A training aid system for quantifying and analyzing the human-gun-target relationship in skeet shooting sports.
By deploying camera arrays and IMU modules within the skeet shooting training area, synchronous acquisition and fusion analysis of multi-source data from humans, firearms, and skeets were achieved. This solved the problems of insufficient data accuracy and interference in existing technologies, provided in-depth technical diagnosis and optimization suggestions, and supported routine training.
Patent Information
- Application Number
- CN202611580846.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-13
- Publication Date
- 2026-11-17
AI Technical Summary
Existing skeet shooting auxiliary analysis technologies suffer from limitations such as single data acquisition dimensions and insufficient precision, inability to realistically recreate three-dimensional shooting scenarios, lack of deep integration of "human-gun-target" collaborative quantitative analysis capabilities, disconnect between virtual simulation systems and live-fire training, and data acquisition methods that interfere with athletes' normal movements, making them unsuitable for routine long-term tracking and use.
By deploying a camera array within the skeet training area to collect video data, combining this with an IMU module to acquire gun movement data, and simultaneously recording human three-dimensional posture, gun movement, and skeet trajectory, a unified time series and spatial coordinate system is established. Multi-source data fusion analysis is then performed to construct a non-invasive data acquisition system.
It enables strongly coupled computation of biomechanics, kinematics, and dynamics on the same time axis, providing in-depth technical diagnosis and optimization suggestions, improving data processing efficiency, reducing interference with athlete training, and supporting routine training and long-term data accumulation.
Smart Images

Figure CN123111002A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present disclosure relates to the technical field of intelligent training assistance for competitive sports, and more particularly, to a flying disc shooting sport human-gun-target relationship quantification and analysis training assistance system, and the invention name of the present disclosure can also be a flying disc shooting sport human-gun-target coordination analysis method and system. BACKGROUND
[0002] Skeet shooting, a competitive sport demanding extremely high precision, timing, and human-machine coordination, currently employs several technological solutions for motion capture, technical diagnosis, and simulation training. These solutions primarily rely on two-dimensional image analysis, wearable inertial sensor monitoring, and virtual simulation and performance statistics. Two-dimensional image analysis is currently the most commonly used method. Early approaches often employed high-speed cameras with a shooting speed of 100 frames per second, capturing images from a side plane to analyze and obtain data such as the speed and angle of raising and moving the gun. However, when athletes perform movements such as body rotation, displacement in the third dimension cannot be captured, leading to data distortion. More recently, high-speed cameras (with sampling rates typically between 50Hz and 120Hz) mounted to the side of the athlete are commonly used to record the shooting process. This is combined with video processing software, allowing for manual slow-motion playback to calibrate key moments such as gun movement and firing, thereby measuring the athlete's right shoulder angle, stability angle, balance angle, and the gun's movement angle and angular velocity. In terms of gun trajectory capture, one approach involves attaching reflective markers to the underside of the barrel and using two external cameras in conjunction with SIMI Motion analysis software to capture the barrel's horizontal and vertical displacement. Wearable inertial sensor monitoring aims to acquire more refined continuous kinematic data. Existing technology has introduced wearable acquisition systems based on inertial measurement units (IMUs). By wearing multiple inertial sensor nodes at specific locations on the athlete's body and the gun, acceleration and angular velocity information of the limbs and the gun are collected. While this approach can acquire multi-node dynamic data, it requires wearing hardware over a large area of the athlete's torso and joints, making it highly invasive. Furthermore, this system is limited to collecting localized data from the person and the gun, and does not incorporate the true three-dimensional trajectory of the "target" in high-speed flight into synchronous acquisition and strongly coupled calculations. In terms of indoor simulation training and macro-level data statistics, existing technologies tend to employ digital virtual reconstruction or pure data entry methods. For example, existing technology (application number: 202511270589.1, application date: 2025.09.08) discloses a digital immersive skeet shooting simulation system. The software algorithm generates a virtual skeet trajectory using a parabolic equation and projects it onto a white background. A gyroscope or accelerometer mounted on the simulated gun captures the muzzle aiming coordinates in real time, and a computer determines whether the aiming coordinates and the simulated skeet coordinates are within the error range to decide whether a hit has occurred. This type of virtual system is detached from the real outdoor target-throwing environment and lacks real weather interference and recoil feedback from live-fire shooting. In terms of performance analysis, data is mainly collected through traditional manual note-taking, photo uploads, or simple electronic scoring methods. This allows for macro-level trend analysis of hit rate, miss rate, and ammunition usage at different positions, but it fails to deeply reveal the micro-level spatial misalignment and technical issues of the "person-gun-target" relationship at the moment of firing.
[0003] Therefore, the existing UFO shooting auxiliary analysis technology has the following problems:
[0004] 1. Data acquisition is limited in dimension and lacks precision, making it difficult to recreate a realistic 3D shooting scene: Traditional video analysis methods and wearable sensor-based acquisition systems can only acquire data in isolation from a single dimension. They cannot simultaneously acquire human posture, gun line 3D vector, and disc target 3D spatial coordinates with high precision. As a result, the existing systems provide very little data and cannot accurately reproduce a realistic 3D shooting scene on the same timeline.
[0005] 2. Lack of in-depth quantitative analysis capabilities for the synergistic relationship between the human, gun, and target: Existing technologies typically study shooting actions in isolation, failing to couple human biomechanics, gun kinematics, and target trajectory; they also fail to quantify the core dynamic indicators between the human, gun, and target. This makes it difficult to uncover the underlying causes of misses and to provide in-depth technical diagnostics.
[0006] 3. The virtual simulation system is seriously out of sync with live-fire training and cannot solve the technical optimization problems in real competition: The digital immersive skeet shooting simulation system cannot capture and correct the slight biomechanical deformations of athletes under real shooting environment and high-pressure confrontation, and cannot solve the problems of improving hit rate and technical optimization in real competition.
[0007] 4. Data collection methods significantly interfere with athletes' normal movements: Current motion capture systems or advanced testing equipment use highly invasive wearable devices that not only alter the original weight and balance of the firearm, but also severely interfere with athletes' muscle memory and real shooting feel, resulting in distorted data.
[0008] 5. Inability to support routine long-term tracking and use: Existing skeet shooting analysis systems involve cumbersome setup, debugging, and data processing workflows. Video analysis heavily relies on manual frame-by-frame playback and calibration, resulting in significant delays in technical feedback. This interferes with daily live-fire training, leads to inefficient data flow, and is only suitable for short-term research or occasional sampling tests, making it unsuitable for integration into regular training. Furthermore, the system cannot continuously monitor athletes' technical movements over long periods, hindering the accumulation of coherent individual technical data and making it difficult for coaching teams to grasp the long-term, longitudinal patterns of athletes' technical fluctuations.
[0009] Therefore, the problems existing in the above-mentioned prior art have become problems that urgently need to be solved in this field. Summary of the Invention
[0010] In view of this, this disclosure provides a method and system for analyzing the coordination between the athlete, shotgun, and target in skeet shooting. By synchronously recording the athlete, shotgun, and skeet target during live-fire training, the system acquires three-dimensional human posture data, shotgun motion data, shotgun line spatial vector, and skeet three-dimensional flight trajectory data. The system then performs quantitative analysis of the coordination between the athlete, shotgun, and target under a unified time series and a unified spatial coordinate system to solve the aforementioned technical problems.
[0011] One aspect of this disclosure provides a method for analyzing the human-target coordination in skeet shooting, comprising:
[0012] Multi-source data acquisition includes: acquiring video data using a camera array deployed within the skeet training area; acquiring raw IMU sampling data using IMU modules deployed below the muzzle or barrel of the firearm; and acquiring skeet launch time data using a skeet launch time acquisition module. Prior to data acquisition, the process also includes: calibrating the intrinsic and extrinsic parameters of the cameras in the camera array; establishing a transformation relationship between the camera coordinate system and the unified spatial coordinate system of the target range; and performing time synchronization processing on the skeet trajectory data and the raw video data acquired by the camera array.
[0013] The multi-source data obtained from the data acquisition is processed to obtain structured basic data. The multi-source data includes the original video data, image frame sequences, camera calibration parameters, IMU raw sampling data, and UFO launch time data. The structured basic data includes: human three-dimensional posture data, gun three-dimensional feature point data, gun line vector data, UFO three-dimensional trajectory data, firing time data, and data quality identifiers.
[0014] The structured basic data is spatiotemporally matched to obtain a multi-source fusion dataset of human, gun, and target under the same shooting process. Based on the dataset, the human-gun-target synergy analysis index is calculated, and the athlete's shooting technique is diagnosed according to the index results to generate diagnostic results. The synergy analysis includes athlete movement process analysis, gun movement process analysis, skeet target movement process analysis, gun-target synergy analysis, and overall human-gun-target synergy analysis.
[0015] Based on the diagnostic results, a technical action optimization plan is generated.
[0016] Another aspect of this disclosure provides a human-target coordination analysis system for skeet shooting, comprising: a camera array module, an IMU module, a skeet launch time acquisition module, a data acquisition module, a data processing module, a database module, a coordination analysis module, and a report output module, wherein...
[0017] The camera array module is coupled to the acquisition module and is used to acquire multi-view video data of athlete movements, gun movements, and the flight process of the skeet shooting training area; the camera array module is deployed in the skeet shooting training area.
[0018] The IMU module, coupled to the acquisition module, is installed below the muzzle or barrel of the gun and is used to acquire the acceleration, angular velocity and vibration impact signals generated at the moment of firing, and to identify the firing time of the first target and the firing time of the second target.
[0019] The UFO launch time acquisition module is coupled to the acquisition module and is deployed at the launch port or covering the launch port area to acquire the physical launch time of the UFO.
[0020] The acquisition module is coupled to the data processing module and is used to uniformly control and record the video acquisition tasks, athlete information, training items, target information, target type, hit results and acquisition status during the skeet shooting training process.
[0021] The data processing module is coupled to the acquisition module, the database module, and the collaboration analysis module. The data processing module processes the raw video data, camera calibration parameters, training task information, and firing time event data transmitted by the acquisition module and the IMU module to generate structured basic data. The data processing module also transmits the structured basic data to the database module for storage and to the collaboration analysis module for calculating and analyzing human-gun-target collaboration indicators.
[0022] The database module is coupled to the acquisition module, the data processing module, the collaborative analysis module, and the report output module, respectively, and is used to receive data generated by the acquisition module, the data processing module, the collaborative analysis module, and the report output module, and to store, manage, associate, and retrieve the data.
[0023] The coordination analysis module is coupled to the data processing module, calculates the human-gun-target coordination analysis index based on the structured basic data, and outputs the technical diagnostic results; wherein, the coordination analysis includes athlete movement process analysis, gun movement process analysis, skeet target movement process analysis, gun-target coordination analysis, and overall human-gun-target coordination analysis;
[0024] The report output module is coupled to the collaborative analysis module and the database module, respectively, and is used to summarize, organize and display the analysis results of the collaborative analysis module and the stored data of the database.
[0025] Compared with existing technologies, the method and system for analyzing the coordination between humans and targets in skeet shooting provided in this disclosure achieve at least the following beneficial effects:
[0026] First, the method disclosed herein breaks through the barriers of single-dimensional technical analysis, strongly coupling biomechanics (force generation and center of gravity displacement of various joints in the human body), equipment kinematics (spatial transfer of the gun line vector), and target dynamics (parabolic flight state of the flying saucer) on the same time axis. This in-depth fusion analysis allows instructors to clearly see the causal relationships between various shooting stages, providing data support for analyzing the correlations between shooting stages and possible technical reasons. It achieves a deep deconstruction of the dynamic synergistic relationships throughout the entire shooting process, providing a completely new perspective for exploring the laws of shooting victory.
[0027] Secondly, this disclosure addresses the highly coupled "person-gun-target" characteristics of skeet shooting by constructing a collaborative evaluation index system oriented towards the real training process. It quantitatively evaluates the entire process of an athlete completing a single shot from multiple dimensions. A scientific diagnostic logic is established, moving from index results to technical problems. Athletes' technical issues can be categorized based on the type of index anomaly, the direction of deviation, the magnitude of deviation, changes in timing and rhythm, and comparisons with historical data. Through this approach, this disclosure transforms the somewhat vague judgments of coaches' experience into measurable, comparable, and traceable quantitative indicators and diagnostic conclusions, elevating skeet shooting technical analysis from outcome evaluation to process diagnosis. The results of this scientific technical diagnosis provide objective evidence for subsequent movement optimization, training plan adjustments, and the establishment of long-term technical files for athletes.
[0028] Third, the method disclosed herein abstracts the gun body into a dynamic spatial vector (gun line vector) formed by the line connecting the gun's rear to its front, and combines this with an IMU module to achieve millisecond-level precise anchoring of the firing moment. Through spatial analytical geometry algorithms, subjective judgment is transformed into a quantifiable spatial geometric error equation, which can intuitively determine the cause of the miss.
