An indoor dynamic shooting training control system and method based on a projected virtual target

Abstract

This invention provides an indoor dynamic shooting training control system and method based on a projected virtual target, belonging to the field of indoor dynamic shooting training control technology. To support shooters in dynamic shooting while moving and to accurately calculate the three-dimensional spatial hit position, a main control server generates a three-dimensional virtual scene and a dynamic virtual target with three-dimensional spatial coordinates. The target is then projected and displayed on a physical display medium by an image presentation unit. During the shooter's movement, a spatial positioning module tracks and acquires the shooter's current physical three-dimensional spatial coordinates in real time. The shooter fires at the physical display medium using a laser shooting simulator, and an image acquisition module instantly captures the image containing the laser spot and extracts the two-dimensional pixel coordinates of the spot. The main control server performs a reverse mapping calculation between the shooter's physical three-dimensional spatial coordinates and the two-dimensional pixel coordinates of the spot to generate a three-dimensional shooting ray. This invention is applied to indoor dynamic shooting training.

Description

Technical Field

[0001] This invention provides an indoor dynamic shooting training control system and method based on a projected virtual target, belonging to the field of indoor dynamic shooting training control technology. Background Technology

[0002] Shooting training is a core subject in military tactics, police combat, and shooting sports. With the development of computer vision and virtual reality technologies, indoor simulated shooting systems based on projection screens and laser guns have been widely used. However, existing indoor laser shooting systems generally suffer from the following technical defects:

[0003] Unable to support shooting while moving (limited to fixed point): The current photoelectric calibration algorithm establishes a two-dimensional mapping matrix based on a fixed geometric relationship between the "shooting point and the projection screen". Once the shooter shoots while moving (shooting while moving), the change in the shooter's perspective and distance will cause severe perspective distortion in the projected image, causing the original two-dimensional coordinate mapping algorithm to completely fail and the hit determination error to be extremely large.

[0004] Lack of realistic three-dimensional spatial depth: Traditional projected targets only move within a two-dimensional plane coordinate system, which cannot achieve the tactical depth simulation of "approaching-moving away" in three-dimensional space. Even if the distance of the target is simulated by zooming in and out of the screen, the system cannot perceive the relative three-dimensional distance between the shooter and the screen, and cannot make accurate hit judgments based on physical laws. This will make it difficult to meet the needs of modern high-mobility combat tactical training.

[0005] Therefore, there is an urgent need to develop a control scheme that can support shooters in dynamic shooting while on the move and can accurately reverse calculate the three-dimensional spatial hit position. Summary of the Invention

[0006] To address the technical problems existing in the background art, the present invention adopts the following technical solution: providing an indoor dynamic shooting training control system based on a projected virtual target, comprising:

[0007] The main control server is used to run 3D virtual scenes and generate dynamic virtual targets with 3D spatial coordinates;

[0008] An image presentation unit, connected to the main control server, is used to project the three-dimensional virtual scene and dynamic virtual target from perspective onto a physical display medium.

[0009] The spatial positioning module is used to acquire the three-dimensional spatial coordinates of the moving shooter in the real physical space in real time and transmit the coordinate data to the main control server;

[0010] A laser shooting simulator, held by a shooter, is used to fire a laser beam at the physical display medium when aiming and firing;

[0011] The image acquisition module is used to acquire images of laser spots on the physical display medium in real time, and extract the two-dimensional pixel coordinates of the laser spots and transmit them to the main control server.

[0012] The main control server combines the shooter's three-dimensional spatial coordinates, the projection system's intrinsic parameter matrix, and the laser spot's two-dimensional pixel coordinates to construct a three-dimensional shooting space ray, calculates the spatial intersection of the three-dimensional shooting space ray and the dynamic virtual target in the three-dimensional coordinate system, and outputs the shooting hit result.

[0013] The image presentation unit is a planar projection device, and the physical display medium is a projection screen or a wall.

[0014] The main control server adjusts the perspective distortion and scaling ratio of the dynamic virtual target in the projected image in real time according to the shooter's three-dimensional spatial coordinates, so as to visually simulate the depth displacement of the target in three-dimensional space on a two-dimensional physical display medium.

[0015] The spatial positioning module uses an ultra-wideband positioning base station cluster or an infrared motion capture camera group based on multi-view vision to perform three-dimensional coordinate tracking of the shooter at a continuous high-frequency refresh rate.

