Method and system for assisting calibration and aiming training of sights based on tag recognition

Abstract

The present invention discloses a method and system for assisting calibration and aiming training of a scope based on label recognition. A telephoto video acquisition device is fixedly installed on a live-fire shooting gun, a QR code label is arranged in a blank area of ​​a target, a real-time picture of the aiming area is captured by the video acquisition device, and the real-time image is transmitted to an image recognition device, and the label and position coordinates in the picture are marked by image recognition, and the image is corrected by perspective transformation according to the recognized position parameters; the image that has been aimed and collected before shooting is intercepted and saved as a base map, and the position of the bullet hole is marked on the base map after shooting, so that the position of the actual shooting crosshair in the collected image at the current shooting distance can be obtained, and the position is compared with the target position recognized by the image in real time, so that assisting calibration of the scope and real-time collection of the aiming trajectory can be realized to assist aiming training.

Description

Technical Field

[0001] The present invention relates to the fields of image recognition, image processing and intelligent analysis, and in particular to a method and system for assisting calibration and aiming training of a sight based on label recognition. Background Art

[0002] Before using a live-fire gun, it is necessary to carry out aiming calibration. The traditional calibration method relies on manual experience. Through multiple shootings, the position difference between the bullet hole and the aiming point is checked to adjust the sight. When the aiming point and the shooting bullet hole are relatively close and stable, it is considered that the sight calibration is completed. This traditional method has many disadvantages. The first is that there is no basis for adjusting the parameters. It depends entirely on the experience of the personnel, and often there is a situation of over-adjustment and repeated adjustment. The second is that the cost is high and the efficiency is low. It is necessary to observe the bullet hole through multiple live-fire shootings. When the target is far away, it is very inconvenient to observe the bullet hole. Multiple live-fire shootings will also cause ammunition waste. In this context, the present invention proposes a method for auxiliary calibration and aiming training of the sight based on tag recognition. Through image recognition technology, only one live-fire shooting is needed to complete the calibration of the sight, which not only greatly reduces the cost of calibration, but also greatly improves the overall efficiency and calibration accuracy through image recognition technology. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for assisting calibration and aiming training of a sight based on tag recognition in view of the deficiencies of the technical solutions in existing application scenarios.

[0004] The objective of the present invention is achieved through the following technical solutions:

[0005] In a first aspect, the present invention provides a method for assisting calibration and aiming training of a sight based on tag recognition, comprising:

[0006] A telephoto video acquisition device installed on the gun collects a real-time image of a target with a QR code label and transmits it to an image recognition device; the image recognition device recognizes the received image in real time and assists in determining the coordinate position of the target in the image based on the QR code label;

[0007] The captured image before the trigger is pulled for the first shot is saved as a base map, and the real position of the target impact point is calibrated on the base map after the first shot, so as to obtain the pixel position of the target impact point in the video captured image at the current shooting distance, that is, the target impact point position coordinates, and the current shooting distance is paired with the corresponding target impact point position coordinates and stored in the image recognition device;

[0008] During the scope calibration or aiming training, a real-time picture of a target with a QR code label is captured by a telephoto video acquisition device, and the center position of the target in the real-time picture is marked by image recognition. A target impact point crosshair is statically drawn at the target impact point position coordinates corresponding to the current shooting distance on the display screen of the image recognition device, and the distance difference between the target impact point and the target center position is displayed in real time on the display screen. The shooter adjusts the gun posture until the target impact point crosshair on the display screen is observed to be aligned with the target center, and then keeps the gun posture still and adjusts the scope crosshair until the scope crosshair is aligned with the target center, thereby completing the scope calibration;

[0009] After the sight is calibrated, during the aiming training process, the coordinates of the target center point are used as the initial position of the new coordinate axis, and the relative coordinates of the impact point in the image relative to the target center position are recorded as the coordinate data under the new coordinate axis, so as to obtain the aiming trajectory of the aiming point around the target center point and realize aiming training assistance.

[0010] Furthermore, the telephoto video acquisition device is a fixed-focus device or a zoom device, which is stably fixed on the gun. The actual coordinate position of the impact point in the acquired image is related to the current shooting distance, the installation angle of the acquisition device and the focal length parameters. When the acquisition device is reinstalled, adjusted and tightened, and the focal length is adjusted, the previously collected and recorded data is invalid and needs to be re-collected and calibrated.