[0029] Fourth, the method disclosed herein improves the efficiency of converting raw video into structured analysis data. It transforms the traditional inefficient process of relying on manual video recording and analysis into a semi-automated, structured, and repeatable data processing workflow, thereby increasing the speed of training data feedback and providing support for routine testing and rapid technical diagnostics in skeet shooting.
[0030] Fifth, this disclosure presents a multi-source heterogeneous data synchronous acquisition architecture for live-fire shooting scenarios, integrating light, electricity, and shadow. The system uses a camera array to cover the target positions at the firing range and the flight airspace of the high and low target houses. Simultaneously, it combines a laser light curtain sensor and an IMU module to construct a three-dimensional perception matrix. It synchronizes the visually captured 3D trajectory of the flying saucer, the markerless motion-captured 3D skeletal coordinates of the human body, and the high-frequency vibration and angular velocity signals of the gun acquired by the IMU at millisecond levels. This innovative multi-source acquisition completely breaks down the data silos of traditional equipment, fully reconstructing the real shooting scene within the same three-dimensional spatiotemporal coordinate system, laying an extremely solid and irreplaceable data foundation for subsequent in-depth "human-gun-target" collaborative analysis.
[0031] Sixth, this disclosure adheres to a "zero-interference" non-intrusive design philosophy. The system primarily relies on a camera array to collect the athlete's three-dimensional posture and weapon movement information, using only a lightweight IMU module on the weapon. This allows for the acquisition of weapon movement data without significantly affecting the weapon's original weight distribution and motion state, thus minimizing the impact on the athlete's normal training movements. This enables athletes to conduct live-fire shooting under completely realistic competitive conditions and psychological pressure. It completely overcomes the limitations of "short-term laboratory testing" and can be perfectly integrated into daily, all-day routine training, thereby establishing a highly valuable long-term digital archive of technological evolution for each elite athlete.
[0032] Of course, any product implementing this disclosure does not necessarily need to achieve all of the technical effects described above at the same time.
[0033] Other features and advantages of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description
[0034] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments of the present disclosure and, together with their description, serve to explain the principles of the present disclosure.
[0035] Figure 1 This is a schematic diagram of a method for analyzing the coordination between humans, guns, and targets in skeet shooting.
[0036] Figure 2 This is a flowchart of a method for analyzing the human-gun-target synergy in skeet shooting.
[0037] Figure 3 This is a schematic diagram of the camera array layout in a method for analyzing the coordination between humans and targets in skeet shooting sports, as disclosed in this paper.
[0038] Figure 4 This discloses the connection relationship between the modules of a human-gun-target coordination analysis system for skeet shooting.
[0039] Figure 5 This is a structural diagram of a human-gun-target coordination analysis system for skeet shooting sports. Detailed Implementation
[0040] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values of the components and steps set forth in these embodiments do not limit the scope of the present disclosure.
[0041] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use.
[0042] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.
[0043] In all the examples shown and discussed herein, any specific values should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.
[0044] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.
[0045] Reference Figures 1-3 This disclosure provides a method for analyzing the human-gun-target coordination in skeet shooting, including:
[0046] S1. Recording the motion process: Multi-source data acquisition is performed, including video data acquisition using a camera array deployed within the skeet training area, raw IMU sampling data acquisition using IMU modules deployed below the muzzle or barrel of the gun, and acquisition of skeet launch time data through a skeet launch time acquisition module; before data acquisition, the following steps are also included: calibrating the intrinsic and extrinsic parameters of the cameras in the camera array, establishing the transformation relationship between the camera coordinate system and the unified spatial coordinate system of the target range; and performing time synchronization processing on the skeet trajectory data and the raw video data obtained by the camera array.
[0047] S2. Obtaining Structured Basic Data: Processing the multi-source data acquired through data collection to obtain structured basic data. The multi-source data includes the original video data, image frame sequences, camera calibration parameters, IMU raw sampling data, and UFO launch time data. The structured basic data includes at least: human 3D posture data, firearm 3D feature point data, gun line vector data, UFO 3D trajectory data, firing time data, and data quality identifiers. The data quality identifiers refer to the data validity and reliability markers added by the system to each type of data when generating basic data for collaborative analysis. These markers are used to determine whether the data can be used for subsequent indicator calculations, whether manual review is required, or whether it should be discarded. The processing specifically includes:
[0048] S2.1 Based on the label-free motion capture method, human key points are identified in video images of athletes from at least two camera perspectives to obtain the coordinates of the athletes' two-dimensional key points in different video images. After data cleaning of the two-dimensional key point coordinates, combined with the stereo calibration parameters of the cameras, three-dimensional reconstruction is performed on the two-dimensional key points from different camera perspectives to obtain the three-dimensional coordinate sequence of the athletes' human key points in the unified spatial coordinate system of the shooting range. Among them, the human key points include at least the left shoulder joint, right shoulder joint, left elbow joint, right elbow joint, left wrist joint, right wrist joint, left hip joint, right hip joint, left knee joint, right knee joint, left ankle joint, right ankle joint, seventh cervical vertebra, and top of the head.
[0049] S2.2. Obtain the two-dimensional coordinates of the gun head and gun tail in video images from at least two camera perspectives, automatically match them according to the frame number, and perform three-dimensional reconstruction in combination with the camera's stereo calibration parameters to obtain the three-dimensional coordinates of the gun head and gun tail. Take the direction from the three-dimensional coordinates of the gun tail to the three-dimensional coordinates of the gun head as the gun line direction to obtain the gun line vector. Based on the three-dimensional coordinates of the gun head, the three-dimensional coordinates of the gun tail, and the gun line vector, calculate the gun kinematic data. The gun kinematic data includes at least one of the following: gun trajectory, change of gun line direction, angular velocity of gun line, speed of gun line movement, gun movement path, change of muzzle elevation, and the spatial relationship between the gun and the target at the moment of firing.
[0050] S2.3 Select the frame interval to be processed in the video image, obtain the UFO trajectory from at least two camera perspectives, generate the UFO's two-dimensional coordinates, combine the camera's stereo calibration parameters, perform three-dimensional reconstruction, and obtain the UFO's three-dimensional coordinate sequence in the unified spatial coordinate system of the target range.
[0051] S3. Perform spatiotemporal matching on the structured basic data to obtain a multi-source fusion dataset of the shooter, rifle, and target under the same shooting process. Based on the dataset, calculate the shooter-gun-target synergy analysis index, and diagnose the athlete's shooting technique based on the index results, generating diagnostic results. The synergy analysis includes athlete movement process analysis, rifle movement process analysis, skeet target movement process analysis, rifle-target synergy analysis, and overall shooter-gun-target synergy analysis. Spatiotemporal matching includes:
[0052] Perform time matching on the structured basic data and convert it into a unified timeline with a unified time point as the zero point;
[0053] Two-dimensional pixel coordinates of the same target point are extracted from at least two camera viewpoints. Based on the established transformation relationship between each camera pixel coordinate system and the unified spatial coordinate system of the target range, the two-dimensional pixel coordinates under different viewpoints are matched. The three-dimensional coordinates of the target point in the unified spatial coordinate system of the target range are obtained by projection matrix intersection or triangulation. The target point includes human joint points, gun head points, gun tail points and UFO center of mass points.
[0054] Using a unified time series as an index, multi-source data is organized according to each time point (or each frame) to form a multi-source fusion dataset of human, gun, and target under the same shooting process. The dataset includes at least: timestamp, relative target release event, three-dimensional coordinates of each joint of the human body, three-dimensional coordinates of the gun head, three-dimensional coordinates of the gun tail, gun line vector, IMU three-axis acceleration, IMU three-axis angular velocity, UFO three-dimensional coordinates, and firing time identifier.
[0055] S4. Based on the diagnostic results, generate a technical action optimization plan.
[0056] Specifically, in step S1, multiple industrial cameras or high-speed network cameras are deployed within the skeet shooting training area to form a multi-view acquisition array covering the athletes, firearms, and the skeet flight area. Specifically, for capturing the athletes' human movements, multiple cameras are set up around the two-way skeet shooting range to record images from multiple perspectives, providing an image foundation for subsequently obtaining the athletes' three-dimensional posture data. For capturing the skeet target trajectory, in the aforementioned camera array, at least two cameras are used to capture the athlete's complete shooting action at each target position. At least two cameras are set up to cover the launch areas of both the high and low target houses and the main flight airspace of the skeet, enabling a complete record of the skeet's spatial motion from launch, flight, to the moment of firing. This setup provides image data for subsequent reconstruction of the skeet's three-dimensional flight trajectory. It should be noted that the camera shooting angles are arranged to ensure complete imaging of the skeet, effective matching of left and right views, and reconstruction of the three-dimensional trajectory. Specifically, at least two cameras used for UFO trajectory acquisition should share a common field of view, ensuring that the main motion of the UFO from launch, flight, to near firing or fragmentation is simultaneously captured in the images of at least two cameras. The two cameras should have a certain spatial baseline and angle to avoid their shooting directions being completely parallel or too close, ensuring subsequent 3D reconstruction based on binocular or multi-view visual geometry. In a preferred embodiment, the two UFO trajectory acquisition cameras can be deployed in a diagonally lateral, cross-shooting manner, meaning the two cameras are located at different lateral positions within the main flight area of the UFO, with their lenses both facing the main flight airspace from launch to firing, creating a parallax-based two-dimensional trajectory for the UFO in the images from the left and right cameras. Depending on site conditions and the UFO's flight altitude, one or both cameras can also be set to a slightly higher or lower position for overhead or low-angle shooting, but should avoid shooting only from the front or back along the UFO's flight direction to reduce issues such as insufficient depth information, UFO obstruction, or trajectory overlap. For two-way skeet shooting training, camera groups for acquiring skeet trajectories in the high-target room and the low-target room can be deployed around the corresponding launch point and the main flight airspace of the skeet, respectively. The camera lens direction should cover the launch point, the skeet intersection area, and the area near the firing point, ensuring that the high-target or low-target skeet can be captured simultaneously by at least two camera perspectives within the effective analysis frame. After camera deployment, the system calibrates the skeet trajectory acquisition cameras using intrinsic parameters (including camera focal length, principal point coordinates, pixel ratio, tilt factor, radial distortion parameters, and tangential distortion parameters), extrinsic parameters (including the camera's rotation matrix and translation vector relative to the unified spatial coordinate system of the target range), and a joint calibration, obtaining the spatial relationship between the cameras. Subsequently, the system reconstructs the 3D trajectory of the skeet in the unified spatial coordinate system of the target range based on the 2D coordinates of the skeet in the left and right camera images, the camera calibration parameters, and the multi-view stereo vision geometry.
[0057] In step S1, time synchronization processing refers to configuring multi-device time synchronization, including network time synchronization, hardware synchronization triggering, or setting unified timestamp recording rules, so that each camera video frame, UFO launch signal, and IMU firing signal has a unified time reference. During data acquisition, the system records the camera number, frame number, and timestamp for each camera's acquired video frame, and records the corresponding time for the UFO physical launch event and IMU firing event. Time synchronization processing enables human, firearm, and UFO data to be analyzed correspondingly within the same frame sequence or on the same timeline.
[0058] Specifically, step S2, obtaining structured basic data, refers to processing the data of the human body, firearms, and flying saucer targets after completing the recording of the motion process to obtain basic data for collaborative analysis.
[0059] Step S2.1: Obtain Athlete's 3D Human Posture Data: The athlete's 3D human posture data includes a sequence of 3D coordinates of key points on the athlete's body during shooting. The sequence of 3D coordinates of key points represents the 3D coordinates of these key points at each frame or time stamp. Using these key points, a skeletal model of the athlete during shooting can be constructed. Further data cleaning includes confidence filtering, outlier removal, short-term missing point interpolation, and smoothing filtering to reduce the impact of image recognition errors, occlusion errors, and key point jitter on subsequent 3D reconstruction results. The athlete's 3D human posture data is organized by frame number or time stamp and includes at least frame sequence information, the name of each key point, and the corresponding 3D coordinate values.
[0060] Step S2.2: Obtain raw shotgun data, including shotgun line vector data and shotgun kinematic data. The shotgun line vector data is obtained through visual recognition. Specifically, the system imports shooting videos of athletes from two camera perspectives and segments the videos into single-frame images according to the frame sequence. Then, the system processes the gun feature points in each frame. In each image, the target area where the gun is located is first selected (frame annotation), and then the positions of the muzzle and butt are clicked or identified (control point annotation). The center points at both ends of the double barrel are used for the muzzle and butt to improve recognition accuracy. Finally, the system outputs the gun feature point recognition results, which may include the frame number, the two-dimensional coordinates of the muzzle and butt in the left and right views, the reconstructed three-dimensional coordinates of the muzzle and butt, and the shot line vector.
