real-time collection of various data in the smart shooting range, and use big data analysis and artificial intelligence technology to conduct in-depth analysis of the integrated data and tap the data value are essential for the operation and management of the smart shooting range.

[0004] However, in reality, the data in the shooting range are scattered, and it is impossible to establish effective connections between the various data. When analysis is required, it is very difficult to collect and call data. Summary of the invention

[0005] The present invention quickly collects data from different data sources and stores them uniformly, pre-associates related different data, and improves the efficiency of data management.

[0006] The technical solution proposed by the present invention is: a smart shooting range data management method, the method comprising:

[0007] Collecting video data 1 from a first data source at a preset frequency, and storing the video data 1 locally;

[0008] Collecting video data 2 from a second data source, and transmitting the video data 2 to cloud storage;

[0009] Download video data 2 and video data 1 from the cloud to form a sample data set and store it locally;

[0010] The video data 1 is divided into a first video data group and a second video data group, and feature data are extracted from the first video data group, the second video data group and the video data 2 respectively to form a feature data set 1, a feature data set 2 and a feature data set 3;

[0011] Establish an association between feature data set 1 and feature data set 2 in the time dimension;

[0012] Establish an association between feature data set 1 and feature data set 3 in the time dimension;

[0013] Based on the existing historical data analysis records and data access records, build multiple data links that can support various specific analysis needs.

[0014] Preferably, the video data one includes weapon shooting video data one and shooter video data, and the video data two includes weapon shooting video data two;

[0015] The step of dividing the video data into a first video data group and a second video data group comprises:

[0016] The weapon shooting video data 1 is divided into the first video data group, and the acquisition timestamp 1 is added; the shooter video data is divided into the second video data group, and the acquisition timestamp 2 is added;

[0017] The construction of multiple data links that can support various specific analysis needs is specifically: establishing data links of feature data set 1, feature data set 2 and feature data set 3, and storing them locally for analyzing shooting behavior, predicting weapon performance degradation and optimizing shooting strategies.

[0018] Preferably, extracting feature data from the first video data set, the second video data set and the second video data set respectively to form feature data set one, feature data set two and feature data set three comprises the following steps:

[0019] Extract feature data from the first video data set to form feature data set one ,in, Indicates weapon one The motion vector of each structure; Indicates the total number of weapon structures;

[0020] Extract feature data from the second video data set to form feature data set 2 ,in, Indicates shooter The motion vectors of the joints;

[0021] Extract feature data from video data 2 to form feature data set 3 ,in, Indicates weapon 2 The motion vector of the structure.

[0022] Preferably, establishing association between the feature data set 1 and the feature data set 2 in the time dimension comprises the following steps:

[0023] Use sliding window one to extract multiple feature data from feature data set one to form time series one;

[0024] Use sliding window 2 to extract multiple feature data from feature data set 2 to form time series 2;

[0025] Use the DTW algorithm to match time series 1 and time series 2 in time and output the matching results;

[0026] According to the matching result, the time series one and the time series two are combined into a shooting sample data set, and the shooting sample data set is used to establish the relationship between the movement of each structure of the weapon one and the corresponding arm movement of the shooter.

[0027] Preferably, establishing association between the feature data set 1 and the feature data set 3 in the time dimension comprises the following steps:

[0028] Use sliding window three to extract multiple feature data from feature data set three to form time series three;

[0029] Use the DTW algorithm to match time series 1 and time series 3 in time, and output the matching results;

[0030] According to the matching results, the weapon sample dataset is constructed using time series 1 and time series 3;

[0031] The weapon sample data set is used to analyze the degree of weapon wear and the probability of wear occurring.

[0032] Preferably, analyzing the second feature data set to obtain the arm movement trajectory of the shooter includes the following steps:

[0033] Extracting a plurality of continuous frames of images from the collected video data of the second video data group; preprocessing the extracted plurality of continuous frames of images, wherein the preprocessing includes graying and denoising;

[0034] Use the pre-trained YOLOv5 to identify the position of the shooter’s arm in each frame and add multiple identification points to the arm image;

[0035] Use Kalman filter to associate the identified points of the arm in consecutive frames to form a trajectory;

[0036] Smooth the trajectory data to obtain the trajectory data of the arm and add the shooting timestamp;