[0011] Furthermore, for different shooting distances, the pixel position of the target impact point in the video capture image is different. When the telephoto video capture device is not disassembled and reassembled and the focal length is not adjusted, the pixel position of the target impact point in the captured video image is considered to be fixed at the same shooting distance. Therefore, the pixel positions of the target impact point collected at multiple different shooting distances are stored in the image recognition device.

[0012] Furthermore, the image recognition device performs real-time recognition on the received image and assists in determining the coordinate position of the target in the image based on the QR code label, specifically:

[0013] The coordinate position of the QR code label in the captured image is obtained through image recognition. Combined with the coordinates and size of the QR code label on the real target paper, the real-time captured image is transformed into a two-dimensional frontal plane image. The converted image is the target paper plane image. The coordinate position of the target center point in the image can be determined in the converted image if the coordinates of the target ring center in the target paper and the radius of the target ring are known.

[0014] Furthermore, the QR code label may be an Aruco label, but is not limited thereto.

[0015] Furthermore, there is at least one QR code label, which is set in a blank area of ​​the target paper, and its size needs to ensure that it is recognizable to the human eye in the video screen; when a label is hit and damaged during the shooting process and cannot be recognized, as long as there is another recognizable label, the subsequent perspective transformation can still be completed.

[0016] Furthermore, the image recognition device performs real-time recognition of the received image, identifies the ID information of the QR code tag and the pixel coordinate values ​​of the four corner points of the current tag in the current video screen, and numbers the coordinates of the four corner points in a clockwise order of upper left, upper right, lower right, and lower left.

[0017] Furthermore, the perspective transformation of the real-time collected image is specifically as follows: when there are four tags in the picture, one corner point coordinate of each tag is selected as the parameter of the perspective transformation; when only one tag is available in the picture, the four corner point coordinates of the current tag are selected as the parameters of the perspective transformation; when there are two tags in the picture, each tag selects two corner point coordinates as the parameters of the perspective transformation; when there are three tags in the picture, one of the tags selects two corner point coordinates as the parameters of the perspective transformation, and the other two tags each select one corner point coordinate as the parameter of the perspective transformation.

[0018] Furthermore, four Aruco labels with different IDs are arranged in the upper left, upper right, lower right and lower left blank areas of the target paper respectively. The image processing Opencv library is used to recognize the Aruco labels in the captured image, and the label ID and the corresponding four corner point coordinates are used as image perspective transformation parameters.

[0019] Furthermore, during the calibration of the scope, a static target impact point crosshairs is displayed on the display screen of the image recognition device. During the shooter's aiming process, the image of the real target will also be displayed on the display screen. The shooter can clearly see the distance difference between the expected target impact point and the actual target center position on the display screen, which can help the shooter to determine whether the scope has been calibrated and avoid confirming the calibration result through multiple live-fire shooting. The shooter adjusts the gun posture and the scope crosshairs so that the scope crosshairs are aimed at the center of the real target, and the static target impact point crosshairs on the display screen also coincides with the center of the real target in the captured image. At this time, the scope calibration is completed.

[0020] Furthermore, after completing the calibration of the sight, the image recognition device can be used to record and display the motion trajectory of the shooter's entire aiming process in real time, helping the shooter to improve the efficiency of aiming training and improve the aiming technique; specifically: during the shooting aiming training process, as the shooter adjusts the gun posture for aiming, the video acquisition device fixed on the gun will also transmit the real-time image of the target captured to the image recognition device. By using the image recognition device to complete the image correction, a planar target image can be obtained. At this time, the relative position of the pixel point position of the target impact point at the shooting distance recorded previously and the identified target center point point is recorded in chronological order to obtain a set of trajectory coordinates, with the target center point as the origin coordinate and the relative position coordinate difference as the coordinate value. The trajectory coordinates are converted into a trajectory image running in the target image, which can depict the movement posture trajectory of the shooter aiming around the target.