[0061] Step S2.3: Obtain the 3D trajectory data of the UFO target: After importing the video, select the frame interval to be processed. The preferred frame interval is the segment between the UFO being launched and breaking up or flying out of the effective analysis area. It should be noted that the selection of the frame interval is a semi-automatic process. The operator can manually select the start and end frames to be processed based on the UFO's launch, firing, breaking up, or flying out of the effective analysis area; the system can also provide recommended frame segments based on the launch time, target location, trajectory continuity, or preset rules, which are ultimately confirmed or adjusted by the operator. After selecting the frame interval, the UFO trajectory is automatically identified in the camera images from two different perspectives based on the previous deep learning, generating the UFO's 2D coordinates at the corresponding frame number. For erroneous trajectories generated during the automatic identification process, the system can automatically identify and remove them, or the operator can manually delete or confirm them. Based on rules such as the continuity of the UFO trajectory, speed changes, spatial range, and identification confidence level, obvious jump points, false detection points, or points that do not conform to the UFO's motion patterns can be automatically marked or removed. For trajectory points that cannot be accurately judged, the operator can manually review and delete them to improve the accuracy of the trajectory data. For missed frames or missing points, there are two methods for filling in the missing points. These methods can be used independently or in combination. The first method involves the operator selecting the missing frame segment or the range for filling in the missing points, and the system performing linear interpolation, spline interpolation, or trajectory fitting to fill in the missing points based on the preceding and following trajectory points. The second method involves the operator manually marking the UFO's position in the corresponding image frame, and the system using the manually marked points as supplementary trajectory points. Once the UFO trajectories in both the left and right views have been corrected, the operator clicks the matching function. The system automatically matches the two-dimensional coordinates of the UFO in the corresponding frames in the left and right views according to the frame number or timestamp, and performs three-dimensional reconstruction based on the camera calibration parameters to obtain the three-dimensional coordinate sequence of the UFO in the unified spatial coordinate system of the target range. Ultimately, the system generates the UFO trajectory, which includes at least the frame number, timestamp, and three-dimensional coordinates of the UFO.
[0062] Specifically, step S3, the human-gun-target collaborative analysis of the motion process, refers to matching the structured basic data according to a unified time axis and a unified spatial coordinate system to form a multi-source fusion dataset of human-gun-target under the same shooting process.
[0063] The specific spatiotemporal matching method is as follows: First, the structured basic data undergoes temporal and spatial matching to ensure that data from different sources can be collaboratively analyzed under the same time reference and spatial coordinate system. Temporal matching addresses the time asynchrony problem between data from multiple cameras, IMUs, and UFO target detection. It transforms the structured basic data to a unified time axis with the physical launch time of the UFO as zero. Through temporal matching, the system can obtain human, firearm, and UFO data at the same frame sequence or time point, thus avoiding errors in the human-firearm-target correspondence caused by acquisition delays, frame rate differences, or sampling frequency differences between different devices. Spatial matching addresses the inconsistency of coordinate systems between different camera perspectives, measurement areas, and data sources. Before system use, the intrinsic and extrinsic parameters of each camera are calibrated. For binocular camera groups with a common field of view, a calibration board can be used for joint calibration; for UFO trajectory acquisition cameras with a large coverage area, range control points, RTK control points, or other known 3D coordinate points can be used for calibration. Through the above calibration, the transformation relationship between the pixel coordinate system of each camera, the camera coordinate system, and the unified spatial coordinate system of the shooting range is established. For human posture data, the system spatially matches the same human key points identified from different camera perspectives to reconstruct the three-dimensional coordinates of the human key points. For firearm data, the system spatially matches the two-dimensional coordinates of the gun head and butt from different camera perspectives to reconstruct the three-dimensional coordinates of the gun head and butt, forming a gun line vector. For UFO trajectory data, the system matches the two-dimensional coordinates of the UFO in the same frame or at the same timestamp in the left and right cameras to reconstruct the three-dimensional trajectory points of the UFO in the unified spatial coordinate system of the shooting range. It should be noted that in the generation of the spatiotemporal fusion dataset, for data with short-term missing data, the system can interpolate and complete it based on data from adjacent time points; for data with severe occlusion, insufficient recognition confidence, or that cannot be effectively reconstructed, the system sets an invalid mark and does not participate in the calculation of the corresponding time point's collaborative index. Through the aforementioned spatiotemporal matching method, the system can correspond human movements, gun movements, and UFO flight processes on the same time axis and in the same spatial coordinate system, enabling subsequent indicators such as gun-target spatial deviation, gun-target motion synchronization, human posture control, gun holding stability, and shooting rhythm to be accurately calculated based on the same shooting event.
[0064] Step S4, Technical Movement Optimization: After completing the above human-gun-target synergy analysis, the system further generates technical movement optimization plans based on the results of various synergy indicators, which are used to guide athletes in targeted training and movement correction.
[0065] In some optional embodiments provided in this disclosure, processing the multi-source data obtained from data acquisition to obtain structured basic data further includes:
[0066] Based on the three-dimensional coordinates of the athlete's key human body points in the unified spatial coordinate system of the shooting range, the kinematic parameters of the athlete during the shooting process are calculated. These kinematic parameters include, but are not limited to, at least one of joint angles, trunk posture, center of gravity position, displacement, velocity, and angular velocity. In other words, based on the three-dimensional coordinates of the key human body points, the kinematic parameters of the athlete during the shooting process can be selectively calculated further.
[0067] In some optional embodiments provided in this disclosure, the coordination analysis includes, but is not limited to, the following: 1. Human motion process analysis: based on the three-dimensional coordinates of each joint of the athlete, calculate the athlete's body posture changes, joint angle changes, center of gravity movement trajectory, support stability, and movement rhythm characteristics during the shooting process, to determine the athlete's movement stability and coordination during initiation, rotation, raising the gun, tracking the target, and firing; 2. Gun motion process analysis: based on the three-dimensional coordinates of the gun head and tail and video analysis data, calculate the shotgun's trajectory direction, angular velocity, rate of change of angular velocity, transient vibration characteristics during firing, and gun holding stability, to determine the athlete's gun control ability during tracking the target and firing; 3. Skeet target motion process analysis: based on the three-dimensional trajectory data of the skeet target... 4. Calculate the changes in the position, speed, flight direction, and trajectory of the flying saucer to describe its actual spatial motion under different target positions and launch directions; 5. Calculate the spatial geometric calculations of the gun line vector and the three-dimensional trajectory of the flying saucer to obtain the spatial relative relationship between the gun line and the flying saucer at the firing moment or a preset key moment. This relative relationship includes the shortest Euclidean distance between the gun line and the center of mass of the flying saucer, horizontal deviation, vertical deviation, depth deviation, and the degree of synchronization between the direction of gun line movement and the direction of flying saucer flight; 6. Calculate the overall coordination characteristics of the human-gun-target, integrating human posture, gun movement, and flying saucer trajectory, to calculate the coordination characteristics of the athlete from target reaction, start-up, gun handling, target tracking to firing, and identify the key technical factors that lead to hit or miss.
[0068] The specific indices for calculating human-gun-target coordination include: target spatial deviation, target temporal deviation, deviation direction, reaction time index, body stability index, gun-holding stability index, shooting rhythm index, and posture control index; the calculation methods include:
[0069] Construct the gun line vector This includes: after obtaining the two-dimensional coordinates of the gun head and tail from two camera perspectives, the system automatically matches them according to the frame number and reconstructs the three-dimensional coordinates of the gun head and tail by combining them with camera calibration parameters. Gun line vector. The line vector from the butt of the gun to the tip is: ; for the line vector of the gun Perform normalization to obtain the unit gun line vector. Guns in The gun line at time (frame) is ;in, Let be the three-dimensional coordinates of the gun head at time t; Let t be the three-dimensional coordinates of the gun tail. The parameter is a linear parameter; that is, the gun is in... The gun line at a given moment can be represented as a straight line in space, when When the value is >0, it means that the gun line extends forward from the rear of the gun along the direction of the gun head;
[0070] UFO trajectory modeling is performed. Based on the characteristics of relatively fixed bidirectional UFO trajectories and predictable motion patterns, the trajectories of the high-target-room and low-target-room UFOs are modeled as parabolas located in a fixed trajectory plane in three-dimensional space. Specifically, this includes fitting the UFO trajectory plane based on the UFO's three-dimensional coordinate point set data. UFO trajectory plane for: Where a, b, c, and d are trajectory plane parameters. The three-dimensional coordinate components in the unified spatial coordinate system of the target range can be determined by least-squares fitting of the three-dimensional trajectory points of the flying saucer; in the flying saucer trajectory plane Inside, the relative time after the flying saucer leaves the target. The trajectory points below are ,in, For UFOs in The three-dimensional spatial position at a given moment. The relative time is defined as the moment the UFO physically leaves the target. For UFOs in The coordinate values along the X-axis in the unified spatial coordinate system of the target range at all times. For UFOs in The coordinate value along the Y-axis in the unified spatial coordinate system of the target range at any time. For UFOs in The coordinate value along the Z-axis in the unified spatial coordinate system of the target range at any time; this trajectory can be obtained by interpolation of measured three-dimensional coordinate points or by fitting a parabolic model in the trajectory plane;
[0071] Calculate the straight line of the gun. and the plane of the UFO trajectory The intersection point, the intersection point is ,in, When the denominator is not 0 and When the value is greater than 0, it indicates that the gun line along the gun tip direction and the UFO trajectory plane have a valid intersection point; if the UFO moves to a certain point at a certain moment... If the gun trajectory is near the location of the flying saucer, the direction of the gun trajectory corresponds to the location of the flying saucer. If there is a deviation between the actual location of the flying saucer and the intersection point, it indicates that there is a spatial or temporal mismatch between the gun's direction and the location of the flying saucer. If the intersection point of the gun trajectory is higher or lower than the flying saucer's trajectory, the system judges that the athlete has a problem with the gun muzzle being too high or too low.
[0072] Calculate the spatial deviation between the gun and the target: the spatial deviation between the gun and the target at the moment of firing. ,in, The point where the firing line intersects the plane of the UFO's trajectory at the moment of firing; The three-dimensional coordinates of the flying saucer at the moment of firing; the projectile's flight time is not considered. ), , The smaller the value, the closer the gun is to the location of the flying saucer at the moment of firing; The larger the value, the greater the deviation between the gun's pointing direction and the flying saucer's position at the moment of firing; considering the projectile's flight time better aligns with the technical meaning of "lead" in real shooting. A smaller value indicates a better match between the athlete's firing aim and the future location of the flying saucer. A larger deviation indicates a problem with the lead, muzzle height, or direction of gun movement.
[0073] Calculating the time deviation of the target shot includes: finding the target shot in the UFO trajectory parameter curve. The UFO trajectory point with the smallest distance ,in, The moment when the distance between the UFO trajectory point and the intersection of the gun line is minimum; calculate the target-gun time deviation. , For the projectile's flight time, when , If the projectile's flight time is not considered, then ≈0 indicates that the intersection of the gun lines matches the current position of the UFO; if A value >0 (or the upper threshold) indicates that the firing line is pointing towards the location the UFO will eventually reach, which is manifested as excessive lead or premature firing; if A value less than 0 indicates that the firing line is pointing towards the position already passed by the flying saucer, which manifests as target lag or insufficient lead; if the projectile's flight time is considered, when When the value is close to 0, it indicates that the firing line of the gun at the moment of firing is well matched with the position of the skeet when the projectile arrives; when it deviates significantly from 0, it can be judged that the athlete has insufficient or excessive lead; when it exceeds the preset reasonable upper limit, it indicates that the lead is too large.
[0074] The deviation direction decomposition includes: decomposing the deviation vector E into the unified spatial coordinate system of the target range or the local coordinate system of the UFO trajectory to obtain the deviation components in the vertical, tangential, and normal directions, wherein... , or ; the deviation vector The system decomposes the trajectory of the UFO into its local coordinate system: the tangential deviation along the UFO's flight direction is used to determine whether the lead is insufficient or excessive; the deviation perpendicular to the flight direction in the trajectory plane is used to determine the muzzle elevation control deviation, for example, when the vertical deviation is greater than 0, it indicates that the muzzle is too high; the out-of-plane deviation perpendicular to the UFO's trajectory plane is used to determine whether the firing line deviates from the UFO's main flight plane. Through the above deviation decomposition, the system can further explain the "miss" as a specific spatial cause, such as insufficient lead, excessive lead, muzzle too high, muzzle too low, or firing line deviating from the UFO's flight plane.