[0037] After obtaining the running trajectory of the arm, the shooting performance data corresponding to this shooting is read from the target system in real time, and the shooting performance timestamp is added;

[0038] Align the shooting timestamp and shooting result timestamp on the timeline;

[0039] The arm motion trajectory data and the shooting performance are combined as sample data to train a shooting performance prediction model, and the shooting performance is predicted based on the shooter's arm motion trajectory;

[0040] Obtain a sample data set again, and extract a time series 1 matching the arm motion trajectory from the sample data set;

[0041] Align time series 1 with the shooting performance data on the time axis to form a shooting action improvement data set;

[0042] The arm motion trajectory data and the shooting performance are combined as sample data to train a shooting performance prediction model, and the shooting performance is predicted according to the shooter's arm motion trajectory, including the following steps:

[0043] Use Keras to build an LSTM model as a shooting performance prediction model;

[0044] Extracting key arm feature data at multiple time points from the arm motion trajectory data to form an arm motion feature time series, wherein the key arm feature data includes movement speed, acceleration, and movement angle change rate;

[0045] After normalizing the key characteristic data of the arm, the data is input into the shooting performance prediction model;

[0046] The shooting performance prediction model outputs the predicted shooting performance.

[0047] Preferably, identifying potential failures or wear of weapons by analyzing data in a weapon sample data set comprises the following steps:

[0048] Get weapon sample dataset ;

[0049] in, Represents a sliding window The data sequence obtained by moving ;

[0050] Indicates the sliding window The data sequence obtained by moving , and Respectively represent the step size of sliding window one and sliding window three; ;

[0051] in, , express Moment Weapon 1 The speed of the structure, express Moment Weapon 1 The direction angle of the structure's motion velocity, express Moment Weapon 1 The movement distance of each structure;

[0052] in, , express Time Weapon II The speed of the structure, express Time Weapon Part 2 The direction angle of the structure's motion velocity, express Time Weapon Part 2 The movement distance of each structure;

[0053] from A key feature parameter set 1 is formed after normalization; the key feature parameter set 1 includes a weapon 1 and a The maximum movement speed of the structure , maximum direction angle , minimum movement speed and minimum direction angle ;

[0054] from Extract key feature parameter 2 from the weapon, and after normalization, form key feature parameter set 2; the key feature parameter 2 includes weapon 2 Maximum movement speed of the structure , maximum direction angle , minimum movement speed and minimum direction angle ;

[0055] Import the pre-trained weapon evaluation model:

[0056] , in, They represent weight one, weight two, weight three and weight four respectively; represents the intercept, Regression coefficients; represents the error term;

[0057] Inputting key feature parameter set 1 and key feature parameter set 2 into the weapon assessment model includes the following steps:

[0058] Key Characteristics Parameter Set 1 and Key Characteristics Parameter Set 2 are input into the weapon assessment model: , output the wear prediction value;

[0059] Setting the evaluation threshold ,if , then the weapon is judged to be slightly worn;

[0060] if , then the weapon is judged to be moderately worn;

[0061] if , then the weapon is judged to be severely worn;

[0062] Weapon Assessment Model Taking the key characteristic parameter set 1 as input variables, the probability value of weapon wear is output;

[0063] Setting probability thresholds ,if , then the weapon is predicted to wear out

[0064] Preferably, the establishment of the data chain of feature data set 1, feature data set 2 and feature data set 3 comprises the following steps:

[0065] Acquire key feature parameter set 1, key feature parameter set 2 and shooter's arm motion trajectory data, and perform standardization processing;

[0066] Associating the data points in the key feature parameter set 1 with the data points in the key feature parameter set 2 by calculating the Euclidean distance between the key feature parameter set 1 and the key feature parameter set 2;

[0067] Associating the shooter's arm motion trajectory data with data points in the key feature parameter set one by calculating the Euclidean distance between the shooter's arm motion trajectory data and the key feature parameter set one;

[0068] Arrange the associated data points in chronological order to form a data chain.

[0069] The present invention also provides a smart shooting range data management system, which is used to execute the smart shooting range data management method.

[0070] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the smart shooting range data management method.