[0021] In a second aspect, the present invention provides a system for assisting calibration and aiming training of a sight based on tag recognition, comprising:

[0022] Label recognition and image correction transformation module: Use the telephoto video acquisition device installed on the gun to capture the target paper with the QR code label, and transmit it to the image recognition device, and use image recognition to mark the label and position coordinates in the real-time picture; Combine the coordinates and size of the label on the real target paper, perform perspective transformation on the real-time picture image, and complete the correction of the captured image;

[0023] Target impact point calibration module: save the collected image before pulling the trigger for the first shot as the base map, and calibrate the real position of the target impact point on the base map after the first shot, so as to obtain the pixel position of the target impact point in the video collection image at the current shooting distance, that is, the target impact point position coordinates, and add a static target impact point crosshair mark at the target impact point position coordinates on the display screen of the image recognition device, and store the current shooting distance and the corresponding target impact point position coordinates in the image recognition device after pairing;

[0024] Sight calibration module: The video image transmitted by the video acquisition device and the target impact point crosshair corresponding to the current shooting distance are displayed in real time on the display screen of the image recognition device, and the image perspective transformation and correction are completed by identifying and locating the QR code label in the image, so as to complete the recognition of the target in the image and the calculation of the target center position, and display the distance difference between the target center point and the target impact point in real time; the shooter adjusts the gun posture until the target impact point crosshair on the display screen is observed to be aligned with the target center, and then keeps the gun posture unchanged and adjusts the sight crosshair until the sight crosshair is aligned with the target center, and the sight calibration is completed;

[0025] Aiming training auxiliary analysis module: Real-time analysis of the images transmitted by the video acquisition device, through the recognition and positioning of the QR code label in the image, to complete the image perspective transformation and correction, and use the recognized real target center as the coordinate origin, and the relative coordinates of the target impact point and the target center as the coordinate value, to record and store them in chronological order, and record and draw the entire running trajectory of the aiming process in the target image.

[0026] In a third aspect, the present invention provides a device for performing assisted calibration and aiming training of a sight based on tag recognition, comprising a memory and one or more processors, wherein the memory stores executable code, and when the processor executes the executable code, it is used to implement the method for performing assisted calibration and aiming training of a sight based on tag recognition as described in the first aspect.

[0027] In a fourth aspect, the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, is used to implement the method for assisting calibration and aiming training of a sight based on tag recognition as described in the first aspect.

[0028] The beneficial effects of the present invention are as follows: the present invention describes a method for realizing target image correction by using special target paper printed with labels during live-fire shooting scope calibration and aiming training, describes a method for calibrating the center of a target ring on a current image, describes a scheme for completing target impact point position calibration by using the bullet hole position hit in the first shot, describes a method for completing scope calibration by using the position difference between the target center point and the target single point, describes a method for completing aiming trajectory recording and drawing by storing the relative relationship between the target center point and the target impact point position; the present invention can realize scope auxiliary calibration and aiming training based on label recognition, with high calibration accuracy and training efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A flow chart of a method for performing auxiliary calibration of a sight based on tag recognition provided by an exemplary embodiment;

[0030] Figure 2 A schematic diagram of a hardware device for implementing the method of the present invention provided by an exemplary embodiment;

[0031] Figure 3 A schematic diagram of the location of the impact point of a target provided by an exemplary embodiment;

[0032] Figure 4 A flow chart of a method for aiming training based on tag recognition provided by an exemplary embodiment;

[0033] Figure 5 A diagram showing the trajectory of aiming training provided by an exemplary embodiment. DETAILED DESCRIPTION

[0034] In order to better understand the technical solution of the present application, the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0035] It should be clear that the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0036] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0037] The embodiment of the present invention provides a method for assisting calibration of a sight based on tag recognition, such as Figure 1 As shown, the specific process of this method is as follows:

[0038] 1. Install a telephoto video acquisition device on the gun

[0039] The telephoto video capture device mounting base is firmly mounted on the metal rail of the gun, which is a Picatinny rail in this embodiment. The video capture device mounting base may have a fine-tuning knob to achieve fine-tuning of the angle up, down, left, and right. The current video image of the video capture device can be roughly aligned with the image observed by the scope without precise adjustment.

[0040] 2. Determine the distance and arrange the target with QR code label

[0041] A target with a positioning identification two-dimensional code label is arranged at the target distance, and at least one two-dimensional code label is arranged in a blank area of ​​the target paper. In this embodiment, two-dimensional code labels are arranged at four corners of the target paper.