[0075] Calculate reaction time index , ,in, The moment the UFO physically launches from the target. This refers to the moment of gun activation or the moment of human activation. It should be noted that reaction time is the time interval between the moment the skeet physically leaves the target and the moment the athlete's gun or human movement shows a clear activation. The "clear activation" moment can be determined by the moment of gun activation or the moment of human movement activation, and is obtained through a threshold determination method. Specifically... The starting point can be determined based on continuous over-threshold changes in the angular velocity of the gun line, the gun head velocity, or the velocity of key points on the human body. If the training analysis focuses more on gun activation, the gun activation moment is used; if it focuses more on the athlete's body activation, the human movement activation moment is used. The gun activation moment can be determined based on IMU angular velocity or acceleration data. For example, the root mean square of the angular velocity can be calculated using a sliding time window. When several consecutive windows exceed a preset threshold, the first over-threshold sampling point is defined as the activation moment. This indicator is used to distinguish between problems such as slow athlete reaction, late activation, premature activation, or abnormal activation rhythm. When the standard deviation of the gun line angular velocity is greater than the warning threshold, it indicates that the gun line angular velocity fluctuates greatly and the gun handling is unstable.
[0076] Calculating the body's dynamic stability margin MoS includes: calculating the extrapolated center of mass. , ,in, For the athlete's overall well-being, For speed, , It is the acceleration due to gravity. The height is the center of mass; construct the support base BoS based on the bipedal support area, and calculate... The minimum distance to the BoS boundary is used to obtain the dynamic stability margin MoS. It should be noted that the body stability index is used to evaluate an athlete's dynamic balance control ability during target tracking, body rotation, and firing. This index can preferably be calculated using the dynamic stability margin MoS. A positive MoS value indicates that the athlete is in a relatively stable state; the larger the MoS, the higher the stability reserve; a negative MoS value indicates a risk of instability. This index is particularly suitable for evaluating lower limb support and body center of gravity control during body rotation in skeet shooting; when MoS is close to or below the lower limit of the reference range, it indicates insufficient body support.
[0077] Calculate gun stability index , The maximum Lyapunov index, ,in, The initial distance between adjacent trajectories. For the process Trajectory distance after each sampling interval; For the first Each sampling interval The time interval is specified. It should be noted that the rifle stability index is used to evaluate an athlete's ability to maintain steady-state control of the rifle during target tracking and firing; the maximum Lyapunov index is a preferred candidate for this index. The local dynamic stability of the barrel motion system is characterized. In the calculation, the IMU three-axis angular velocity and three-axis acceleration data from the time the flying saucer leaves the target to the completion of firing are first extracted and preprocessed by filtering and resampling. Then, the barrel motion feature sequence is selected for phase space reconstruction. Finally, the maximum Lyapunov exponent is calculated using the nearest neighbor trajectory divergence rate. The smaller the value, the slower the divergence of the barrel's trajectory, and the better the stability when holding the gun; The larger the value, the more unstable the barrel movement, and the easier it is to cause deviation of the firing line at the moment of firing; The time interval refers specifically to the time interval between two adjacent sampling points; if the data has already been resampled along a unified time axis, then... The time interval after resampling;
[0078] Shooting rhythm indicators, including time to one target, time to two targets, and total shooting time, are calculated by: separately recording the time to one target for at least two shots. Second target time Total shooting time Calculate the mean, standard deviation, and coefficient of variation for the time to shoot one target, the time to shoot two targets, and the total shooting time. The moment the UFO physically launches from the target. For the moment of firing a target, This is the firing time of the two targets;
[0079] Posture control indicators, calculated using methods including: calculating joint angles and the relative torsional angle of the shoulder and pelvis based on three-dimensional human posture data, where joint angles... A, B, and C are the coordinates of the joints involved in the joint angle calculation; the method for calculating the relative torsion angle between the shoulder and pelvis is as follows: construct the vector connecting the two shoulders and the vector connecting the pelvis; after projecting them onto the horizontal plane, calculate the relative angle. ; Calculate angular velocity It should be noted that posture control indicators are used to evaluate an athlete's joint angles and trunk control during shooting. These indicators are calculated based on three-dimensional human posture data, including shoulder / elbow / knee / ankle joint angles, head tilt angle, and shoulder-pelvis relative torsional angle; angular velocity, etc. It can be used to evaluate an athlete's body rotation range, rotation speed, movement stability, and trunk control at the moment of firing during the target pursuit process.
[0080] Specifically, the target spatial deviation and target temporal deviation are used to evaluate the degree of matching between the intersection of the gun line and the actual or predicted position of the flying saucer at the moment of firing; the target motion synchronization is used to evaluate the consistency between the direction of motion of the intersection of the gun line and the flight direction of the flying saucer; the human posture control index is used to evaluate whether the athlete's body movements are stable and coordinated; and the gun holding stability index is calculated based on the change of the gun line vector and is used to evaluate the athlete's ability to control the direction of the gun.
[0081] In some optional embodiments provided in this disclosure, the frame range is the frame segment between when the UFO is ejected and when the UFO breaks apart or flies out of the effective analysis area.
[0082] In some optional embodiments provided in this disclosure, time matching is performed on the structured basic data to convert it into a unified timeline with a unified time point as the zero point, including:
[0083] During the data acquisition phase, a unified time reference is established, binding each frame of video data acquired by each camera with its corresponding acquisition time to form a "frame number - timestamp" correspondence; the sampling time corresponding to each sampling point in the data acquired by the IMU module is recorded; the physical launch time of the UFO is recorded and defined as the zero point of the firing process; the first... The relative time of the frame to the moment the UFO physically leaves the target is: ,in, For the first The relative time of frame video capture compared to the moment the UFO physically leaves the target. For the first The absolute time of frame video capture. The moment the UFO physically launches from the target; IMU No. The relative time of each sampling point with respect to the moment the UFO physically leaves the target is: ,in, For IMU The relative time of each sampling point with respect to the moment the UFO physically leaves the target. For IMU The absolute time of each sampling point;
[0084] For data with different sampling frequencies, a unified time series is established based on the camera frame rate or a preset analysis frequency. Data from different data sources are aligned to the same time series using timestamp proximity matching, linear interpolation, or spline interpolation. Specifically, when the timestamp of the data to be matched is consistent with the target time point, the data is used directly. When the timestamps are not completely consistent, two adjacent sampling points before and after the target time point are selected for interpolation to obtain the original data at the target time.
[0085] In some optional embodiments provided in this disclosure, obtaining the three-dimensional coordinates of the target point in the unified spatial coordinate system of the target range further includes:
[0086] Obtain the 3D coordinates of the target point in the local 3D coordinate system;
[0087] Based on common control points or known target range coordinates, calculate the rigid body transformation relationships from each local coordinate system to the unified spatial coordinate system of the target range; obtain the three-dimensional coordinates of the target point in the unified spatial coordinate system of the target range, and the rigid body transformation relationships are as follows: ,in, For the three-dimensional coordinates in the unified spatial coordinate system of the target range, Three-dimensional coordinates in a local coordinate system. For rotation matrix, It is a translation vector;
[0088] Rotation matrix Translation vector The confirmation methods include: selecting several common control points or known target range coordinate points whose coordinates can be obtained in both the local coordinate system and the unified spatial coordinate system of the target range, and solving the rotation matrix through rigid body registration or least squares methods. Translation vector The rotation matrix is verified by checking the reprojection error, control point residuals, or 3D reconstruction error. When the error meets a preset threshold, the rotation matrix is confirmed. Translation vector It can be used to perform coordinate transformation calculations in rigid body transformation relationships.
[0089] That is, if different camera groups first establish their own local 3D coordinate systems, then based on common control points or known target range coordinate points, the rigid body transformation relationship from each local coordinate system to the unified spatial coordinate system of the target range is calculated. Through this coordinate transformation, the 3D posture of the human body, the 3D vector of the gun line, and the 3D trajectory of the flying saucer are all unified into the same unified spatial coordinate system of the target range. It should be noted that the translation vector... For camera calibration and coordinate system integration, the system uses fixed transformation parameters determined during the process. When a camera group or local 3D reconstruction results need to be transformed to the unified spatial coordinate system of the target range, the system calculates the rotation matrix of the local coordinate system relative to the unified spatial coordinate system of the target range using calibration boards, common control points, RTK control points, or other known 3D coordinate points. Translation vector .
[0090] In some optional embodiments provided in this disclosure, a technical action optimization plan is generated based on the diagnostic results, including: identifying technical problems of the athlete during shooting based on coordination indicators; classifying the technical problems and generating optimization suggestions based on the type of technical problem; wherein the type of technical problem includes at least one of gun-target relationship deviation, target following rhythm deviation, body posture control deviation, insufficient gun holding stability, and firing timing deviation.
[0091] For example, if there is a significant tangential deviation between the point of intersection of the firing line and the actual or predicted position of the flying saucer (center of mass) at the moment of firing, it indicates that the athlete's lead control is insufficient. If the point of intersection is located where the flying saucer has already passed, it is judged as insufficient lead or delayed target tracking; if the point of intersection is significantly too far from the future position of the flying saucer, it is judged as excessive lead or firing too early. For such problems, action optimization suggestions can be generated, such as "adjusting the target contact point position," "establishing a consistent relationship between the firing line and the flying saucer's flight direction in advance," "shortening or extending the target tracking time before firing," and "optimizing the rhythm of switching between the first and second targets."
[0092] When the intersection point of the gun trajectory shows a significant vertical or off-plane deviation relative to the skeet trajectory, it indicates a deficiency in the athlete's control over muzzle height or spatial direction of the gun trajectory. If the intersection point is higher than the skeet trajectory, it suggests the athlete is raising the muzzle too high or over-controlling their upper limbs; if the intersection point is lower than the skeet trajectory, it suggests the athlete is lowering the muzzle too low or lacking support. For these issues, optimization suggestions can be generated, such as "adjust gun holding height," "optimize shoulder, elbow, and wrist coordination," "control the muzzle elevation or depression," and "keep the gun trajectory near the plane of the skeet trajectory."
[0093] When the synchronization between the gun and target is low, it indicates that the athlete's gun-carrying direction or speed is not matching the flight pattern of the skeet. To address this issue, the speed of the gun's trajectory intersection and the skeet's flight speed can be combined to determine whether the athlete's tracking speed is insufficient, too fast, or deviating from the skeet's path. Correspondingly, training suggestions can be generated such as "improving the continuity of the body's gun-carrying movement," "reducing the arm's independent gun-pulling," "maintaining the gun's trajectory in the same direction as the skeet's flight," and "optimizing the continuous rhythm from viewing the target, raising the gun, to firing."
[0094] When an athlete's reaction time, first-target time, second-target transition time, or total shooting time is found to be abnormal, it indicates a problem with the athlete's shooting rhythm organization. If the reaction time is too long, it suggests a slow start after viewing the target; if the first-target firing time is too long, it suggests slow target reception, excessive tracking time, or hesitation in firing; if the second-target transition time is too long, it suggests disjointed recovery, body rotation, or second-target reception after firing the first target. To address these issues, action optimization suggestions can be generated, such as "shortening the time from viewing the target to starting," "optimizing the first-target firing rhythm," "strengthening the transition from one target to two targets," and "establishing a stable target release-gun-tracking-firing rhythm chain."
[0095] When an athlete's center of gravity fluctuates excessively, their dynamic stability margin is insufficient, or there are abnormal changes in the lower limb support area, it indicates a problem with their body stability. To address this, we can analyze changes in hip, knee, ankle, and torso posture to determine if the athlete is experiencing forward / backward swaying, lateral shifting of the center of gravity, unstable lower limb support, or insufficient body alignment with the weapon during rotation. Correspondingly, training suggestions can be generated such as "stabilize lower limb support," "reduce large shifts in the center of gravity before firing," "enhance the hip's ability to drive the torso and weapon in synchronized rotation," and "maintain body axis stability at the moment of firing."
[0096] When the angular velocity of the rifle fluctuates significantly, the instantaneous vibration during firing is obvious, or the stability indicators of holding the rifle are abnormal, it indicates that the athlete has deficiencies in rifle control. To address these issues, optimization suggestions can be generated, such as "reducing ineffective shaking during rifle handling," "reducing sudden stops or abrupt changes in the muzzle before firing," "improving the overall control of the rifle by the shoulder, elbow, and wrist," and "maintaining a stable firing line at the moment of firing."
[0097] Furthermore, personalized technical benchmarks can be established based on different target positions, different target launch directions, and different athletes' historical data. By comparing current training data with the athlete's historical best data, average data for the same period, or reference data of outstanding athletes, weaknesses in current training can be identified, and individualized training priorities can be generated.
[0098] During the training feedback process, a closed loop can be formed by following the steps of "problem identification - cause analysis - action suggestion - training retest". First, the main technical problems of this shooting are identified based on the synergy index; second, the causes of the problems are analyzed based on the correspondence between human posture, gun line vector, IMU data and skeet trajectory; third, action optimization suggestions are output for coaches and athletes; finally, data is collected again in subsequent training and the same indexes are calculated to evaluate the effect of technical action optimization.
[0099] Through the aforementioned technical optimization steps, this disclosure not only identifies whether an athlete hits the target or misses, but also further explains the spatial, temporal, and action-related reasons for the miss or technical fluctuations, transforming these into actionable training recommendations. As a result, athletes and coaches can make precise adjustments to key aspects, thereby improving training feedback efficiency and the targeted nature of shooting performance improvement.