[0071] Beneficial effects of the present invention:

[0072] 1. The present invention collects information from different data sources in real time, and uses the Internet of Things technology to download this information to a local server for centralized management and storage in the case where there is no direct data connection between them. By performing preprocessing (such as preliminary association) on two data with potential correlation, the retrieval efficiency in the subsequent data management process can be significantly improved. In addition, based on the historical data analysis results and access log records, we will build multiple data channels, each corresponding to a specific type of data set, to meet the diverse data analysis needs.

[0073] 2. The present invention identifies potential failures or wear of weapons by analyzing the data in the weapon sample data set. In addition, by studying the motion characteristics of the corresponding structures of the weapon and the motion characteristics of the shooter's arm, we can explore how these structures affect the motion trajectory of the arm. The shooter can be guided to adjust his arm posture and force method to improve shooting performance. At the same time, after analyzing the correlation between the arm movement characteristics and shooting results, we can better understand which factors affect the results, and then take corresponding measures to optimize the training process. This method not only helps to speed up the improvement of shooting level, but also reduces the demand for actual targets to a certain extent, thereby reducing the cost of operating the shooting range. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 The present invention is a flow chart of a smart shooting range data management method. DETAILED DESCRIPTION

[0075] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.

[0076] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0077] refer to Figure 1 The technical solution provided by the present invention is: a smart shooting range data management method, comprising the following steps:

[0078] 1. Collect video data 1 from the first data source at a preset frequency and store the video data 1 locally. The video data 1 includes weapon shooting video data 1 and shooter video data.

[0079] 2. Collect video data 2 from a second data source, and transmit the video data 2 to cloud storage; the video data 2 includes weapon shooting video data 2.

[0080] 3. Download video data 2 and video data 1 from the cloud to form a sample data set and store it locally;

[0081] 4. Divide the video data 1 into a first video data group and a second video data group. Specifically, divide the weapon shooting video data 1 into the first video data group and add a collection timestamp 1; divide the shooter video data into the second video data group and add a collection timestamp 2.

[0082] Feature data are extracted from the first video data group, the second video data group and the video data group 2 respectively to form feature data set 1, feature data set 2 and feature data set 3. Specifically, feature data is extracted from the first video data group to form feature data set 1 ,in, Indicates weapon one The motion vector of each structure; Indicates the total number of weapon structures;

[0083] Extract feature data from the second video data set to form feature data set 2 ,in, Indicates shooter The motion vectors of the joints;

[0084] Extract feature data from video data 2 to form feature data set 3 ,in, Indicates weapon 2 The motion vector of the structure.

[0085] In this embodiment, weapon one is a weapon used for shooting at a shooting range, and video data one is a shooting video of weapon one shot by a high-speed camera during shooting training at the shooting range.

[0086] Weapon 2 is a brand new weapon, the same model as Weapon 1. Video data 2 is a video of Weapon 2 shooting test shot by a high-speed camera.

[0087] 5. Establish an association between feature data set 1 and feature data set 2 in the time dimension, specifically:

[0088] Use sliding window one to extract multiple feature data from feature data set one to form time series one, and use sliding window two to extract multiple feature data from feature data set two to form time series two;

[0089] Use the dynamic time warping (DTW) algorithm to match time series 1 and time series 2 in time and output the matching result;

[0090] According to the matching result, the time series one and the time series two are combined into a shooting sample data set, and the shooting sample data set is used to establish the relationship between the movement of each structure of the weapon one and the corresponding arm movement of the shooter.

[0091] 6. Establish an association between feature data set 1 and feature data set 3 in the time dimension, specifically:

[0092] Use sliding window three to extract multiple feature data from feature data set three to form time series three;

[0093] The dynamic time warping (DTW) algorithm is used to temporally match time series one and time series three, and a matching result is output; the so-called matching result is a distance matrix, which represents the minimum distance between time series one and time series three.

[0094] According to the matching results, a weapon sample dataset is constructed using time series 1 and time series 3. Specifically, if the distance between a certain time series 3 and time series 1 is less than a preset threshold, they are considered to belong to the same category and are combined into one sample.

[0095] The degree of weapon wear and the probability of wear occurring are analyzed through the weapon sample dataset.

[0096] In this embodiment, the arm movement trajectory of the shooter is obtained by analyzing the feature data set 2, including the following steps:

[0097] Extracting a plurality of continuous frames of images from the collected video data of the second video data group; preprocessing the extracted plurality of continuous frames of images, wherein the preprocessing includes graying and denoising;

[0098] Use pre-trained YOLOv5 to identify the position of the shooter’s arm in each frame.