[0042] In this embodiment, Aruco Marker is used as a QR code label, and a Marker with ID 3 is printed on the upper left of the target paper, a Marker with ID 1 is printed on the upper right, a Marker with ID 2 is printed on the lower right, and a Marker with ID 0 is printed on the lower left. The image processing Opencv library is used to realize the recognition and positioning of the Aruco Marker in the captured image.

[0043] 3. Connect the video acquisition device to the image recognition device

[0044] The video acquisition device installed on the gun can be connected to the image recognition device wirelessly or by wire. When the wireless method is used, the video acquisition device has its own battery and wireless communication module.

[0045] In this embodiment, a wired connection is adopted, and the video acquisition device is equivalent to an external camera for the image recognition device, which transmits real-time images to the image recognition device and displays them on the display screen of the image recognition device.

[0046] The hardware equipment involved in the entire implementation process is as follows Figure 2 shown.

[0047] 4. Take a screenshot after aiming to save the current screen and complete the shooting action

[0048] Before the first aiming live-fire shooting, the current captured image is captured and saved as a base image, and the gun and the scope are kept stable, that is, the aiming position is kept unchanged, and a live-fire shooting is completed. The function of capturing and saving the current captured image can be completed through the input device of the image recognition device, such as a keyboard or a mouse, or through an external sensor, such as a micro switch linked to the trigger.

[0049] In this embodiment, the action of capturing and saving the currently captured image is performed by an external keyboard of the image recognition device.

[0050] 5. Mark the target impact point in the saved image

[0051] After the shooting is completed, the hit point is determined by observing the target in the real-time display image on the display screen of the image recognition device or viewing the target through a target mirror, and the base map captured and saved before the shooting is called out, and the real position of the hit point is manually calibrated on the base map to obtain the target impact point position coordinates at the current shooting distance; specifically, the live ammunition hit point position calibration can be completed by moving the cursor or inputting coordinates; the target impact point position coordinate data and the current shooting distance parameter data are stored in the image recognition device as a set of bound data, and a target impact point crosshair is statically drawn at the target impact point position coordinates on the display screen, such as Figure 3 As shown, it is to assist the shooter in observing the position of the expected shooting impact point pointed by the current gun in the entire target image.

[0052] 6. Adjust the sight until the crosshairs are aligned with the target impact point.

[0053] After the crosshairs of the target impact point are calibrated, the crosshairs of the sight can be adjusted to make the aiming positions of the crosshairs of the sight and the target impact point coincide with each other, and the sight calibration is completed. After the sight calibration, aiming training can be carried out.

[0054] In this embodiment, the gun posture is first adjusted until the static target impact point crosshairs on the image recognition device display screen are observed to be aligned with the target center, and then the gun posture is kept stable and the sight crosshairs are adjusted to align with the target center to achieve the coincidence of the two.

[0055] The embodiment of the present invention also provides a method for aiming training based on tag recognition, such as Figure 4 As shown, the specific process of this method is as follows:

[0056] 1. Collect the target image and implement perspective transformation correction based on the QR code label

[0057] The image processing Opencv library is used to realize the recognition and positioning of Aruco Marker in the acquired image, and the coordinates of the QR code label in the original target paper image are used as perspective transformation parameters to perform image transformation correction.

[0058] When there are four labels in the picture, select one corner point coordinate of each label as the parameter of perspective transformation; when there is only one label available in the picture, select the four corner point coordinates of the current label as the parameters of perspective transformation; when there are two labels in the picture, select two corner point coordinates of each label as the parameters of perspective transformation; when there are three labels in the picture, one of the labels selects two corner point coordinates as the parameters of perspective transformation, and the other two labels each select one corner point coordinate as the parameters of perspective transformation.

[0059] In this embodiment, the coordinates of the upper left corner point of Aruco Marker (ID3), the upper right corner point of (ID1), the lower right corner point of (ID2), and the lower left corner point of (ID0) are taken as perspective transformation parameters, and the coordinates of these four corner points in the original target paper image are: [0,0], [800,0], [800,830], [0,830], respectively.