[0100] Reference Figures 3-5 In another embodiment, this disclosure also provides a human-gun-target coordination analysis system for skeet shooting. This system implements the aforementioned human-gun-target coordination analysis method for skeet shooting. It can simultaneously collect motion data of the athlete, shotgun, and skeet target in a real skeet shooting training scenario, and perform quantitative analysis and technical diagnosis of the human-gun-target coordination relationship during the shooting process based on a unified time axis and a unified spatial coordinate system. The system includes: a camera array module 100, an IMU module 300, a skeet launch time acquisition module 200, a data acquisition module 400, a data processing module 500, a database module 800, a coordination analysis module 600, and a report output module 700.
[0101] The camera array module 100 is coupled to the acquisition module 400 to acquire multi-view video data of athlete movements, gun movements, and skeet flight within the skeet shooting training area. The camera array module 100 is deployed within the skeet shooting training area and includes no fewer than four cameras. At least two cameras capture the athlete's complete shooting action at each target position, and at least two cameras cover the entire skeet flight area. The acquisition module 400 controls the start / stop, exposure, frame rate, and data storage of the cameras in the camera array module 100. It also transmits multi-view video data to the acquisition module, which receives, stores, forwards, or creates data indexes, and provides the data to the data processing module 500.
[0102] The IMU module 300 is coupled to the acquisition module 400 and is placed below the muzzle or barrel of the gun. It is used to acquire the triaxial acceleration, triaxial angular velocity, vibration and impact signals generated at the moment of firing and the corresponding sampling timestamp, and to identify the moment of the first and second shots.
[0103] The UFO launch time acquisition module 200 is coupled to the acquisition module 400 and is deployed at or covering the launch port area to acquire the physical launch time of the UFO and use this time as a reference time point for subsequent time matching.
[0104] The acquisition module 400 is coupled to the data processing module 500 and is used to uniformly control and record the video acquisition tasks, athlete information, training items, target information, target type, hit results and acquisition status during the skeet shooting training process.
[0105] The data processing module 500 is coupled to the acquisition module 400, the database module 800, and the collaboration analysis module 600. The data processing module 500 processes the raw video data, camera calibration parameters, training task information, and firing time event data transmitted by the acquisition module 400 and the IMU module 300 to generate structured basic data. The data processing module 500 also transmits the raw data to the database module 800 for storage and to the collaboration analysis module 600 for calculating and analyzing the human-gun-target collaboration index.
[0106] The database module 800 is coupled to the acquisition module 400, the data processing module 500, the collaborative analysis module 600 and the report output module 700 respectively, and is used to receive the data generated by the acquisition module 400, the data processing module 500, the collaborative analysis module 600 and the report output module 700 and to store, manage, associate and retrieve the data.
[0107] The synergy analysis module 600 is coupled to the data processing module 500. It calculates human-gun-target synergy analysis indicators based on structured basic data and outputs technical diagnostic results. The synergy analysis includes athlete movement process analysis, gun movement process analysis, skeet target movement process analysis, gun-target synergy analysis, and overall human-gun-target synergy analysis. This module analyzes human biomechanics, gun kinematics, and skeet trajectory information on the same time axis and in the same spatial coordinate system, thereby transforming the coach's experience judgment into quantifiable indicator results.
[0108] The report output module 700 is coupled to the collaboration analysis module 600 and the database module 800, respectively. It is used to summarize, organize and display the analysis results of the collaboration analysis module 600 and the stored data in the database, and output training diagnostic reports for coaches, athletes and researchers.
[0109] Specifically, the camera array module serves as the foundation for the system's front-end visual acquisition, enabling human 3D pose reconstruction, gun line recognition, and UFO 3D trajectory reconstruction. Each camera connects to the acquisition module via hardware trigger cables, a synchronization controller, network switching equipment, or a synchronization acquisition gateway. The video data acquired by each camera includes the camera number, frame number, timestamp, and acquisition channel information, and is transmitted to the data processing module for subsequent target recognition, 3D reconstruction, and collaborative analysis. This camera array module is suitable for both two-way and multi-directional UFO shooting operations conducted within the same training area. In the two-way skeet shooting training mode, the camera array module preferably includes ten cameras, which are arranged in sections according to the high target room, low target room, eight shooting positions, and the skeet intersection area of the two-way skeet shooting range. Among them, some cameras are used to cover the main flight airspace of the skeet after it leaves the target in the high and low target rooms, so as to record the trajectory changes of the skeet from the time it leaves the target, through flight, to firing or breaking apart; some cameras are used to cover the athlete's complete shooting action process at different target positions, so that the process of the athlete from preparation, starting, carrying the gun, following the target, firing, to the end of the follow-up can be recorded by at least two camera perspectives at each target position; and some cameras can be used to assist in recording the spatial attitude changes of the gun and the changes in the direction of the gun line during the shooting process.
[0110] The camera array module and the acquisition module transmit raw video streams, camera numbers, frame numbers, timestamps, exposure parameters, and acquisition status information. After receiving, storing, forwarding, or creating data indexes, the acquisition module provides the video file path, image frame sequence, camera number, frame number, timestamp, and camera calibration parameter index to the data processing module. The camera array module and the data processing module also transmit video files, image frame sequences, and camera calibration parameter indexes stored or forwarded by the acquisition module. Based on this data, the data processing module performs human posture recognition, gun feature point recognition, UFO trajectory recognition, and 3D reconstruction. The camera array module indirectly participates in collaborative analysis through the human 3D posture data, gun 3D feature point data, and UFO 3D trajectory data output by the data processing module.
[0111] Specifically, intrinsic parameters describe the imaging relationship when a point in three-dimensional space is projected onto the camera's two-dimensional image plane. Extrinsic parameters refer to the position and attitude parameters of the camera relative to the unified spatial coordinate system of the target range. Among these, the rotation matrix... The translation vector is used to describe the directional relationship between the camera coordinate system and the unified spatial coordinate system of the target range. This describes the positional relationship between the origin of the camera coordinate system and the origin of the unified spatial coordinate system of the shooting range. Through the aforementioned camera array module, the system can simultaneously record multiple perspectives of athletes, firearms, and skeets without requiring athletes to wear reflective markers or human body sensors. This provides a unified visual data source for subsequent markerless human motion capture, firearm feature point recognition, skeet trajectory recognition, 3D reconstruction, and human-firearm-target collaborative analysis.
[0112] In this disclosure, the acquisition module serves as the data entry point between the camera array module, data processing module, and database module. It ensures that each shooting training session contains complete identity information, event information, target location information, time information, and acquisition status information. The acquisition module transmits acquisition start commands, acquisition stop commands, synchronization trigger signals, camera number, camera status, and video data with the camera array module. It transmits video file paths, image frame sequences, athlete information, event type, target number, target miss type, hit result, and acquisition time information with the data processing module. It transmits training batch information, athlete identity information, training task information, acquisition status information, and data index information with the database module. The acquisition module is indirectly linked to the coordination analysis module through the data processing and database modules, providing training scene labels and result labels for subsequent calculations of human-gun-target coordination indicators. Through the acquisition module, the system can uniformly record information during on-site training and correlate it one-to-one with the video data acquired by the camera array. Therefore, the system can not only complete the acquisition of raw video, but also provide a structured data foundation for subsequent human 3D pose reconstruction, gun line recognition, UFO trajectory recognition, collaborative index calculation, training data query and technical report generation.
[0113] In this disclosure, the data processing module is used to process the raw video data, image frame sequences, camera calibration parameters, training task information, and firing time event data transmitted by the acquisition module and the IMU module. The IMU module is primarily used to identify the firing times of the first and second shots and to calculate time-related indicators such as the time to the first target, the time to the second target, and the total shooting time. The gun motion information is obtained from the three-dimensional coordinates of the gun head, the three-dimensional coordinates of the gun tail, and the gun line vector data. The data processing module can be deployed on a data processing server, a training workstation, or a local computer.
[0114] The database module establishes a specialized training database for skeet shooting. Within this database, athlete information, training task information, raw data collection, processed result data, synergy indicator data, and report data are categorized and managed according to data type, data table, or data set format. The skeet shooting training database can be deployed on a local server, training site workstation, cloud server, or a combined local and cloud data storage environment. The database module receives data generated by other modules and provides data query, data retrieval, historical data comparison, and training file management functions to these modules. Athlete information data: used to record and manage basic athlete information. Training task data: used to record training scenario information corresponding to each training session, each set of data collection tasks, or each shot. Raw data collection index: used to record the raw data obtained by the data collection module and its storage location. Data processing result data: used to store the structured results output by the data processing module. Synergy indicator data: used to store the human-gun-target synergy indicator results, indicator thresholds, indicator levels, anomaly markers, and technical problem types calculated by the synergy analysis module. Diagnostic report data: used to store training diagnostic reports, periodic analysis reports, and long-term athlete technical files generated by the report output module. Operation Logs and Data Traceability Information: Used to record the addition, modification, deletion, and review of data during system use. Data Association Method: The skeet shooting training database associates different types of data through unified task numbers, athlete numbers, training batch numbers, shot numbers, and data file numbers. Data Query and Retrieval Functions: The database module is equipped with data query and retrieval functions to support athlete training data queries, single-shot data queries, target or stance data queries, event type data queries, hit result queries, and key indicator queries. It can also provide historical benchmark data for the collaborative analysis module. Through the above database modules, the system can uniformly manage field-collected data, 3D reconstruction data, collaborative indicator data, and training report data according to different data types within a single skeet shooting training database, avoiding the establishment of multiple fragmented databases.
[0115] The report output module can be deployed on local workstations, training ground servers, web platforms, or handheld terminals, and connects to the database module and the collaboration analysis indicator module. The report output module retrieves athlete information, training task information, raw video indexes, human 3D posture data, skeet 3D trajectory data, gun line vector data, IMU time event data, hit results, and collaboration indicator results from the database module, and generates single-shot reports, single-group training reports, daily training reports, and phased training reports according to preset report templates. Single-shot reports are used for detailed review of a single shooting process. Single-group training reports are used to summarize and analyze the results of multiple shots by the same athlete in the same training task. Daily training reports summarize all training data for an athlete within a specific training day. Phased training reports are used to track the athlete's technical changes over a training cycle. Visualized output content: The report output module can visualize key data. Technical diagnosis and movement optimization suggestions output: The report output module generates technical diagnostic opinions and movement optimization suggestions based on the collaboration analysis indicator results. The technical diagnostic report includes at least the following: rifle-target relationship diagnosis, rhythm diagnosis, posture diagnosis, stability diagnosis, and training focus suggestions. When the target spatial deviation is large, the report indicates a mismatch between the rifle's pointing and the clay pigeon's position. When the target time deviation is negative, the report indicates delayed target tracking or insufficient lead. When the target time deviation is significantly positive and exceeds a reasonable range, the report indicates excessive lead or premature firing. When the rifle's angular velocity fluctuates significantly, the report indicates unstable rifle handling or insufficient muzzle control before firing. When the time for one or two targets fluctuates significantly, the report indicates an unstable shooting rhythm. When there are abnormal changes in the body's center of gravity or key joint angles, the report indicates unstable body support or uncoordinated rifle handling. Corresponding training suggestions are generated simultaneously with the diagnostic report.
[0116] The report output module can generate electronic reports according to preset templates. After the report is generated, the report output module writes the report file, report number, generation time, corresponding athlete number, training task number, indicator result number, and original data index into the database module. Users can query historical reports by athlete name, training date, event type, target position or stance, hit result, and key indicators. Through the above report output module, the system can transform multi-source data and collaborative analysis results of human-gun-target into diagnostic reports that can be directly used for training feedback. This allows coaches and athletes to clearly understand the spatial deviation, temporal deviation, movement problems, and training priorities in each shot, thereby achieving a closed-loop training assistance from data collection and indicator calculation to the output of technical optimization suggestions.
[0117] Specifically, the UFO launch time acquisition module can be a laser light curtain module, deployed at the launch port. It obtains a relatively accurate physical launch time by blocking the signal generated by the beam as the UFO passes through the launch port. Alternatively, it can acquire video footage from cameras covering the launch port area and observe the launch port region to obtain the UFO launch time. The system can utilize video footage from cameras covering the launch ports of either high or low target houses, identifying the first frame in which the UFO appears at the launch port, pixel changes in the launch port area, changes in indicator lights on the target building, or manual annotation to determine the UFO launch time, and use this time as a reference point for time matching. All of these methods provide a unified time reference for human posture data, gun line vector data, UFO trajectory data, and IMU firing time event data.
[0118] In some optional embodiments provided in this disclosure, the acquisition module 400 includes an acquisition main unit and a handheld terminal. The acquisition main unit is coupled to the camera array module 100 and is used to control the start and stop, synchronization triggering, data reception and local storage of the cameras in the camera array.