[0099] Add multiple identification points to the arm image; use the Kalman filter to associate the identification points of the arm in consecutive frames to form a trajectory; smooth the trajectory data to obtain the trajectory data of the arm movement, and add a shooting timestamp.

[0100] After obtaining the arm's running trajectory, the shooting performance data corresponding to this shooting is read from the target system in real time, and the shooting performance timestamp is added. The relationship between arm movement and shooting performance is analyzed using the two types of data obtained. Specifically:

[0101] Align the shooting timestamp and shooting score timestamp on the time axis, use the arm motion trajectory data and shooting score combination as sample data to train the shooting score prediction model, and predict the shooting score based on the shooter's arm motion trajectory. The specific steps include:

[0102] Use Keras to build an LSTM model as a shooting performance prediction model; extract the key feature data of the arm at multiple time points from the arm motion trajectory data to form an arm motion feature time series, wherein the key feature data of the arm includes movement speed, acceleration and movement angle change rate; normalize the key feature data of the arm and input it into the shooting performance prediction model; the shooting performance prediction model outputs the predicted shooting performance.

[0103] Then, the sample data set is obtained again, and the time series one matching the arm movement trajectory is extracted from the sample data set; the time series one is aligned with the shooting performance data on the time axis to form a shooting action improvement data set; and the data basis is used to improve the shooting training program.

[0104] In this embodiment, the steps of forming the shooting performance prediction model are as follows:

[0105] Use the Sequential class to create a sequential model; set up the first LSTM layer, set up 50 neurons in this layer, return the output of each time step, and specify the shape of the input data; set up the second LSTM layer, set up 50 neurons in this layer, but only return the output of the last time step.

[0106] Add a fully connected layer using the Dense layer with an output dimension of 1. Compile the model using the Adam optimizer and the mean squared error loss function. After denormalization, output the predicted score.

[0107] In reality, the shooting range needs to understand the degree of wear of weapons so as to perform maintenance or replace weapons in advance. The present invention provides a method, which is specifically:

[0108] By analyzing the data in the weapon sample dataset, potential failures or wear of the weapon are identified, including the following steps:

[0109] Get weapon sample dataset ;

[0110] in, Represents a sliding window The data sequence obtained by moving ;

[0111] Indicates the sliding window The data sequence obtained by moving , and Respectively represent the step size of sliding window one and sliding window three; ;

[0112] in, , express Moment Weapon 1 The speed of the structure, express Moment Weapon 1 The direction angle of the structure's motion velocity, express Moment Weapon 1 The movement distance of each structure;

[0113] in, , express Time Weapon Part 2 The speed of the structure, express Time Weapon Part 2 The direction angle of the structure's motion velocity, express Time Weapon Part 2 The movement distance of each structure;

[0114] from A key feature parameter set 1 is formed after normalization; the key feature parameter set 1 includes a weapon 1 and a The maximum movement speed of the structure , maximum direction angle , minimum movement speed and minimum direction angle ;

[0115] from Extract key feature parameter 2 from the weapon, and after normalization, form key feature parameter set 2; the key feature parameter 2 includes weapon 2 Maximum movement speed of the structure , maximum direction angle , minimum movement speed and minimum direction angle ;

[0116] Import the pre-trained weapon evaluation model:

[0117] , in, They represent weight one, weight two, weight three and weight four respectively; represents the intercept, Regression coefficients; represents the error term;

[0118] Inputting key feature parameter set 1 and key feature parameter set 2 into the weapon assessment model includes the following steps:

[0119] Key Characteristics Parameter Set 1 and Key Characteristics Parameter Set 2 are input into the weapon assessment model: , output the wear prediction value;

[0120] Setting the evaluation threshold ,if , then the weapon is judged to be slightly worn;

[0121] if , then the weapon is judged to be moderately worn;

[0122] if , then the weapon is judged to be severely worn;

[0123] Weapon Assessment Model Taking the key characteristic parameter set 1 as input variables, the probability value of weapon wear is output;

[0124] Setting probability thresholds ,if , then it is predicted that the weapon will wear out.