[0016] The physical display medium includes multiple layers of high-transmittance holographic gauze arranged in a sequential array along a direction perpendicular to the shooting direction. The main control server switches the focal plane of the projection device in real time according to the depth coordinates of the dynamic virtual target, so that the target is physically imaged on the holographic gauze at different depths.

[0017] The main control server is embedded with a trained image segmentation neural network model, which is used to extract the sub-pixel coordinates of the laser spot against a complex dynamic projection background.

[0018] A method for controlling an indoor dynamic shooting training control system based on a projected virtual target includes the following control steps:

[0019] Step S1: The main control server generates a three-dimensional virtual scene and a dynamic virtual target with three-dimensional spatial coordinates, and displays it on the physical display medium after projection transformation by the image rendering unit;

[0020] Step S2: During the archer's movement, the spatial positioning module tracks and obtains the archer's current physical three-dimensional spatial coordinates in real time;

[0021] Step S3: The shooter uses the laser shooting simulator to fire at the physical display medium. The image acquisition module instantly captures the image containing the laser spot and extracts the two-dimensional pixel coordinates of the spot.

[0022] Step S4: The main control server performs reverse mapping calculation of the shooter's physical three-dimensional spatial coordinates and the two-dimensional pixel coordinates of the light spot based on the camera's intrinsic and extrinsic parameter matrix, generating a three-dimensional shooting ray with a definite spatial direction.

[0023] Step S5: In the virtual three-dimensional coordinate system, the main control server calculates whether the three-dimensional shooting ray and the three-dimensional collision bounding box of the dynamic virtual target intersect geometrically. If they intersect, it is determined that a hit has occurred, and an evaluation result is generated based on the hit location.

[0024] The reverse mapping solution method used in step S4 specifically includes:

[0025] Using the shooter's current three-dimensional spatial coordinates as the starting point of the ray, the direction vector of the ray in three-dimensional space is calculated using the spatial pose matrix of the physical display medium plane and the two-dimensional pixel coordinates of the light spot, thereby establishing the spatial analytical equation of the three-dimensional shooting ray.

[0026] In step S5, a comprehensive tactical action score is generated based on the number of hits on the target, the shooter's movement speed trajectory obtained in step S2, and the firing time parameters.

[0027] The advantages of this invention compared to existing technologies are as follows: This invention provides an indoor dynamic shooting training control system and method based on a projected virtual target. It acquires the three-dimensional coordinates of the shooter in real time through a spatial positioning module, combines this with the two-dimensional pixel coordinates of the laser spot extracted by an image acquisition module, and uses a main control server to perform multi-coordinate system inverse mapping calculation to generate a three-dimensional shooting ray. This ray is then intersected with the three-dimensional collision box of the virtual target to determine a hit. This invention breaks through the limitation of traditional indoor shooting requiring a fixed standing position. By employing real-time spatial positioning and three-dimensional ray inverse calculation techniques, it fundamentally eliminates perspective distortion errors and coordinate calculation offsets caused by shooting while moving. It not only achieves dynamic adaptation of the viewing angle but also restores the true ballistic depth logic through three-dimensional bounding box intersection detection, significantly improving the practicality and flexibility of tactical training. Attached Figure Description

[0028] The present invention will be further described below with reference to the accompanying drawings:

[0029] Figure 1 This is a flowchart illustrating the steps of the indoor dynamic shooting training control method based on a projected virtual target according to the present invention. Detailed Implementation

[0030] like Figure 1As shown, in order to solve the problem that indoor simulated shooting in the prior art cannot support high-precision dynamic shooting while moving, the present invention provides an indoor dynamic shooting training and control system based on a projected virtual target, which mainly consists of a physical display medium, an image presentation unit, a spatial positioning module, an image acquisition module, a laser shooting simulator, and a main control server.

