Array type high-speed camera shooting track parameter measuring method, device and system

Abstract

The invention discloses a method, a device and a system for measuring ballistic parameters of array type high-speed camera shooting, and aims to solve the problem of limited measurement means at the initial stage and the final stage of launching of a high-rotation flying body. According to the method, a multi-stage trajectory detection algorithm combining a reverse dynamic threshold value and trajectory constraint guidance is adopted, a high-speed camera image sequence is intelligently processed, and trajectory parameters of the high-rotation flying body are calculated; on the device, a high-speed camera array, a muzzle initial-speed radar and a communication unmanned aerial vehicle network which are deployed in a launching area and a drop point area are combined, so that multi-angle and high-frame-rate synchronous observation and reliable data transmission of a trajectory key area are realized; in the system, through a data center deployed at a ground workstation, high-precision position, attitude and speed information of an initial segment obtained by high-speed camera shooting are fused under a unified time reference, a measurement bottleneck is broken through, a data blank is filled, and a key data support is provided for researching launching disturbance, flight stability and control system response.

Description

Technical Field

[0001] This invention relates to the field of flight parameter measurement technology for high-speed rotating aircraft, and more specifically, to a method, device, and system for measuring trajectory parameters using an array-type high-speed camera. Background Technology

[0002] Obtaining the motion parameters of high-spinning projectiles, such as guided artillery shells, during flight is crucial for their dynamic characteristic analysis, control system evaluation, and optimization. From a practical measurement perspective, the entire trajectory can be divided into five key stages based on the flight state and the applicability of measurement methods: the launch transient stage, the fusion transition stage, the integrated navigation stage, the control identification stage, and the landing point constraint stage. Among these, the launch transient stage and the landing point constraint stage present particularly severe measurement challenges.

[0003] During the launch transient phase (≤600ms), the aircraft is subjected to extreme high overload, severe impact vibration, and high-speed rotation, causing instantaneous saturation or failure of the internal in-situ measurement system. High engine exhaust temperatures and high-speed aerodynamic heat cause electronic components to operate beyond their limits, while strong electromagnetic interference and satellite signal loss further exacerbate the difficulty of data acquisition. During the landing constraint phase (approximately 2 seconds before landing), the aircraft is in a high-speed dive, and drastic attitude changes and environmental interference significantly increase measurement complexity. The available measurement methods are severely limited during these two phases, necessitating external measurement technologies to fill the data gaps.

[0004] High-speed camera measurement technology is unaffected by the extreme mechanical, thermal, and electromagnetic environments inside the aircraft, and can directly capture the transient spatial position and attitude changes of high-speed rotating bodies at a rate of tens of thousands of frames per second, effectively filling the data gaps during the launch transient phase and the landing point constraint phase. However, the accuracy of high-speed camera measurement alone is easily affected by lighting, image quality, calibration accuracy, and background interference, making it difficult to adapt to the complex and variable environment during the launch of high-speed rotating aircraft. Therefore, there is an urgent need to develop an innovative method to provide high-precision trajectory parameter measurement results when measurement methods are limited during the launch transient phase and the landing point constraint phase, ultimately achieving comprehensive acquisition of high-precision and complete flight parameters of high-speed rotating aircraft throughout their entire flight. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide an array-type high-speed camera trajectory parameter measurement method, device and system to achieve high-precision and complete acquisition of flight parameters of high-rotation flying objects throughout their entire flight.

[0006] An array-type high-speed camera ballistic parameter measurement system includes a high-speed camera array 100 for the launch area, a high-speed camera array 200 for the impact area, a communication UAV array 300, a muzzle velocity radar 400, and a data center 500 installed on a ground workstation.