[0125] 7. Establish a data link of feature data set 1, feature data set 2 and feature data set 3, and store them locally for analyzing shooting behavior, predicting weapon performance degradation and optimizing shooting strategies. Specifically:

[0126] Acquire key feature parameter set 1, key feature parameter set 2 and shooter's arm motion trajectory data, and perform standardization processing;

[0127] Associating the data points in the key feature parameter set 1 with the data points in the key feature parameter set 2 by calculating the Euclidean distance between the key feature parameter set 1 and the key feature parameter set 2;

[0128] Associating the shooter's arm motion trajectory data with data points in the key feature parameter set one by calculating the Euclidean distance between the shooter's arm motion trajectory data and the key feature parameter set one;

[0129] The associated data points are arranged in chronological order to form a data chain. The data chain integrates the motion vector data of each structure of the weapon, the arm motion trajectory data of the shooter, and the motion vector data of each structure during the reference weapon test. The smart shooting range can quickly obtain the corresponding data and its associated data when needed, providing a data basis for monitoring the degree of weapon wear and optimizing the shooter's shooting skills. It is convenient for the corresponding algorithms and models to quickly call the corresponding data, analyze the impact of weapon wear on shooting results, the impact of the shooter's arm movement on shooting results, and predict the shooting results based on the swing of the arm, so that the shooting results can be quickly obtained.

[0130] In this embodiment, the association between key feature parameter set 1 and key feature parameter set 2 is taken as an example for description.

[0131] By calculating the Euclidean distance between the key feature parameter set 1 and the key feature parameter set 2, the data points in the key feature parameter set 1 and the key feature parameter set 2 are associated, including the following steps:

[0132] Create an empty list to store the association results; traverse the data set, and for each data point in key feature parameter set one, calculate its Euclidean distance with all data points in key feature parameter set two.

[0133] Find the data point with the smallest distance to the current data point and record it in an empty list; finally return all associated pairs.

[0134] In some preferred embodiments, for each data analysis task, the task analysis log information records the task start time, end time, status (success / failure), error information, etc. At the same time, the data call log records the call time, call parameters, return results, etc.

[0135] By writing scripts, relevant information is extracted from task analysis logs and data call logs at a preset frequency and time, and the most executed analysis tasks and the analysis tasks with the most failures and error messages are analyzed. According to the data call logs, the call data of the task analysis is obtained, and multiple data chains are automatically constructed, as well as the corresponding pre-association between the two data sources.

[0136] The present invention also provides a smart shooting range data management system, which is used to execute the smart shooting range data management method.

[0137] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the smart shooting range data management method.

[0138] The embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. The embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by the central processing unit (CPU), the above-mentioned functions defined in the method of the present application are executed. It should be noted that the above-mentioned computer-readable medium of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of an electrical, magnetic, optical, electromagnetic, infrared segment, or semiconductor, or any combination of the above. More specific examples of computer-readable storage media may include: an electrical connection with one or more wire segments, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including wireless, electrical wire, optical cable, RF, etc., or any suitable combination of the foregoing.

[0139] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0140] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments. Without departing from the principles, the implementation methods of the present invention may be deformed or modified in any way.