[0031] In practical deployment, an anti-glare high-gain projection screen is used as the physical display medium to establish a world coordinate system. Let the plane where the curtain is located be The reference plane; the image presentation unit uses an ultra-short-throw laser projector with a brightness of no less than 6000 ANSI lumens; the spatial positioning module uses an indoor positioning array based on ultra-wideband (UWB) technology or an infrared multi-view vision motion capture system, with positioning base stations deployed at the top edge and four corners of the training room; the shooter wears a positioning tag, and the system outputs the shooter's three-dimensional coordinates at a sampling rate of no less than 100Hz. To eliminate coordinate jitter during movement, the main control server uses Kalman filtering to smooth the raw positioning data, ensuring the spatial stability of the ray starting point; the laser shooting simulator emits an invisible infrared laser beam, while the image acquisition module uses a high frame rate global shutter industrial infrared camera and is equipped with a narrow bandpass filter to shield visible light interference.

[0032] Based on the above system hardware architecture, the core control method adopted in this invention lies in "deriving three-dimensional rays from two-dimensional light spots and performing three-dimensional spatial collision detection". The specific execution and calculation derivation process is as follows:

[0033] Step S1: The main control server generates a 3D virtual scene and a dynamic virtual target with 3D spatial coordinates. The image rendering unit projects and transforms these elements, displaying them on the physical display medium. During the perspective calibration process of the virtual scene and target projection, the main control server runs a 3D physics engine to generate data with depth information. The system uses a virtual dynamic target and pre-calibrates the infrared camera's intrinsic parameter matrix using a camera calibration algorithm. And the extrinsic parameters (rotation matrix) of the camera coordinate system relative to the world coordinate system. With translation vector ).

[0034] Step S2: To achieve high-precision spot extraction against complex backgrounds, after the shooter fires while moving, the spatial positioning module tracks and acquires the shooter's current physical three-dimensional spatial coordinates in real time, and the infrared camera captures the target area image. The main control server uses a deep convolutional neural network based on image segmentation (such as the U-Net model architecture), takes the current frame image as input, and outputs a high-confidence mask containing only the laser spot. Then, the sub-pixel-level two-dimensional coordinates of the mask area are calculated using the gray-scale centroid method. This significantly improves the extraction accuracy of the light spot center.

[0035] Step S3: The shooter fires at the physical display medium using a laser shooting simulator. The image acquisition module instantly captures the image containing the laser spot, extracts the two-dimensional pixel coordinates of the spot, and performs a reverse spatial mapping from two-dimensional to three-dimensional, including:

[0036] Based on the pinhole camera model, the projection transformation relationship from a physical point to the pixel plane is as follows:

[0037] ;

[0038] in, The depth scaling factor is in the camera coordinate system; since it is known that the laser must fall on the physical plane of the projection screen (i.e., =0), combined with the camera's intrinsic and extrinsic parameter matrices, the main control server will determine the two-dimensional pixel coordinates of the light spot. Inverse mapping is used to calculate the three-dimensional world coordinates of the actual laser strike on the physical screen. .

[0039] Step S4: Based on the camera's intrinsic and extrinsic parameter matrices, the main control server performs a reverse mapping calculation between the shooter's physical 3D spatial coordinates and the light spot's 2D pixel coordinates to generate a 3D shooting ray with a defined spatial direction. The method for constructing the 3D shooting ray's spatial equation is as follows:

[0040] The smoothed three-dimensional coordinates of the shooter at the moment of firing are obtained through the spatial positioning module. ,by As the starting point of the ray, with Define the 3D spatial shooting ray as the point of penetration. Its parametric equation can be expressed as:

[0041] ;

[0042] At this point, regardless of how the shooter moves or how their position deviates, the ray represents the actual physical trajectory of the ballistics.

[0043] Step S5: In the virtual 3D coordinate system, the main control server calculates whether the 3D shooting ray and the 3D collision bounding box of the dynamic virtual target geometrically intersect. If they intersect, a hit is determined, and an evaluation result (score) is generated based on the hit location. Specifically, this includes:

[0044] In the 3D physics engine, a dynamic target is assigned an Oriented Bounding Box (OBB), and the main control server executes ray casting in real time. The algorithm for intersecting the 3D space of the target OBB, if the ray parameters If a real solution exists and the intersection point is located on or inside the OBB surface, it is considered a hit. The three-dimensional Euclidean distance between the intersection point and the bullseye is further calculated to determine the hit score. Simultaneously, the system retrieves the shooter's high-frequency movement velocity vector and trajectory fluctuation variance before firing. If the shot is fired and hit while moving at high speed, the system will assign a higher tactical difficulty weighted score.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An indoor dynamic shooting training control system based on a projected virtual target, characterized in that: include: The main control server is used to run 3D virtual scenes and generate dynamic virtual targets with 3D spatial coordinates; An image presentation unit, connected to the main control server, is used to project the three-dimensional virtual scene and dynamic virtual target from perspective onto a physical display medium. The spatial positioning module is used to acquire the three-dimensional spatial coordinates of the moving shooter in the real physical space in real time and transmit the coordinate data to the main control server; A laser shooting simulator, held by a shooter, is used to fire a laser beam at the physical display medium when aiming and firing; The image acquisition module is used to acquire images of laser spots on the physical display medium in real time, and extract the two-dimensional pixel coordinates of the laser spots and transmit them to the main control server. The main control server combines the shooter's three-dimensional spatial coordinates, the projection system's intrinsic parameter matrix, and the laser spot's two-dimensional pixel coordinates to construct a three-dimensional shooting space ray, calculates the spatial intersection of the three-dimensional shooting space ray and the dynamic virtual target in the three-dimensional coordinate system, and outputs the shooting hit result.

2. The indoor dynamic shooting training control system based on a projected virtual target according to claim 1, characterized in that: The image presentation unit is a planar projection device, and the physical display medium is a projection screen or a wall. The main control server adjusts the perspective distortion and scaling ratio of the dynamic virtual target in the projected image in real time according to the shooter's three-dimensional spatial coordinates, so as to visually simulate the depth displacement of the target in three-dimensional space on a two-dimensional physical display medium.

3. The indoor dynamic shooting training control system based on a projected virtual target according to claim 1, characterized in that: The spatial positioning module uses an ultra-wideband positioning base station cluster or an infrared motion capture camera group based on multi-view vision to perform three-dimensional coordinate tracking of the shooter at a continuous high-frequency refresh rate.

4. The indoor dynamic shooting training control system based on a projected virtual target according to claim 1, characterized in that: The physical display medium includes multiple layers of high-transmittance holographic gauze arranged in a sequential array along a direction perpendicular to the shooting direction. The main control server switches the focal plane of the projection device in real time according to the depth coordinates of the dynamic virtual target, so that the target is physically imaged on the holographic gauze at different depths.

5. The indoor dynamic shooting training control system based on a projected virtual target according to claim 1, characterized in that: The main control server is embedded with a trained image segmentation neural network model, which is used to extract the sub-pixel coordinates of the laser spot against a complex dynamic projection background.

6. The method for controlling an indoor dynamic shooting training control system based on a projected virtual target according to claim 1, characterized in that: The control steps include the following: Step S1: The main control server generates a three-dimensional virtual scene and a dynamic virtual target with three-dimensional spatial coordinates, and displays it on the physical display medium after projection transformation by the image rendering unit; Step S2: During the archer's movement, the spatial positioning module tracks and obtains the archer's current physical three-dimensional spatial coordinates in real time; Step S3: The shooter uses the laser shooting simulator to fire at the physical display medium. The image acquisition module instantly captures the image containing the laser spot and extracts the two-dimensional pixel coordinates of the spot. Step S4: The main control server performs reverse mapping calculation of the shooter's physical three-dimensional spatial coordinates and the two-dimensional pixel coordinates of the light spot based on the camera's intrinsic and extrinsic parameter matrix, generating a three-dimensional shooting ray with a definite spatial direction. Step S5: In the virtual three-dimensional coordinate system, the main control server calculates whether the three-dimensional shooting ray and the three-dimensional collision bounding box of the dynamic virtual target intersect geometrically. If they intersect, it is determined that a hit has occurred, and an evaluation result is generated based on the hit location.

7. The method for controlling an indoor dynamic shooting training control system based on a projected virtual target according to claim 6, characterized in that: The reverse mapping solution method used in step S4 specifically includes: Using the shooter's current three-dimensional spatial coordinates as the starting point of the ray, the direction vector of the ray in three-dimensional space is calculated using the spatial pose matrix of the physical display medium plane and the two-dimensional pixel coordinates of the light spot, thereby establishing the spatial analytical equation of the three-dimensional shooting ray.

8. The method for controlling an indoor dynamic shooting training control system based on a projected virtual target according to claim 6, characterized in that: In step S5, a comprehensive tactical action score is generated based on the number of hits on the target, the shooter's movement speed trajectory obtained in step S2, and the firing time parameters.

Images