[0007] The high-speed camera array 100 in the launch area includes at least three high-speed cameras, which are deployed on both sides and behind the launch point, respectively. One of the high-speed cameras is located directly behind the launch point and is used to monitor the attitude, speed and position of the high-spiral flying object in the yaw direction after launch. Two of the high-speed cameras are located directly in front of the launch point and are symmetrically distributed along the launch surface to monitor the attitude, speed and position of the high-spiral flying object in the trajectory after launch. The high-speed camera array 200 in the landing area includes at least three high-speed cameras, which are deployed on both sides and in front of the preset landing point, respectively. One of the high-speed cameras is located directly in front of the preset landing point and is used to monitor the attitude, speed and position of the high-speed spinning object in the yaw direction before landing. The other two high-speed cameras are located directly behind the preset landing point and are symmetrically distributed along the launch surface to monitor the attitude, speed and position of the high-speed spinning object in the launch direction before landing. The communication drone array 300 is used to provide a relay node for the communication link, and can maintain communication with the in-situ measurement device and the measurement drone array of the high-rotation flying body throughout the process, and transmit the measurement information to the ground workstation in real time to form an air-to-ground communication link for real-time networking communication. The muzzle velocity radar 400 is placed in front of the muzzle, with its beam axis strictly parallel and aligned with the barrel axis, and uses the Doppler effect to measure the initial velocity of the high-rotation flying object when it leaves the muzzle. The data center 500 receives measurement data in real time from the high-speed camera array 100 in the launch area and the high-speed camera array 200 in the landing area via a communication drone, and then fuses them under the same time reference.

[0008] Preferably, the high-speed camera array 100 in the launch area and the high-speed camera array 200 in the impact area are equipped with multiple optical image acquisition units. Each unit is equipped with a high frame rate camera and an optical lens and is arranged on both sides of the ballistic trajectory to synchronously acquire image sequences of the high-spinning flight object as it leaves the barrel and during its flight.

[0009] Preferably, the high-speed camera array 100 in the launch area and the high-speed camera array 200 in the landing area also integrate communication modules for transmitting image data to the data center 500 via the communication drone array 300.

[0010] Preferably, the muzzle velocity radar 400 has a built-in Doppler measurement unit for detecting the initial velocity of a high-speed flying object at the moment of exiting the barrel, and transmits the initial velocity data to the data center 500 through a communication module.

[0011] Preferably, the communication drone array 300 is deployed over the measurement area and is equipped with a communication module.

[0012] Preferably, the data center 500 is the core processing unit of the system, including a system initialization and parameter configuration module, an image frame reading and preprocessing module, an inverse dynamic threshold trajectory detection module, a ballistic modeling module, a target detection module based on ballistic constraints, a target trajectory optimization and coordinate transformation module, and a velocity calculation and result output module.

[0013] A method for measuring ballistic parameters of a measurement system, comprising: Step S21: Extract the video data transmitted back from each high-speed camera in the high-speed camera array 100 in the transmission area and the high-speed camera array 200 in the landing point, and perform system initialization and parameter configuration; Step S22: Image frame reading and preprocessing; Step S23: Inverse dynamic threshold trajectory detection: Based on the shooting environment and image quality, a decreasing grayscale threshold sequence is set; the frames in the video are divided into two parts, the first and the second, and the frames in the second half of the video are extracted and detected using the following method: The process proceeds in reverse order from the end frame to the middle frames. For each frame, the following operations are performed: the first threshold in the grayscale threshold sequence is read; the preprocessed image is binarized and segmented based on the current threshold; a disk structuring element is used to perform a closing operation to connect adjacent regions; connected components are marked and morphological features are extracted; the area, aspect ratio, and solidity of the extracted target are sequentially determined to be within the corresponding threshold range; then, the displacement per unit time is determined to be within the velocity threshold range. If all the above conditions are met, the detection is successful, and the extracted target coordinates are stored; if the detection fails, the next threshold is used for binarization. If the detection fails after all thresholds in the grayscale threshold sequence have been traversed, the frame is marked as an abnormal frame, and the next frame is switched for detection. Step S24: Model the trajectory of the high-spinning flight object; Step S25: Target detection based on trajectory constraints: First, based on the trajectory model, the trajectory is translated both upwards and downwards. This forms two trajectory constraint zones: (6) In the formula, Add boundaries to the trajectory constraints. The trajectory constraint includes a lower boundary; This is the set y-axis translation amount; in a 1080p image, Pixel; Secondly, based on the target position detected in the previous frame and speed Establish velocity-based constraint bands: (7) In the formula, The frame rate of the video; The x-axis translation amount is set; in a 1080p image, Pixel.