[0119] The handheld terminal is coupled to the main acquisition unit and is used for training information input, acquisition task triggering, hit result recording, and acquisition status viewing.
[0120] Specifically, the main acquisition unit is deployed in the data acquisition host, server, or workstation at the training site and connected to the camera array module via a wired network, synchronization control cable, or acquisition gateway. Handheld terminals are deployed on-site at the training site and can be used by coaches, researchers, or auxiliary testing personnel. They connect to the main acquisition unit via a wireless or wired network. The handheld terminals include interfaces for athlete management, event selection, training content selection, target number selection, data acquisition control, line status display, and hit recording.
[0121] Example 1: Analysis of Person-Gun-Target Coordination in Two-Way Skeet Shooting Training
[0122] In this embodiment, the training program is two-way skeet shooting. The training content can be selected from high-target single-target training, low-target single-target training, or double-target training. The target position can be any of the target positions from 1 to 8. Taking the athlete performing double-target training at target position 7 as an example, the athlete completes 5 consecutive double-target shootings, with the 3rd shooting being the subject of detailed analysis.
[0123] The specific methods include: the system first selects the athlete number, training item, target number, training content, and hit record recording method through the handheld terminal in the acquisition module, and then binds the training task with the corresponding camera group, target type, and data processing rules.
[0124] Before data acquisition, a multi-view camera array should be deployed, with specific requirements as described above.Figure 3 As shown, ten cameras, C10 to C19, were deployed. Cameras C10 and C11 form a low-target room skeet trajectory acquisition camera group, used to record the flight trajectory of skeets in the low-target room; cameras C12 and C13 form a high-target room skeet trajectory acquisition camera group, used to record the flight trajectory of skeets in the high-target room; cameras C14 and C15 are used to record the shooting actions of athletes at targets 1, 2, and 8; cameras C18 and C19 are used to record the shooting actions of athletes at targets 6, 7, and 8; and cameras C16 and C17 are used to record the shooting actions of athletes at targets 3, 4, and 5.
[0125] Before training begins, as mentioned above, the system performs intrinsic parameter calibration, extrinsic parameter calibration, and synchronous calibration on each camera, and establishes the transformation relationship between the pixel coordinate system of each camera, the camera coordinate system, and the unified spatial coordinate system of the target range.
[0126] Once the athlete has finished preparing and begun training, the acquisition module sends a synchronous acquisition command to the camera array module, controlling the camera group corresponding to the current target position and training content to begin acquiring video data. Simultaneously, the laser light curtain module detects the interruption signal generated when the flying saucer passes the exit point and determines the moment corresponding to this signal as the physical exit point of the flying saucer, using it as a unified time zero point. The moment when the athlete's gun begins to exhibit continuous and effective movement is recorded as... The IMU module is installed below the shotgun muzzle to collect the impact signal generated at the moment of firing and to identify the moment of firing of a target. and the firing time of the second target .
[0127] After video acquisition, the data processing module processes the human, firearm, and skeet data separately, as described above. For human data, the system uses video from the camera group covering the athlete's movements at target position 7 for identification and reconstruction, obtaining the athlete's individual key points in the unified coordinate system of the shooting range, forming a three-dimensional coordinate sequence. For firearm data, the system imports shooting videos from two camera perspectives, which are then reconstructed and calculated by the firearm feature point processing unit.
[0128] For UFO data, the system selects the corresponding UFO trajectory camera group for either the high-target or low-target training area based on the current bidirectional UFO training content. For example, when analyzing a high-target UFO, it uses video from cameras C12 and C13; when analyzing a low-target UFO, it uses video from cameras C10 and C11. The system first selects the frame segment between the UFO's launch from the target and its firing, fragmentation, or exit from the effective analysis area, and then identifies the UFO's two-dimensional trajectory in the left and right camera images. After the trajectory correction for the left and right views is completed, the system matches the UFO's two-dimensional coordinates in the left and right images according to the frame number or timestamp, and performs three-dimensional reconstruction to obtain the UFO's three-dimensional coordinate sequence in the unified spatial coordinate system of the range.
[0129] Subsequently, the system performs time matching and space matching. The specific methods are as described above.
[0130] In the collaborative analysis phase, the system performs computational analysis based on the multi-source fusion dataset. In this embodiment, the specific steps are as follows:
[0131] For time-related index calculations: In this embodiment, the reaction time threshold is set within the range of 0.15s to 0.25s, the gun-launching phase threshold is set within the range of 0.25s to 0.32s, the first-target gun-carrying phase threshold is set within the range of 0.15s to 0.22s, the target-following phase threshold is set within the range of 0.20s to 0.30s, the second-target gun-carrying phase threshold is set within the range of 0.27s to 0.40s, and the total time threshold from the release of the clay pigeon to the firing of the second target is set within the range of 1.00s to 1.20s. It is understood that the above threshold ranges are examples of system parameter configurations; specific thresholds can be calibrated based on the athlete's personal historical data, training stage, and target position.
[0132] Calculate reaction time and firing rhythm indicators: It should be noted that if the initial firing phase... For self Until the butt of the rifle touches the face, the rifle carrying stage for one target. From the moment the rifle butt touches the face until firing at the target, the follow-through phase... From the firing of the first target to the start of the reverse gun transport phase, the second target gun transport phase... From the start of the reverse gun handling to the firing of the second target, then ; ;
[0133] Table 1 shows exemplary event moments and time metrics for 5 shots (unit: seconds).
[0134] Table 1
[0135]
[0136] The breakdown of the time for each stage in Table 1 is shown in Table 2 (unit: s), where the check sum is the sum of reaction time, gun-raising stage, gun-carrying stage for the first target, gun-following stage for the second target, and gun-carrying stage for the second target.
[0137] Table 2
[0138]
[0139] Calculate the average value for any time index of 5 shots. Sample standard deviation Coefficient of variation ,in, The calculation results are shown in Table 3: the total time variation coefficient is 2.92%, the first target time variation coefficient is 3.16%, and the second target conversion time variation coefficient is 3.01%, indicating that the time rhythm of the five shots has high repeatability; the average reaction time is 0.196 s, which is used to illustrate the time from when the athlete recognizes the flying saucer leaving the target to when he starts the gun movement.
[0140] Table 3
[0141]
[0142] In the analysis of the spatial relationship between the gun and the target, as mentioned earlier, the spatial deviation, temporal deviation, and directional components of the deviation are obtained by solving for the gun-target spatial deviation. This is then applied to the three-dimensional trajectory points of the same flying saucer. Plane fitting is performed to obtain the UFO trajectory plane Π. The spatial straight line of the gun line at the moment of firing is L(r) = + Find the intersection point of the gun trajectory and the plane of the UFO's trajectory. .
[0143] Considering the projectile's flight time At that time, with = As the reference time for the UFO, define the deviation vector ΔP= - The spatial deviation of the gun and target is =‖ΔP‖. The unit vector tangential to the UFO's motion. Vertical unit vector in the trajectory plane and plane normal vector If a local coordinate base is established, then the tangential deviation, vertical deviation, and out-of-plane deviation are respectively: , , , ;in, The positive direction is set as the direction of flight of the flying saucer. The positive direction is set to vertically upward. Since the trajectory plane is obtained by fitting a discrete trajectory, the original UFO trajectory points may have fitting residuals relative to the fitted plane. It can be used to characterize the out-of-plane deviation of the intersection of the gun lines relative to the actual position of the flying saucer. The calculation method for the gun-target time deviation is as described above.
[0144] For example, in this embodiment, the analysis range of the target spatial deviation is set to 0.03 to 0.30 m, the display range of the target time deviation is set to -0.05 to 0.10 s, and 0 to 0.05 s is set as the exemplary target interval. <0 indicates insufficient tangential lead. >0 indicates that the tangential advance is too large; >0 indicates that the muzzle is too high. <0 indicates that the muzzle is too low; An increase indicates a greater deviation of the gun trajectory from the plane of the UFO's trajectory.
[0145] Taking the third shot and the firing of the target as an example, let's take... =(0.350, -0.620, 1.420)m, =(1.050, -0.100, 1.510) m, calculated to obtain =(0.799, 0.593, 0.103). The fitting plane for the UFO trajectory is y=8.000 m. The intersection point of the gun line and the trajectory plane is calculated. =(11.954, 8.000, 2.912) m. Projectile flight time. =0.035s, then the reference time for one target is... =0.660 + 0.035 = 0.695s. The three-dimensional coordinates of the UFO at this moment are set as follows: =(12.054, 7.970, 2.872)m, therefore ΔP=(-0.100, 0.030, 0.040)m, thus obtaining =-0.100 m、 =0.040 m、 =0.030 m and =0.112 m. The UFO's trajectory is in line with... The moment when the distance is minimum is =0.689s, therefore =0.689-0.695=-0.006 s.
[0146] Table 4 shows exemplary results of the gun-target spatial relationship between targets one and two during the third shot: It can be seen that the gun-target time deviation for both targets one and two is less than 0, and the tangential deviation is negative, corresponding to insufficient lead or target lag. The absolute value of the negative tangential deviation for target two is larger, indicating that this problem is more pronounced after target two conversion. The vertical deviation for both shots is greater than 0, indicating a tendency for the muzzle to be too high; the increased deviation of targets two from the surface indicates a greater degree of deviation of the gun line from the UFO trajectory plane before firing.
[0147] Table 4
[0148]
[0149] In the gun-target motion synchronization analysis, the consistency of their motion directions is calculated based on the velocity of the intersection point of the gun line and the flying saucer trajectory plane, and the flying saucer's flight speed. This is used to evaluate the synchronization relationship between the athlete's gun movement direction and the flying saucer's flight direction. If the gun-target motion synchronization is low, the system can further combine the gun line angular velocity, flying saucer velocity, and changes in human posture to determine whether the athlete has insufficient or excessive following speed, or deviation in gun movement direction. The gun line direction can be determined by the horizontal azimuth angle. and pitch angle The exemplary data for the gun trajectory angle of the third shot are shown in Table 5.
[0150] Table 5
[0151]
[0152] Table 6 provides an example of the statistical analysis of the gun line angular velocity during each action phase.
[0153] Table 6
[0154]
[0155] For example, the average angular velocity of the gun line can be set in the range of 100–150 deg / s, and the standard deviation of the angular velocity can be set in the range of 10–30 deg / s for display and classification. The overall average angular velocity of the third shot is 122.9 deg / s, which is within the example range; the standard deviation during the second-target gun handling phase is 29.4 deg / s, which is close to the upper limit of the example range and higher than the example fluctuation warning threshold of 25 deg / s. Therefore, it can be judged that the angular velocity of the gun line fluctuates greatly before the second target is fired, indicating that the stability of holding the gun has decreased or there is a lot of muzzle correction.
[0156] Target motion synchronization: Calculation of the first... The cosine similarity between the velocity of the intersection point of the gun line and the velocity of the flying saucer at each sampling moment is linearly normalized to the range of 0 to 1 and averaged to obtain the synchronization degree S. For example, S > 0.85 can be set as excellent, 0.70 ≤ S ≤ 0.85 as average, and S < 0.70 as requiring significant improvement. The synchronization degree of the first target in the third shot is 0.86, which is in the excellent range; the synchronization degree of the second target is 0.77, which is in the average range. The decrease in the synchronization degree of the second target is consistent with the increase in the angular velocity fluctuation during the second target gun movement phase, which is used to characterize the reduced consistency between the gun line and the flying saucer's flight direction after the two targets are switched.
[0157] In human motion analysis, the system calculates joint angles, trunk posture, center of gravity movement trajectory, and body stability indicators based on the athlete's key point three-dimensional coordinates. If the system detects significant fluctuations in the athlete's center of gravity before firing, accompanied by instability in the gun trajectory, it determines that the athlete may have problems with lower limb instability or insufficient gun bearing. The specific calculation method is as described above. For example, the shoulder-pelvic torsion angle threshold can be set to 20°–50°, the elbow joint angle threshold can be set to 90°–120°, and the knee joint angle threshold can be set to 120°–150°. The posture indicators for the third shot are shown in Table 7.
[0158] Table 7
[0159]
[0160] In this embodiment, the MoS reference threshold range is set to 0.05-0.15 m, with a larger value indicating a more sufficient dynamic stability reserve. The MoS at the second target firing during the third shot is 0.058 m, close to the lower limit of the example range; for example, when MoS is less than 0.060 m, the system can provide a warning of insufficient body support. This value, together with the large fluctuation in the angular velocity of the gun line during the second target carrying phase, characterizes insufficient lower limb support and torso gun-carrying ability.
[0161] The calculation method for the gun-holding stability index is as described above. In this embodiment, It can be set from 0.2 to 0.8 s. -1 The value is displayed within a certain range; a smaller value indicates a slower local divergence of the barrel's trajectory and higher stability while holding the weapon. Calculations show that the third shot, from initiation to the firing of the second target... =0.68 s -1 Located on the higher side of the example range, corresponding to the standard deviation of angular velocity of 29.4 deg / s during the two-target gun-carrying phase, it serves as a warning of decreased gun line stability before firing.