[0060] 2. Calculate the coordinates of the target impact point in the current image relative to the target center

[0061] The perspective transformation parameters obtained from the two-dimensional code label are used to transform the coordinate values ​​of the target impact point in the original image to obtain the new coordinates M of the target impact point in the corrected image, and the new coordinates N of the target center in the current corrected image are obtained according to the coordinates of the two-dimensional code label in the original target paper image, thereby calculating the coordinate difference X between the coordinates M and N, and then setting the coordinate N as the new coordinate origin, and the coordinate difference X as the new coordinate value of the target impact point.

[0062] 3. Continuously draw the new coordinates of the target impact point and connect them into lines to save

[0063] The new coordinate values ​​of the identified and calculated target impact points relative to the target center are plotted in the electronic image of the target in the order of the images and connected by line segments, such as Figure 5 During the entire aiming training process, the aiming trajectory data can be stored in the image recognition device, or uploaded to the server through the network, and combined with the trigger micro switch linkage, the data collection during shooting can be realized to help the shooter determine whether the shooting time is the best.

[0064] In this embodiment, the trajectory data is only stored in a storage device built into the image recognition device.

[0065] Corresponding to the above-mentioned embodiment of the method for performing auxiliary calibration and aiming training of a sight based on tag recognition, the present invention also provides an embodiment of a system for performing auxiliary calibration and aiming training of a sight based on tag recognition. The system for performing auxiliary calibration and aiming training of a sight based on tag recognition provided by the present invention comprises the following modules:

[0066] Label recognition and image correction transformation module: Use the telephoto video acquisition device installed on the gun to capture the target paper with the QR code label, and transmit it to the image recognition device, and use image recognition to mark the label and position coordinates in the real-time picture; Combine the coordinates and size of the label on the real target paper, perform perspective transformation on the real-time picture image, and complete the correction of the captured image;

[0067] Target impact point calibration module: save the collected image before pulling the trigger for the first shot as the base map, and calibrate the real position of the target impact point on the base map after the first shot, so as to obtain the pixel position of the target impact point in the video collection image at the current shooting distance, that is, the target impact point position coordinates, and add a static target impact point crosshair mark at the target impact point position coordinates on the display screen of the image recognition device, and store the current shooting distance and the corresponding target impact point position coordinates in the image recognition device after pairing;

[0068] Sight calibration module: The video image transmitted by the video acquisition device and the target impact point crosshair corresponding to the current shooting distance are displayed in real time on the display screen of the image recognition device, and the image perspective transformation and correction are completed by identifying and locating the QR code label in the image, so as to complete the recognition of the target in the image and the calculation of the target center position, and display the distance difference between the target center point and the target impact point in real time; the shooter adjusts the gun posture until the target impact point crosshair on the display screen is observed to be aligned with the target center, and then keeps the gun posture unchanged and adjusts the sight crosshair until the sight crosshair is aligned with the target center, and the sight calibration is completed;

[0069] Aiming training auxiliary analysis module: Real-time analysis of the images transmitted by the video acquisition device, through the recognition and positioning of the QR code label in the image, to complete the image perspective transformation and correction, and use the recognized real target center as the coordinate origin, and the relative coordinates of the target impact point and the target center as the coordinate value, to record and store them in chronological order, and record and draw the entire running trajectory of the aiming process in the target image.

[0070] All relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module, and will not be repeated here.

[0071] In addition, the present invention also provides an embodiment of a device for performing assisted calibration and aiming training of a sight based on tag recognition, comprising a memory and one or more processors, wherein the memory stores executable code, and when the processor executes the executable code, it is used to implement the method for performing assisted calibration and aiming training of a sight based on tag recognition in the above-mentioned embodiment.

[0072] An embodiment of the present invention also provides a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, it is used to implement the method for assisting calibration and aiming training of a sight based on tag recognition in the above embodiment.

[0073] The computer-readable storage medium may be an internal storage unit of any device with data processing capability described in any of the aforementioned embodiments, such as a hard disk or a memory. The computer-readable storage medium may also be an external storage device of any device with data processing capability, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capability. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capability, and may also be used to temporarily store data that has been output or is to be output.