[0014] For the region enclosed by the trajectory constraint band and the velocity-based constraint band, target detection is performed on the first half of the frame within the region furthest from the launch point: First, select the centroid of the connected component with the largest area as the target location. When target detection fails, the target's position is extrapolated using a trajectory model, and set to... ;by A secondary detection region of a predetermined size is established around the center, and the target is re-detected. If the secondary detection fails, this frame is recorded as an abnormal frame, and its position is extrapolated. The velocity-based constraint band is updated using equation (7); Step S26: Optimize the target trajectory and transform the coordinates to the physical space coordinate system; Step S27: Calculate and output the actual velocity of the target based on the target trajectory.

[0015] Preferably, the image frame reading and preprocessing in step S22 includes: First, the RGB image is converted to the YUV color space and the luminance component is extracted; then, an S-curve is used to adjust the pixel intensity distribution. Secondly, suppress Gaussian noise and enhance local contrast; Next, a sharpening operation is performed to enhance high-frequency components by comparing the original image with the Gaussian blur image; Finally, based on the target's reflectivity, it is determined whether to perform pixel value inversion processing. That is, the median value of all pixels in the image is calculated, and if it is less than a set threshold, pixel value inversion processing is performed.

[0016] Ideally, in 1080p images, Pixel; Pixel.

[0017] The present invention has the following beneficial effects: This invention discloses a method, device, and system for measuring trajectory parameters using array-type high-speed cameras, aiming to address the limited means of measuring the initial and final stages of launch of high-spiral aircraft. The method employs a multi-level trajectory detection algorithm combining inverse dynamic thresholding and trajectory constraint guidance to intelligently process high-speed camera image sequences and calculate the trajectory parameters of the high-spiral aircraft. The device combines a high-speed camera array deployed in the launch and impact areas, a muzzle velocity radar, and a communication UAV network to achieve multi-angle, high-frame-rate synchronous observation and reliable data transmission in key trajectory areas. The system utilizes a data center deployed at a ground workstation to fuse high-precision position, attitude, and velocity information acquired from the initial phase of high-speed imaging under a unified time reference, overcoming measurement bottlenecks, filling data gaps, and providing crucial data support for research on launch disturbances, flight stability, and control system response. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of an array-type high-speed camera trajectory parameter measurement system according to a first embodiment of the present invention; Figure 2 This is a side view of the high-speed camera array in the transmission area of ​​the array-type high-speed camera trajectory parameter measurement system according to the first embodiment of the present invention; Figure 3 This is a top view of the high-speed camera array in the transmission area of ​​the array-type high-speed camera trajectory parameter measurement system according to the first embodiment of the present invention; Figure 4 This is a side view of the high-speed camera array at the landing point in the array-type high-speed camera trajectory parameter measurement system according to the first embodiment of the present invention; Figure 5 This is a top view of the high-speed camera array for landing points in the array-type high-speed camera trajectory parameter measurement system according to the first embodiment of the present invention; Figure 6 This is a structural diagram of an array-type high-speed camera trajectory parameter measuring device according to a second embodiment of the present invention; Figure 7 This is a structural diagram of the algorithm for measuring trajectory parameters of an array-type high-speed camera according to a third embodiment of the present invention. Figure 8 This is a schematic diagram of a target detection method based on trajectory constraints in an array-type high-speed camera trajectory parameter measurement system according to a third embodiment of the present invention. Detailed Implementation

[0019] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] Example 1: This embodiment provides an array-type high-speed camera trajectory parameter measurement system, such as... Figure 1As shown, it includes a high-speed camera array 100 for the launch area, a high-speed camera array 200 for the impact area, a UAV array 300, a muzzle velocity radar 400, and a data center 500 installed on a ground workstation.