[0162] After completing the above analysis, the system generates technical diagnostic results and action optimization suggestions for this two-way skeet training exercise. For the third shot, targeting two targets... =-0.013 s and =-0.220m, the system identifies this as insufficient lead or target lag; two targets =+0.070 m, the system also provides a minor warning that the muzzle is too high; the standard deviation of the angular velocity of the gun line during the two-target gun handling phase is 29.4 deg / s, and =0.68 s -1 The system identified this as a decrease in gun line stability before firing; when firing at the second target, MoS=0.058 m, and the system gave an associated warning about the small body support reserve.
[0163] System diagnosis: The athlete's main problems in this shooting were insufficient lead control and decreased stability of the gun line before firing.
[0164] Action optimization suggestions: 1. Establish the target contact point in advance; 2. Keep the gun line consistent with the flight direction of the flying saucer; 3. Reduce muzzle adjustments before firing; 4. Strengthen lower limb support and torso carrying ability.
[0165] Through the above methods, this disclosure can uniformly quantify the time rhythm, gun line movement, gun-target spatial relationship, human posture, body stability, and gun holding stability in two-way skeet shooting, and generate technical diagnostic results and action optimization suggestions that can be used for training assistance based on index symbols and threshold relationships.
[0166] Finally, the report output module can generate single-shot reports, single-group training reports, or phased training reports. Through this embodiment, the present disclosure can achieve closed-loop training assistance from data acquisition, 3D reconstruction, collaborative analysis to technical action optimization in real-world skeet shooting training scenarios.
[0167] Example 2: Analysis of Person-Gun-Target Coordination in Multi-Directional Skeet Shooting Training
[0168] This embodiment and Embodiment 1 use the same basic principle of human-gun-target synergy analysis. The main differences are in the training program, athlete's position, target launch direction, camera activation combination, and indicator interpretation method.
[0169] In this embodiment, the training program is multi-target skeet shooting. The training content may include left-facing targets, center-facing targets, right-facing targets, or other target launch directions set according to the rules of multi-target skeet shooting. The athlete's position can be set according to the multi-target skeet training positions. Taking the athlete training on a left-facing target in a certain multi-target skeet shooting position as an example, the training type is single-target shooting. The athlete is located in the 3rd position, and the athlete completes 5 shots in the same direction, with the 3rd shot being the subject of detailed analysis.
[0170] In this embodiment, since multi-directional skeet shooting and two-way skeet shooting can be conducted on the same training ground, based on the multi-directional skeet training's standing area, forward target throwing area, and main skeet flight sector, all or some cameras in the camera array module are selected to form a multi-directional skeet working camera group. Specifically, the system activates the camera group facing the athlete's standing area to record the athlete's complete actions at each shooting position in the multi-directional skeet, including preparation, target calling, target catching, gun handling, target tracking, firing, and the end of follow-up; simultaneously, the camera group facing the forward target throwing area and the main skeet flight sector is activated to record the spatial trajectory of the skeet after it flies out of the target throwing area.
[0171] The specific method differs from that of Example 1 in the following ways:
[0172] For single-target training, the system identifies the firing moment and calculates the time between target release and firing. For multi-shot or continuous training, the system can identify different firing events and calculate the corresponding firing time intervals. During data processing, the system first automatically calls the athlete movement camera group, gun recognition camera group, and skeet trajectory camera group corresponding to the multi-target skeet training based on the event type, station number, target release direction, and batch information recorded by the acquisition module. It then performs video frame extraction, image distortion correction, effective frame segment extraction, and timestamp alignment. For skeet data, based on the target release direction of the current training, it calls the camera group covering the forward target release area and the main flight sector.
[0173] During the coordination analysis phase, when the target launch direction is left-facing, the system focuses on evaluating the athlete's leftward body rotation and gun line following during the gun-carrying process; when the target launch direction is right-facing, the system focuses on evaluating the athlete's rightward body rotation, shoulder-hip coordination, and gun line control; when the target launch direction is center-facing, the system focuses on evaluating the athlete's frontal target contact, gun elevation height, and firing timing control. For different target launch directions, the system can determine whether the athlete has insufficient lead, excessive lead, excessively high muzzle, excessively low muzzle, or a gun line deviating from the plane of the flying saucer trajectory based on the spatial relationship between the intersection of the gun lines and the position of the flying saucer.
[0174] The system performs calculations and analyses based on this multi-source fusion dataset. In this embodiment, the specific steps are as follows:
[0175] Calculation of time-related indicators: In this embodiment, the reaction time threshold range is 0.15–0.30 s, and the shooting time threshold range is 0.70–1.20 s. The calculation process includes: establishing a coordinate system O-XYZ, with the X-axis pointing to the right side of the field, the Y-axis pointing forward, and the Z-axis pointing vertically upward. The left-facing target flies along the negative X-axis direction. This embodiment is a single-target shooting exercise, and the conversion time between two targets is not calculated. Example data table 8 (unit: s) shows the event times and time indicators for 5 multi-directional flying saucer shooting exercises in the same direction.
[0176] Table 8
[0177]
[0178] For any time index in the 5 shots, calculate its mean, sample standard deviation and coefficient of variation. The results are shown in Table 9.
[0179] Table 9
[0180]
[0181] As can be seen, the average reaction time is 0.224 s, the standard deviation is 0.009 s, and the CV is 4.03%; the average shooting time is 0.923 s, the standard deviation is 0.031 s, and the CV is 3.35%, both of which are within the exemplary range.
[0182] In human motion analysis, the exemplary threshold ranges in this embodiment are: shoulder-pelvic torsion angle threshold of 20°–50°, elbow joint angle threshold of 90°–120°, knee joint angle threshold of 120°–150°, and MoS threshold of 0.05–0.15 m. The third shooting index (Table 10) was calculated. At the time of firing, all joint angles were within the exemplary threshold range; MoS decreased from 0.106 m to 0.057 m, approaching the lower limit of the range, indicating a decrease in lower limb support reserve during leftward rotation and firing.
[0183] Table 10
[0184]
[0185] Calculation of the degree of ricochet before firing the firearm; in one embodiment, at the moment of firing... =(0.320,-0.580,1.410) m, =(-0.441, 0.343, 1.506) m, calculated to obtain (-0.634, 0.769, 0.080). For example, the three-dimensional trajectory of the UFO within the analysis time window of the third shot is: =(-15.125-21.600(t-0.937), 17.982+0.015(t-0.937), 3.301+0.850(t-0.937)-0.600(t-0.937)²)m. Fitting the trajectory points yields plane Π: y-18.000=0; the gun line is L(r)= + The intersection point is calculated as follows: =(-14.999, 18.000, 3.342) m. Projectile flight time. =0.032s, then =0.937 s, and =(-15.125, 17.982, 3.301)m. Take the left-hand tangential vector. =(-1, 0, 0), vertical vector =(0,0,1), normal vector With n=(0, 1, 0), we get ΔP=(0.126, 0.018, 0.041) m. =-0.126 m、 =+0.041 m、 =+0.018 m and =0.134 m. and The time of the UFO trajectory with the smallest distance is =0.689s, therefore =0.931-0.937=-0.006s. The spatial relationship indicators of the gun and target for the third shot are summarized in Table 11.
[0186] Table 11
[0187]
[0188] The third shot and All are negative, which corresponds to a delayed firing line and insufficient lead. The result is positive, therefore it also indicates that the gun muzzle is too high.
[0189] Gun stability is measured using the maximum Lyapunov index. Characterization. Exemplarily, in this embodiment, The threshold range is 0.2–0.8 s. -1 A higher value indicates poorer dynamic stability of the gun line. (Third shot) =0.66s -1 The 30.2 deg / s standard deviation of the angular velocity before firing together indicate a decrease in the stability of the firing line.
[0190] The system also calculates the gun-target synchronization based on the velocity of the intersection point of the gun's trajectory and the flight speed of the flying saucer. For example, S > 0.85 indicates high consistency, 0.70 ≤ S ≤ 0.85 indicates average consistency, and S < 0.70 indicates adjustment is needed. The calculation shows that S = 0.74 for the third shot, indicating a deviation between the gun's movement direction and the leftward flight direction of the target.
[0191] In the analysis of gun motion, the calculated indicators of the gun are shown in Table 12 (taking the third shot as an example).
[0192] Table 12
[0193]
[0194] The average angular velocity from the start of the third shot to the firing interval was 121.8 deg / s, with a standard deviation of 27.6 deg / s; the standard deviation within 0.15 s before firing was 30.2 deg / s, indicating that the oscillation of the firing line increased before firing.
[0195] After completing the above analysis, the system generates technical diagnostic results and movement optimization suggestions for multi-directional skeet training based on the synergy index results. For example, when the intersection of the gun line lags behind the skeet position for a long time in left-facing target training, the system can suggest that the athlete establish the target contact point in advance and enhance the continuity of the body's movement to the left while carrying the gun; when the angular velocity of the gun line fluctuates greatly in right-facing target training, the system can suggest that the athlete reduce the arm's independent gun pull and improve the coordination of the hip and torso in carrying the gun; when the vertical deviation of the intersection of the gun line is large in center-facing target training, the system can suggest that the athlete adjust the gun holding height and the muzzle rise; when the firing time fluctuates greatly in the same position, the system can suggest that the athlete establish a stable target call-target contact-target follow-fire rhythm chain. <0 and <0 indicates insufficient lead time; >0 indicates the muzzle is too high; a lower S indicates insufficient directional matching; a smaller MoS indicates... An increase indicates a decrease in support and gun trajectory stability. Based on the system's analysis of the firing results from the multi-directional flying saucer target to the left: the reaction time for the third attempt was 0.219 s, and the firing time was 0.905 s, which is within the exemplary range. =-0.006s and =-0.126 m, indicating insufficient lead time; =+0.041m indicates the muzzle is too high; S=0.74 indicates insufficient matching of the carrying direction; MoS=0.057m indicates insufficient lower limb support reserve; =0.66 s -1 Furthermore, the standard deviation of the angular velocity before firing is 30.2 deg / s, indicating a decrease in the stability of the firing line.
[0196] System diagnosis: The main problems of the athlete in this left-facing single-target shooting exercise were insufficient lead control, insufficient matching between the direction of gun movement and the direction of the flying saucer, decreased lower limb support stability during body rotation, and insufficient dynamic stability of the gun line before firing.
[0197] Action optimization suggestions: 1. Establish target contact points in advance; 2. Enhance body rotation while carrying the gun; 3. Keep the gun line aligned with the direction of the flying saucer's flight; 4. Improve the stability of the gun line before firing.
[0198] Through the above methods, this disclosure can quantify the human-gun-target coordination in multi-directional flying saucer single-target shooting and generate technical diagnostics and motion optimization suggestions.
[0199] Finally, the report output module generates a multi-directional skeet training report. The system can also compare the current training results with the athlete's historical data, average data for the same position, or reference data from outstanding athletes to form individualized training priorities.
[0200] Through this embodiment, the present disclosure enables the synchronous acquisition and unified analysis of athlete movements, shotgun trajectory, and clay target trajectory in a real multi-directional clay target shooting training scenario. It also generates targeted technical diagnoses and training optimization suggestions based on different stances and target launch directions. Therefore, the same human-gun-target coordination analysis system is applicable not only to two-way clay target shooting but also to multi-directional clay target shooting training, enabling data acquisition, coordinated analysis, and training feedback for different clay target events within the same training venue.
[0201] The improvements disclosed herein are not pieced together, nor do they increase the complexity of the overall structure or the complexity of use, nor do they significantly increase the cost. At the same time, the device is easy to install and disassemble, and is not easily damaged, which objectively improves the reliability of the disclosure. It meets the actual structural and application requirements of the field. For those skilled in the art, the improvements to the device are clear and complete, and are easy to implement according to the contents of the instruction manual, making it highly popular and adaptable.
[0202] While specific embodiments of this disclosure have been described in detail by way of example, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of this disclosure. Those skilled in the art should understand that modifications can be made to the above embodiments without departing from the scope and spirit of this disclosure. The scope of this disclosure is defined by the appended claims.