Claims

1. A smart shooting range data management method, characterized in that: The method comprises: Collecting video data 1 from a first data source at a preset frequency, and storing the video data 1 locally; Collecting video data 2 from a second data source, and transmitting the video data 2 to cloud storage; Download video data 2 and video data 1 from the cloud to form a sample data set and store it locally; The video data 1 is divided into a first video data group and a second video data group, and feature data are extracted from the first video data group, the second video data group and the video data 2 respectively to form a feature data set 1, a feature data set 2 and a feature data set 3; Establishing an association between feature data set 1 and feature data set 2 in the time dimension; including the following steps: Use sliding window one to extract multiple feature data from feature data set one to form time series one; Use sliding window 2 to extract multiple feature data from feature data set 2 to form time series 2; Use the dynamic time warping (DTW) algorithm to match time series 1 and time series 2 in time and output the matching result; According to the matching result, the time series 1 and the time series 2 are combined into a shooting sample data set, wherein the shooting sample data set is used to establish a relationship between the movement of each structure of the weapon 1 and the corresponding arm movement of the shooter; Establish an association between feature data set 1 and feature data set 3 in the time dimension; Based on the existing historical data analysis records and data access records, build multiple data links that can support various specific analysis needs; The method of extracting feature data from the first video data set, the second video data set and the second video data set to form feature data set one, feature data set two and feature data set three comprises the following steps: Extract feature data from the first video data set to form feature data set one ,in, Indicates weapon one The motion vector of each structure; Indicates the total number of weapon structures; Extract feature data from the second video data set to form feature data set 2 ,in, Indicates shooter The motion vectors of the joints; Extract feature data from video data 2 to form feature data set 3 ,in, Indicates weapon 2 The motion vector of each structure; The step of establishing association between the feature data set 1 and the feature data set 3 in the time dimension includes the following steps: Use sliding window three to extract multiple feature data from feature data set three to form time series three; Use the DTW algorithm to match time series 1 and time series 3 in time, and output the matching results; According to the matching results, the weapon sample dataset is constructed using time series 1 and time series 3; The weapon sample dataset is used to analyze the degree of weapon wear and the probability of wear, including: By analyzing the data in the weapon sample dataset, potential failures or wear of the weapon are identified, including the following steps: Get weapon sample dataset ; in, Represents a sliding window The data sequence obtained by moving ; Indicates the sliding window The data sequence obtained by moving , and Respectively represent the step size of sliding window one and sliding window three; ; in, , express Moment Weapon 1 The speed of the structure, express Moment Weapon 1 The direction angle of the structure's motion velocity, express Moment Weapon 1 The movement distance of each structure; in, , express Time Weapon II The speed of the structure, express Time Weapon II The direction angle of the structure's motion velocity, express Time Weapon II The movement distance of each structure; from A key feature parameter set 1 is formed after normalization; the key feature parameter set 1 includes a weapon 1 and a The maximum movement speed of the structure , maximum direction angle , minimum movement speed and minimum direction angle ; from Extract key feature parameter 2 from the weapon, and after normalization, form key feature parameter set 2; the key feature parameter 2 includes weapon 2 Maximum movement speed of the structure , maximum direction angle , minimum movement speed and minimum direction angle ; Import the pre-trained weapon evaluation model: in, They represent weight one, weight two, weight three and weight four respectively; represents the intercept, Regression coefficients; represents the error term; Inputting key feature parameter set 1 and key feature parameter set 2 into the weapon assessment model includes the following steps: Key Characteristics Parameter Set 1 and Key Characteristics Parameter Set 2 are input into the weapon assessment model: , output the wear prediction value; Setting the evaluation threshold ,if , then the weapon is judged to be slightly worn; if , then the weapon is judged to be moderately worn; if , then the weapon is judged to be severely worn; Weapon Assessment Model Taking the key characteristic parameter set 1 as input variables, the probability value of weapon wear is output; Setting probability thresholds ,if , it is predicted that the weapon will be worn out; Weapon 1 is a weapon used for shooting at a shooting range; Weapon 2 is a brand new weapon and is the same model as Weapon 1.

2. A smart shooting range data management method according to claim 1, characterized in that: The video data 1 includes weapon shooting video data 1 and shooter video data, and the video data 2 includes weapon shooting video data 2; The step of dividing the video data into a first video data group and a second video data group comprises: The weapon shooting video data 1 is divided into the first video data group, and the acquisition timestamp 1 is added; the shooter video data is divided into the second video data group, and the acquisition timestamp 2 is added; The construction of multiple data links that can support various specific analysis needs is specifically: establishing data links of feature data set 1, feature data set 2 and feature data set 3, and storing them locally for analyzing shooting behavior, predicting weapon performance degradation and optimizing shooting strategies.

3. A smart shooting range data management method according to claim 2, characterized in that: The step of establishing a data chain of feature data set 1, feature data set 2 and feature data set 3 comprises the following steps: Acquire key feature parameter set 1, key feature parameter set 2 and shooter's arm motion trajectory data, and perform standardization processing; Associating the data points in the key feature parameter set 1 with the data points in the key feature parameter set 2 by calculating the Euclidean distance between the key feature parameter set 1 and the key feature parameter set 2; Associating the shooter's arm motion trajectory data with data points in the key feature parameter set one by calculating the Euclidean distance between the shooter's arm motion trajectory data and the key feature parameter set one; Arrange the associated data points in chronological order to form a data chain.

4. A smart shooting range data management system, characterized in that: The system is used to execute a smart shooting range data management method as described in any one of claims 1-3.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement a smart shooting range data management method as described in any one of claims 1 to 3.

Images