[0021] The high-speed camera array 100 in the launch area includes three high-speed cameras 101, 102, and 103, which are deployed on both sides and behind the launch point, respectively. Figure 2 and Figure 3 As shown, the high-speed camera 103 behind the launch point is located at x2 directly behind the launch point and is used to monitor the attitude, speed and position of the high-speed cyclone projectile in the yaw direction after launch; the high-speed cameras 101 and 102 on both sides of the launch point are located at x1 directly in front of the launch point, symmetrically distributed along the launch surface, and at a distance L1 from the launch surface, and are used to monitor the attitude, speed and position of the high-speed cyclone projectile in the trajectory after launch. The high-speed camera array 200 for the landing area includes three high-speed cameras 201, 202, and 203, which are deployed on both sides and in front of the preset landing point, respectively. Figure 4 and Figure 5 As shown, the high-speed camera 203 in front of the preset landing point is located at x4 directly in front of the preset landing point, and is used to monitor the attitude, speed and position of the yaw direction of the high-spiral flying object before landing; the high-speed cameras 201 and 202 on both sides of the preset landing point are located at x3 directly behind the preset landing point, symmetrically distributed along the launch surface, and at a distance of L2 from the launch surface, and are used to monitor the attitude, speed and position of the trajectory of the high-spiral flying object before landing. The UAV array 300 is used to provide relay nodes for the communication link, and can maintain communication with the in-situ measurement device and the measurement UAV array of the high-speed flying body throughout the process, and transmit the measurement information to the ground workstation in real time to form an air-to-ground communication link for real-time network communication.

[0022] The muzzle velocity radar 400 is placed in front of the muzzle, with its beam axis strictly parallel and aligned with the gun barrel axis. It uses the Doppler effect to measure the initial velocity of a high-speed flying object when it exits the muzzle.

[0023] The data center 500 installed on the ground workstation receives measurement data in real time from the high-speed camera array 100 in the measurement launch area and the high-speed camera array 200 in the landing area via a communication drone, and fuses them under the same time reference.

[0024] Example 2: This embodiment provides an array-type high-speed camera trajectory parameter measurement device, such as... Figure 7 As shown, it includes a high-speed camera array unit 101102 / 103 for the launch area, a high-speed camera array unit 201 / 202 / 203 for the impact area, a muzzle velocity radar 400, a communication drone array 300, and a data center 500.

[0025] The high-speed camera array units 101102 / 103 in the launch area and 201 / 202 / 203 in the landing area include multiple optical image acquisition units. Each unit is equipped with a high frame rate camera and an optical lens, and is arranged on both sides of the trajectory to synchronously acquire image sequences of the high-speed spinning aircraft as it leaves the barrel and during its flight. Each unit also integrates a communication module for transmitting image data to a data center.

[0026] The muzzle velocity radar 400 has a built-in Doppler measurement unit for detecting the initial velocity of a high-speed flying object at the moment of exiting the barrel, and transmits the initial velocity data to the data center 500 through a communication module.

[0027] The communication drone array 300 is deployed over the measurement area, equipped with a communication module, and serves as an airborne relay node to enhance the reliability and coverage of communication between ground equipment and the data center.

[0028] The data center 500 is the core processing unit of the system, including the following functional modules: system initialization and parameter configuration module, image frame reading and preprocessing module, inverse dynamic threshold trajectory detection module, trajectory modeling module, target detection module based on trajectory constraints, target trajectory optimization and coordinate transformation module, and velocity calculation and result output module.

[0029] Example 3: This embodiment provides a method for measuring trajectory parameters of an array-type high-speed camera, such as... Figure 8 As shown, the processing of video data transmitted back from each high-speed camera in the transmitting area high-speed camera array 100 and the landing point high-speed camera array 200 includes the following steps: Step S21: System initialization and parameter configuration First, a parallel computing thread pool is created to accelerate computation. Second, multi-dimensional parameters are configured: for the object detection module, parameters such as minimum / maximum area thresholds, aspect ratio range, and solidity requirements are set; for the trajectory detection module, parameters such as boundary spacing and minimum number of trajectory points are set; for the preprocessing module, parameters such as enabling and setting the intensity of image inversion, contrast enhancement, sharpening, and noise reduction are set; and for the physical parameter module, parameters such as video frame rate are set.

[0030] Step S22: Image frame reading and preprocessing Perform preprocessing frame by frame: First, the RGB image is converted to the YUV color space and the luminance component is extracted; then, the pixel intensity distribution is adjusted by S-curve mapping, as shown in formula (1). (1) In the formula, This is the threshold for the brightness range. These are the adjusted pixel values.