Claims
1. A method for analyzing the coordination between the shooter and target in skeet shooting, characterized in that, include: Multi-source data acquisition is carried out, including video data acquisition through a camera array deployed in the flying saucer training area, raw IMU sampling data acquisition through an IMU module deployed below the muzzle or barrel of the gun, and data acquisition of the flying saucer exit time through a flying saucer exit time acquisition module. Before data acquisition, the process also includes: calibrating the intrinsic and extrinsic parameters of the cameras in the camera array, establishing the transformation relationship between the camera coordinate system and the unified spatial coordinate system of the target range; and performing time synchronization processing on the UFO trajectory data and the raw video data obtained by the camera array. The multi-source data acquired through data acquisition is processed to obtain structured basic data. The multi-source data includes the original video data, image frame sequences, camera calibration parameters, IMU raw sampling data, and UFO launch time data. The structured basic data includes: human 3D posture data, firearm 3D feature point data, gun line vector data, UFO 3D trajectory data, firing time data, and data quality identifiers. The processing includes: Based on a markerless motion capture method, human keypoints are identified in video images of the athlete from at least two camera perspectives to obtain the two-dimensional keypoint coordinates of the athlete in different video images. After data cleaning of the two-dimensional keypoint coordinates, three-dimensional reconstruction is performed in combination with the stereo calibration parameters of the camera to obtain the three-dimensional coordinate sequence of the athlete's human keypoints in a unified spatial coordinate system of the shooting range. The human keypoints include at least the left shoulder joint, right shoulder joint, left elbow joint, right elbow joint, left wrist joint, right wrist joint, left hip joint, right hip joint, left knee joint, right knee joint, left ankle joint, right ankle joint, seventh cervical vertebra, and top of the head. Two-dimensional coordinates of the gun head and gun tail in video images from at least two camera perspectives are acquired. Combined with the stereo calibration parameters of the cameras, three-dimensional reconstruction is performed to obtain the three-dimensional coordinates of the gun head and gun tail. The direction from the three-dimensional coordinates of the gun tail to the three-dimensional coordinates of the gun head is taken as the gun line direction to obtain the gun line vector. Based on the three-dimensional coordinates of the gun head, the three-dimensional coordinates of the gun tail, and the gun line vector, the kinematic data of the gun is calculated. The kinematic data of the gun includes at least one of the following: gun trajectory, change of gun line direction, angular velocity of gun line, speed of gun line movement, gun movement path, change of muzzle elevation, and the spatial relationship between the gun and the target at the moment of firing. Select the frame interval to be processed in the video image, obtain the UFO trajectory from at least two camera perspectives, generate the UFO's two-dimensional coordinates, and combine the stereo calibration parameters of the camera to perform three-dimensional reconstruction to obtain the three-dimensional coordinate sequence of the UFO in the unified spatial coordinate system of the target range. The structured basic data is spatiotemporally matched to obtain a multi-source fusion dataset of the shooter, rifle, and target under the same shooting process. Based on the dataset, a human-gun-target synergy analysis index is calculated. The athlete's shooting technique is diagnosed based on the index results, generating a diagnostic result. The synergy analysis includes athlete movement process analysis, rifle movement process analysis, skeet target movement process analysis, rifle-target synergy analysis, and overall human-gun-target synergy analysis. The spatiotemporal matching includes: The structured basic data is time-matched and converted into a unified timeline with a unified time point as the zero point. Two-dimensional pixel coordinates of the same target point are extracted from at least two camera viewpoints. Based on the established transformation relationship between the camera pixel coordinate systems, the camera coordinate system and the unified spatial coordinate system of the target range, the two-dimensional pixel coordinates under different viewpoints are matched. The three-dimensional coordinates of the target point under the unified spatial coordinate system of the target range are obtained by projection matrix intersection or triangulation. The target point includes the human joint point, the gun nose point, the gun tail point and the center of mass of the flying saucer. Using a unified time series as an index, a multi-source fusion dataset of human, gun, and target under the same shooting process is formed. The dataset includes at least: timestamp, relative target release event, three-dimensional coordinates of each joint of the human body, three-dimensional coordinates of the gun head, three-dimensional coordinates of the gun tail, gun line vector, IMU three-axis acceleration, IMU three-axis angular velocity, UFO three-dimensional coordinates, and firing time identifier. Based on the diagnostic results, a technical action optimization plan is generated.
2. The method according to claim 1, characterized in that, The process of processing the multi-source data obtained from the data acquisition to obtain structured basic data also includes: The kinematic parameters of the athlete during shooting are calculated based on the three-dimensional coordinates of the athlete's personal key points in the unified spatial coordinate system of the shooting range. The kinematic parameters include at least one of joint angles, trunk posture, center of gravity position, displacement, velocity, and angular velocity.
3. The method according to claim 1, characterized in that, The calculated human-gun-target synergy analysis indicators include: The calculated human-gun-target coordination analysis indicators include: gun-target spatial deviation, gun-target temporal deviation, deviation direction, reaction time index, body stability index, gun-holding stability index, shooting rhythm index, and posture control index. The calculation method includes: Construct the gun line vector The gun line vector The direction from the breech to the muzzle; for the aforementioned gun line vector Perform normalization to obtain the unit gun line vector. Guns in The gun line at that moment was ;in, Let t be the three-dimensional coordinates of the gun tail. For linear parameters; Based on the UFO's three-dimensional coordinate point set data, the UFO's trajectory plane is obtained through fitting. ; in the plane of the UFO trajectory Within, the relative time after the flying saucer leaves the target. The trajectory points below are ,in, For UFOs in The three-dimensional spatial position at a given moment. The relative time is the time after the UFO leaves the target, with the physical moment of its launch as the zero point. For UFOs in The coordinate value along the X-axis in the unified spatial coordinate system of the target range at any time. For UFOs in The coordinate value along the Y-axis in the unified spatial coordinate system of the target range at any time. For UFOs in The coordinate value along the Z-axis in the unified spatial coordinate system of the target range at any time; Calculate the straight line of the gun. and the plane of the UFO trajectory intersection ; Calculate the spatial deviation between the gun and the target: the spatial deviation between the gun and the target at the moment of firing. ,in, The point where the firing line intersects the plane of the UFO's trajectory at the moment of firing; The three-dimensional coordinates of the flying saucer at the moment of firing. For the projectile's flight time ; Calculating the target time deviation includes: finding the target trajectory parameter curve with... The UFO trajectory point with the smallest distance ,in, The moment when the distance between the UFO trajectory point and the intersection of the gun line is minimum; calculate the target-gun time deviation. ; The deviation vector E is decomposed to obtain deviation components in the vertical, tangential, and normal directions, where... , or ; Calculate reaction time index , ,in, The moment the UFO physically launches from the target. The moment of activation of the firearm or the moment of activation of the human body; Calculating the body's dynamic stability margin MoS includes: calculating the extrapolated center of mass. , ,in, For the athlete's overall well-being, For speed, , It is the acceleration due to gravity. The height is the center of mass; construct the support base BoS based on the bipedal support area, and calculate... The minimum distance to the BoS boundary is used to obtain the dynamic stability margin MoS; Calculate gun stability index , The maximum Lyapunov index, ,in, The initial distance between adjacent trajectories. For the process Trajectory distance after each sampling interval; For time intervals; Statistically record the time to shoot at one target, the time to shoot at two targets, and the total shooting time for at least two shooting sessions; calculate the mean, standard deviation, and coefficient of variation for the time to shoot at one target, the time to shoot at two targets, and the total shooting time. The joint angles and the relative torsional angle between the shoulder and pelvis are calculated based on human 3D posture data. The method for calculating the relative torsional angle between the shoulder and pelvis is as follows: construct the vector connecting the two shoulders and the vector connecting the pelvis; project them onto the horizontal plane and then calculate the relative angle. Calculate angular velocity .
4. The method according to claim 1, characterized in that, The frame interval is the frame segment between when the UFO is thrown out and when the UFO breaks or flies out of the effective analysis area.
5. The method according to claim 1, characterized in that, The step of performing time matching on the structured basic data, converting the structured basic data into a unified timeline with a unified time point as zero, includes: During the data acquisition phase, a unified time reference is established through network time synchronization, hardware synchronization triggering, or unified timestamp recording. Each frame of the original video data is bound to its corresponding acquisition time, forming a correspondence. The sampling time corresponding to each sampling point in the data acquired by the IMU module is recorded. The physical launch time of the flying saucer is recorded and defined as the zero point of the firing process. The relative time of the nth frame of any camera with respect to the physical launch time of the flying saucer is: ,in, For the first The relative time of frame video capture compared to the moment the UFO physically leaves the target. For the first The absolute time of frame video capture. The physical launch time of the flying saucer; IMU number The relative time of each sampling point with respect to the moment when the UFO physically leaves the target is: ,in, For IMU The relative time of each sampling point with respect to the moment the UFO physically launches from the target. For IMU The absolute time of each sampling point; For the original video data with different sampling frequencies, timestamp proximity matching, linear interpolation, or spline interpolation are used to align the data from different data sources to the same time series.
6. The method according to claim 1, characterized in that, The process of obtaining the three-dimensional coordinates of the target point in the unified spatial coordinate system of the target range further includes: Obtain the three-dimensional coordinates of the target point in the local three-dimensional coordinate system; Based on common control points or known target range coordinates, calculate the rigid body transformation relationship from each local coordinate system to the unified spatial coordinate system of the target range; obtain the three-dimensional coordinates of the target point in the unified spatial coordinate system of the target range, and the rigid body transformation relationship is as follows: ,in, For the three-dimensional coordinates in the unified spatial coordinate system of the target range, Three-dimensional coordinates in a local coordinate system. Let be a rotation matrix. It is a translation vector; The rotation matrix and the translation vector The confirmation method includes: selecting several common control points or known target range coordinate points whose coordinates can be obtained in both the local coordinate system and the unified spatial coordinate system of the target range, and solving the rotation matrix by rigid body registration or least squares method. and the translation vector The rotation matrix is verified by reprojection error, control point residual, or 3D reconstruction error. When the error meets a preset threshold, the rotation matrix is confirmed. and the translation vector It can be used to perform coordinate transformation calculations in the rigid body transformation relationship.
7. The method according to claim 1, characterized in that, The step of generating a technical action optimization plan based on the diagnostic results includes: Based on the synergy index, identify the technical problems of the athlete during the shooting process, classify the technical problems, and generate optimization suggestions according to the type of technical problem.
8. A system for analyzing the coordination between humans, guns, and targets in skeet shooting, characterized in that: include: The system includes a camera array module, an IMU module, a UFO launch time acquisition module, a data acquisition module, a data processing module, a database module, a collaborative analysis module, and a report output module. The camera array module, coupled to the acquisition module, is used to acquire multi-view video data of athlete movements, gun movements, and the flight process of the skeet shooting training area. The camera array module is deployed within the skeet shooting training area and includes no fewer than four cameras. At least two cameras capture the athlete's complete shooting motion at each target position, and at least two cameras cover the entire skeet target flight area. The acquisition module controls the start / stop, exposure, frame rate, and data storage of the cameras within the camera array module. The acquisition module also transmits the multi-view video data to the acquisition module, which receives, stores, forwards, or establishes a data index, and then provides it to the data processing module. The IMU module, coupled to the acquisition module, is installed below the muzzle or barrel of the gun and is used to acquire the acceleration, angular velocity and vibration impact signals generated at the moment of firing, and to identify the firing time of the first target and the firing time of the second target. The UFO launch time acquisition module is coupled to the acquisition module and is deployed at the launch port or covering the launch port area to acquire the physical launch time of the UFO. The acquisition module is coupled to the data processing module and is used to uniformly control and record the video acquisition tasks, athlete information, training items, target information, target type, hit results and acquisition status during the skeet shooting training process. The data processing module is coupled to the acquisition module, the database module, and the collaborative analysis module. The data processing module processes the raw video data, camera calibration parameters, training task information, and firing time event data transmitted by the acquisition module and the IMU module, generating structured basic data. This structured basic data includes human 3D posture data, firearm 3D feature point data, gun line vector data, UFO 3D trajectory data, firing time data, and corresponding data quality identifiers. The data processing module also transmits the structured basic data to the database module for storage and to the collaborative analysis module for calculating and analyzing human-firearm-target collaborative performance indicators. The database module is coupled to the acquisition module, the data processing module, the collaborative analysis module, and the report output module, respectively, and is used to receive data generated by the acquisition module, the data processing module, the collaborative analysis module, and the report output module, and to store, manage, associate, and retrieve the data. The coordination analysis module is coupled to the data processing module, calculates the human-gun-target coordination analysis index based on the structured basic data, and outputs the technical diagnostic results; wherein, the coordination analysis includes athlete movement process analysis, gun movement process analysis, skeet target movement process analysis, gun-target coordination analysis, and overall human-gun-target coordination analysis; The report output module is coupled to the collaborative analysis module and the database module, respectively, and is used to summarize, organize and display the analysis results of the collaborative analysis module and the stored data of the database.
9. The human-gun-target coordination analysis system for skeet shooting sport according to claim 8, characterized in that, The acquisition module includes an acquisition main unit and a handheld terminal. The acquisition main unit is deployed in a workstation in the skeet shooting training area and is coupled to the camera array module. It is used to control the camera's start and stop, synchronization triggering, data reception, and local storage. The handheld terminal is deployed on-site at the skeet shooting training ground and coupled to the main data acquisition unit. It is used for training information input, data acquisition task triggering, hit result recording, and data acquisition status viewing.
Citation Information
Patent Citations
Digital immersive flying saucer shooting analogue simulation system
CN120947423A