[0074] The above description is merely a preferred embodiment of one or more embodiments of the present specification and is not intended to limit one or more embodiments of the present specification. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of the present specification shall be included in the scope of protection of one or more embodiments of the present specification.

Claims

1. A method for assisting calibration and aiming training of a sight based on tag recognition, characterized in that: include: A real-time picture of a target with a QR code label is collected by a telephoto video acquisition device installed on the gun and transmitted to an image recognition device; the image recognition device recognizes the received image in real time and determines the coordinate position of the target in the image based on the QR code label, specifically: the coordinate position of the QR code label in the collected image is obtained by image recognition, and the coordinate and size of the QR code label on the real target paper are combined to perform perspective transformation on the real-time collected image, and the collected image is converted into a two-dimensional frontal plane image. The converted image is the target paper plane image. The coordinates of the target ring center in the target paper and the radius of the target ring are known, and the converted image can be used to obtain the coordinates of the target ring center in the target paper. The coordinate position of the target center point in the image is determined in the image after the image is acquired; the perspective transformation of the real-time acquired image is specifically as follows: when there are four labels in the picture, one corner point coordinate of each label is selected as the parameter of the perspective transformation; when only one label is available in the picture, the four corner point coordinates of the current label are selected as the parameters of the perspective transformation; when there are two labels in the picture, two corner point coordinates of each label are selected as the parameters of the perspective transformation; when there are three labels in the picture, one of the labels selects two corner point coordinates as the parameters of the perspective transformation, and the other two labels each select one corner point coordinate as the parameter of the perspective transformation; The captured image before the trigger is pulled for the first shot is saved as a base map, and the real position of the target impact point is calibrated on the base map after the first shot, so as to obtain the pixel position of the target impact point in the video captured image at the current shooting distance, that is, the target impact point position coordinates, and the current shooting distance is paired with the corresponding target impact point position coordinates and stored in the image recognition device; During the scope calibration or aiming training, a real-time picture of a target with a QR code label is captured by a telephoto video acquisition device, and the center position of the target in the real-time picture is marked by image recognition. A target impact point crosshair is statically drawn at the target impact point position coordinates corresponding to the current shooting distance on the display screen of the image recognition device, and the distance difference between the target impact point and the target center position is displayed in real time on the display screen. The shooter adjusts the gun posture until the target impact point crosshair on the display screen is observed to be aligned with the target center, and then keeps the gun posture still and adjusts the scope crosshair until the scope crosshair is aligned with the target center, thereby completing the scope calibration; After the sight is calibrated, during the aiming training process, the coordinates of the target center point are used as the initial position of the new coordinate axis, and the relative coordinates of the impact point in the image relative to the target center position are recorded as the coordinate data under the new coordinate axis, so as to obtain the aiming trajectory of the aiming point around the target center point and realize aiming training assistance.

2. The method for performing auxiliary calibration and aiming training of a sight based on tag recognition according to claim 1, characterized in that: The telephoto video acquisition device is a fixed-focus device or a zoom device, which is stably fixed on the gun. The actual coordinate position of the impact point in the acquired image is related to the current shooting distance, the installation angle of the acquisition device and the focal length parameters. When the acquisition device is reinstalled, adjusted and tightened, and the focal length is adjusted, the previously collected and recorded data is invalid and needs to be re-collected and calibrated.

3. The method for performing auxiliary calibration and aiming training of a sight based on tag recognition according to claim 1, characterized in that: For different shooting distances, the pixel positions of the target impact point in the video capture image are different. When the telephoto video capture device is not disassembled and reassembled and the focal length is not adjusted, the pixel position of the target impact point in the captured video image is considered to be fixed at the same shooting distance. Therefore, the pixel positions of the target impact point collected at multiple different shooting distances are stored in the image recognition device.

4. The method for performing auxiliary calibration and aiming training of a sight based on tag recognition according to claim 1, characterized in that: The number of the QR code label of the target with the QR code label is at least one, which is set in the blank area of ​​the target paper, and its size needs to ensure that it is recognizable to the human eye in the video screen; when a label is hit and damaged during the shooting process and cannot be recognized, as long as there is another recognizable label, the subsequent perspective transformation can still be completed.