[0031] Secondly, a nonlocal mean filtering algorithm is applied to calculate a weighted average within an 11×11 window to suppress Gaussian noise; contrast-limited adaptive histogram equalization (CLAHE) is used to enhance local contrast within an 8×8 block. Next, perform unsharpened mask (USM) sharpening to enhance high-frequency components by using the difference between the original image and the Gaussian blur image, as shown in formula (2): (2) In the formula, Ieq is the input image. λ is a Gaussian filter, and λ is the sharpening intensity factor.

[0032] Finally, based on the target's reflectivity, it is determined whether to perform pixel value inversion processing. That is, the median of all pixels in the image is calculated. If it is less than a set threshold (the threshold is set to 60 in this embodiment), then pixel value inversion processing is performed.

[0033] Step S23: Inverse dynamic threshold trajectory detection Since high-spinning projectiles are often accompanied by dense smoke during launch, the presence of dense smoke can greatly affect image processing and the recognition of high-spinning projectiles. Therefore, a reverse dynamic threshold trajectory detection method is adopted to detect the trajectory of high-spinning projectiles in reverse order.

[0034] A decreasing grayscale threshold sequence is set based on the shooting environment and image quality. The video frames are divided into two parts (first and second) according to the set parameters. Frames in the second half of the video are extracted and detected using the following method: The processing proceeds in reverse order from the end frame to the middle frame. For each frame, the following operations are performed: the first threshold in the grayscale threshold sequence is read, the preprocessed image is binarized based on the current threshold, and the disk structuring element is used to perform a closing operation to connect adjacent regions; after marking the connected components, the centroid coordinates, the bounding rectangle, the length of the major axis / minor axis, and the solidity are extracted as morphological features. A multi-level filtering decision tree is constructed to eliminate various types of noise extracted incorrectly: the area, aspect ratio, and solidity of the extracted target are judged in turn to see if they exceed the corresponding threshold range; then, it is judged whether the displacement per unit time exceeds the speed threshold range, which is the speed of the target movement obtained by using the target position at the previous moment, as shown in formula (3). If all the above conditions are met, the multi-level decision tree is passed, that is, the detection is successful, and the extracted target coordinates are stored. If the multi-level decision tree is not passed, the next threshold is used for binarization. When all thresholds in the grayscale threshold sequence have been traversed and the detection is still unsuccessful, this frame is marked as an abnormal frame, and the next frame is switched for detection.

[0035] (3) In the formula, Let be the position vector of the k-th frame.

[0036] Step S24: Trajectory Modeling In the field of view of a high-speed camera, the trajectory can be approximated as a straight line. The RANSAC robust regression algorithm is used to fit the trajectory model: three points are randomly selected to generate candidate straight line equations; the perpendicular distance from all points to the straight line is calculated; interior points with a distance less than 5 pixels are retained; after 100 iterations, the model with the most interior points is selected. (4) Refitting the accurate trajectory equation using the least squares method based on the interior point set (5) (5) In the formula, These are the pixel coordinates of the high-speed flying object in the image.

[0037] Step S25: Target detection based on trajectory constraints After trajectory modeling, the detection range is constrained using the trajectory model and the position and velocity from the previous moment. Detection is then performed within this constrained range, processing the first half of the frame. First, based on the trajectory model, the trajectory is translated both upwards and downwards. This forms two trajectory constraint zones, as shown in formula (6).

[0038] (6) In the formula, Add boundaries to the trajectory constraints. The lower boundary is used as the trajectory constraint. In a 1080p image, Pixel.

[0039] Secondly, based on the target position detected in the previous frame and speed Establish a velocity-based constraint band, i.e., formula (7): (7) In the formula, The frame rate of the video, in a 1080p image. Pixel.

[0040] For the region enclosed by the trajectory constraint zone and the velocity-based constraint zone, subsequent target detection is performed in the region furthest from the launch point: First, calculate the local Otsu threshold and perform binarization segmentation. Then, perform morphological opening operations and select the centroid of the connected component with the largest area as the target location. When target detection fails, the position of the target is extrapolated using trajectory equation (5), and set as follows: ;by A secondary detection region of a predetermined size (50×50 pixels in this embodiment) is established around the center. A local adaptive threshold is applied for re-detection. If the secondary detection fails, this frame is recorded as an abnormal frame, and its position is extrapolated. The velocity-based constraint band is updated using equation (7).

[0041] Step S26: Target trajectory optimization and coordinate transformation First, target trajectory optimization is performed: outliers in the detected position sequence are scanned and replaced with linear interpolation of the preceding and following valid points. A Savitzky-Golay filter is applied for smoothing, and a cubic polynomial fit is performed within an N-frame window, i.e., formula (8). (8) In the formula, These are the filter coefficients.

[0042] Next, coordinate system transformation is performed: the position of the gun muzzle reference point in the image coordinate system is calculated, and the origin of the coordinate system is translated and the direction is adjusted. Finally, unit conversion is performed based on the calibration parameters to generate the position sequence of the physical space coordinate system. That is, formula (9). (9) In the formula, η is the pixel-to-physical unit conversion coefficient, with units of m / pixel. The coordinates are the reference point coordinates.

[0043] Step S27: Velocity Calculation and Result Output The positional changes of these feature points in consecutive frames are analyzed, and the instantaneous velocity components are calculated using the central difference method, where the time interval is determined according to the frame rate of the video, i.e., formula (10).

[0044] (10) In the formula, This refers to the frame rate of the video.

[0045] The calculated horizontal and vertical velocities are vector-combined to obtain the target's actual velocity, i.e., formula (11).

[0046] (11) The position and velocity results are output after performing cubic polynomial fitting within an N-frame window.

[0047] In summary, the above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An array-type high-speed camera ballistic parameter measurement system, characterized in that, It includes a high-speed camera array for the launch area (100), a high-speed camera array for the impact area (200), a communication drone array (300), a muzzle velocity radar (400), and a data center (500) installed on a ground workstation. The high-speed camera array (100) in the launch area includes at least three high-speed cameras, which are deployed on both sides and behind the launch point, respectively. One of the high-speed cameras is located directly behind the launch point and is used to monitor the attitude, speed and position of the high-spiral flying object in the yaw direction after launch. Two of the high-speed cameras are located directly in front of the launch point and are symmetrically distributed along the launch surface to monitor the attitude, speed and position of the high-spiral flying object in the trajectory after launch. The high-speed camera array (200) in the landing area includes at least three high-speed cameras, which are deployed on both sides and in front of the preset landing point, respectively. One of the high-speed cameras is located directly in front of the preset landing point and is used to monitor the attitude, speed and position of the yaw direction of the high-spiral flying object before landing. The other two high-speed cameras are located directly behind the preset landing point and are symmetrically distributed along the launch surface to monitor the attitude, speed and position of the trajectory of the high-spiral flying object before landing. The communication UAV array (300) is used to provide a relay node for the communication link, and can maintain communication with the in-situ measurement device and the measurement UAV array of the high-speed flying body throughout the process, and transmit the measurement information to the ground workstation in real time to form an air-to-ground communication link for real-time network communication; The muzzle velocity radar (400) is placed in front of the muzzle, with its beam axis strictly parallel and aligned with the barrel axis, and uses the Doppler effect to measure the initial velocity of the high-speed flying object when it leaves the muzzle. The data center (500) receives measurement data from the high-speed camera array (100) in the launch area and the high-speed camera array (200) in the landing area in real time via a communication drone, and then fuses them under the same time reference.

2. The array-type high-speed camera ballistic parameter measurement system as described in claim 1, characterized in that, The high-speed camera array (100) in the launch area and the high-speed camera array (200) in the impact area are equipped with multiple optical image acquisition units. Each unit is equipped with a high frame rate camera and an optical lens and is arranged on both sides of the ballistic trajectory to synchronously acquire image sequences of the high-spinning flight body as it leaves the barrel and during its flight.

3. The array-type high-speed camera ballistic parameter measurement system as described in claim 1, characterized in that, The high-speed camera array (100) in the launch area and the high-speed camera array (200) in the landing area also integrate a communication module for transmitting image data to the data center (500) via the communication drone array (300).