5. The method for performing auxiliary calibration and aiming training of a sight based on tag recognition according to claim 1, characterized in that: The image recognition device recognizes the received image in real time, identifies the ID information of the QR code tag and the pixel coordinate values ​​of the four corner points of the current tag in the current video screen, and numbers the coordinates of the four corner points in a clockwise order of upper left, upper right, lower right, and lower left.

6. The method for performing auxiliary calibration and aiming training of a sight based on tag recognition according to claim 1, characterized in that: During the scope calibration process, a static target impact point crosshair is displayed on the display screen of the image recognition device. When the shooter is aiming, the image of the real target is also displayed on the display screen. The shooter adjusts the gun posture and scope crosshair so that the scope crosshair is aimed at the center of the real target. At the same time, the static target impact point crosshair on the display screen also coincides with the center of the real target in the captured image. At this time, the scope calibration is completed.

7. The method for performing auxiliary calibration and aiming training of a sight based on tag recognition according to claim 1, characterized in that: After completing the calibration of the sight, during the shooting and aiming training, as the shooter adjusts the gun posture to aim, the video acquisition device fixed on the gun will also transmit the real-time image of the target captured to the image recognition device. By using the image recognition device to complete the image correction, a flat target image can be obtained. At this time, the relative position of the pixel point position of the target impact point at the shooting distance recorded previously and the identified target center point position are recorded in chronological order to obtain a set of trajectory coordinates. The target center point is used as the origin coordinate and the relative position coordinate difference is used as the coordinate value. The trajectory coordinates are converted into a trajectory image running in the target image, which can depict the movement posture trajectory of the shooter aiming around the target.

8. A system for assisting calibration and aiming training of a sight based on tag recognition, comprising: Label recognition and image correction transformation module: Use the telephoto video acquisition device installed on the gun to capture the target paper with the QR code label printed on it, and transmit it to the image recognition device. Use image recognition to mark the label and position coordinates in the real-time picture. Specifically: use image recognition to obtain the coordinate position of the QR code label in the captured image, combine the coordinates and size of the QR code label on the real target paper, perform perspective transformation on the real-time captured image, and convert the captured image into a two-dimensional frontal plane image. The converted image is the target paper plane image. If the coordinates of the target ring center in the target paper and the radius of the target ring are known, the converted image can be Determine the coordinate position of the center point of the target in the image; the perspective transformation of the real-time acquired image is specifically as follows: when there are four tags in the picture, select one corner point coordinate of each tag as the parameter of the perspective transformation; when only one tag is available in the picture, select the four corner point coordinates of the current tag as the parameters of the perspective transformation; when there are two tags in the picture, each tag selects two corner point coordinates as the parameters of the perspective transformation; when there are three tags in the picture, one of the tags selects two corner point coordinates as the parameters of the perspective transformation, and the other two tags each select one corner point coordinate as the parameter of the perspective transformation; Target impact point calibration module: save the collected image before pulling the trigger for the first shot as the base map, and calibrate the real position of the target impact point on the base map after the first shot, so as to obtain the pixel position of the target impact point in the video collection image at the current shooting distance, that is, the target impact point position coordinates, and add a static target impact point crosshair mark at the target impact point position coordinates on the display screen of the image recognition device, and store the current shooting distance and the corresponding target impact point position coordinates in the image recognition device after pairing; Sight calibration module: The video image transmitted by the video acquisition device and the target impact point crosshair corresponding to the current shooting distance are displayed in real time on the display screen of the image recognition device, and the image perspective transformation and correction are completed by identifying and locating the QR code label in the image, so as to complete the recognition of the target in the image and the calculation of the target center position, and display the distance difference between the target center point and the target impact point in real time; the shooter adjusts the gun posture until the target impact point crosshair on the display screen is observed to be aligned with the target center, and then keeps the gun posture unchanged and adjusts the sight crosshair until the sight crosshair is aligned with the target center, and the sight calibration is completed; Aiming training auxiliary analysis module: Real-time analysis of the images transmitted by the video acquisition device, through the recognition and positioning of the QR code label in the image, to complete the image perspective transformation and correction, and use the recognized real target center as the coordinate origin, and the relative coordinates of the target impact point and the target center as the coordinate value, to record and store them in chronological order, and record and draw the entire running trajectory of the aiming process in the target image.

Images