4. The array-type high-speed camera ballistic parameter measurement system as described in claim 1, characterized in that, The muzzle velocity radar (400) has a built-in Doppler measurement unit for detecting the initial velocity of a high-speed flying object at the moment of exiting the barrel, and transmits the initial velocity data to the data center (500) through a communication module.

5. The array-type high-speed camera ballistic parameter measurement system as described in claim 1, characterized in that, The communication drone array (300) is deployed over the measurement area and is equipped with a communication module.

6. The array-type high-speed camera ballistic parameter measurement system as described in claim 1, characterized in that, The data center (500) is the core processing unit of the system, including a system initialization and parameter configuration module, an image frame reading and preprocessing module, an inverse dynamic threshold trajectory detection module, a ballistic modeling module, a target detection module based on ballistic constraints, a target trajectory optimization and coordinate transformation module, and a velocity calculation and result output module.

7. A method for measuring ballistic parameters based on the measurement system of any one of claims 1 to 7, characterized in that, include: Step S21: Extract the video data transmitted back from each high-speed camera in the high-speed camera array (100) in the transmission area and the high-speed camera array (200) in the landing area, and perform system initialization and parameter configuration; Step S22: Image frame reading and preprocessing; Step S23: Inverse dynamic threshold trajectory detection: Based on the shooting environment and image quality, a decreasing grayscale threshold sequence is set; the frames in the video are divided into two parts, the first and the second, and the frames in the second half of the video are extracted and detected using the following method: The processing proceeds in reverse order from the end frame to the middle frame. For each frame, the following operations are performed: the first threshold in the grayscale threshold sequence is read, the preprocessed image is binarized based on the current threshold, and a disk structuring element is used to perform a closing operation to connect adjacent regions; after marking the connected components, morphological features are extracted. The system sequentially checks whether the extracted target area, aspect ratio, and solidity exceed the corresponding threshold range; then it checks whether the displacement per unit time exceeds the velocity threshold range. If all the above conditions are met, the detection is successful, and the extracted target coordinates are stored. If the detection fails, the next threshold is used for binarization segmentation. If the detection still fails after all thresholds in the grayscale threshold sequence have been traversed, the frame is marked as an abnormal frame and the next frame is switched for detection. Step S24: Model the trajectory of the high-spinning flight object; Step S25: Target detection based on trajectory constraints: First, based on the trajectory model, the trajectory is translated both upwards and downwards. This forms two trajectory constraint zones: （6） In the formula, Add boundaries to the trajectory constraints. The trajectory constraint includes a lower boundary; This is the set y-axis translation amount; in a 1080p image, Pixel; Secondly, based on the target position detected in the previous frame and speed Establish velocity-based constraint bands: （7） In the formula, The frame rate of the video; The x-axis translation amount is set; in a 1080p image, Pixel. For the region enclosed by the trajectory constraint band and the velocity-based constraint band, target detection is performed on the first half of the frame within the region furthest from the launch point: First, select the centroid of the connected component with the largest area as the target location. ; When target detection fails, the target's position is extrapolated using a trajectory model, and set to... ;by A secondary detection region of a predetermined size is established around the center, and the target is re-detected. If the secondary detection fails, this frame is recorded as an abnormal frame, and its position is extrapolated. The velocity-based constraint band is updated using equation (7); Step S26: Optimize the target trajectory and transform the coordinates to the physical space coordinate system; Step S27: Calculate and output the actual velocity of the target based on the target trajectory.

8. The ballistic parameter measurement method as described in claim 7, characterized in that, Step S22, image frame reading and preprocessing, includes: First, the RGB image is converted to the YUV color space and the luminance component is extracted; then, an S-curve is used to adjust the pixel intensity distribution. Secondly, suppress Gaussian noise and enhance local contrast; Next, a sharpening operation is performed to enhance high-frequency components by comparing the original image with the Gaussian blur image; Finally, based on the target's reflectivity, it is determined whether to perform pixel value inversion processing. That is, the median value of all pixels in the image is calculated, and if it is less than a set threshold, pixel value inversion processing is performed.

9. The ballistic parameter measurement method as described in claim 7, characterized in that, In 1080p images, Pixel; Pixel.

Images