A spatial positioning method and system for moving targets with a large field of view

Abstract

The present invention provides a method and system for spatially locating a moving target with a large field of view. The system includes: a first sensor array, a second sensor array, a synchronous acquisition module connected to the first sensor array and the second sensor array, and a terminal connected to the synchronous acquisition module. The first sensor array and the second sensor array are both dynamic visual sensor arrays. The method includes: pre-calibrating the first sensor array and the second sensor array to obtain calibration information; the synchronous acquisition module collects the pulse sequence generated when the first sensor array and the second sensor array detect the moving target; the terminal obtains the pulse sequence and obtains the spatial three-dimensional coordinates of the moving target based on the calibration information and the pulse sequence. The present invention forms a large field of view by splicing two dynamic visual sensor arrays. When detecting a moving target, the spatial three-dimensional coordinates of the moving target are obtained based on the collected pulse sequence, thereby meeting the demand for spatial positioning of a moving target with a large field of view.

Description

Technical Field

[0001] The present invention relates to the field of computational vision measurement technology, and in particular to a spatial positioning method and system for a moving target with a large field of view. Background Art

[0002] Achieving three-dimensional spatial positioning of high-speed moving targets provides an important observation tool for analyzing insect trajectory, high-speed aircraft trajectory, ballistic analysis, surveillance, and early warning. Freely moving high-speed targets often have a large range of motion, and adapting to these high speeds and large ranges is a challenging problem for achieving high-speed target spatial positioning.

[0003] Conventional photoelectric tracking and measurement instruments use "frame" cameras to image, detect, and track targets. For high-speed targets, not only are the images blurred, but the target can also be easily lost. A dynamic vision sensor is a new type of bionic vision sensor that outputs a pulse data stream only when the scene light intensity changes. It has the advantages of a large dynamic range, high temporal resolution, and no motion blur, making it an ideal sensor for high-speed target perception. Binocular dynamic vision sensors can measure the depth of high-speed targets. Their main principle is to calculate parallax through binocular matching to achieve depth measurement. However, their small field of view makes it difficult to meet the requirements for spatial positioning of high-speed targets with a large field of view.

[0004] Therefore, the existing technology has defects and needs to be improved and developed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method and system for spatial positioning of moving targets with a large field of view in response to the above-mentioned defects of the prior art, aiming to solve the problem that conventional photoelectric tracking measuring instruments and binocular dynamic vision sensors in the prior art are difficult to meet the requirements of spatial positioning of large-field-of-view and high-speed targets.

[0006] The technical solutions adopted by the present invention to solve the technical problems are as follows:

[0007] A method for spatially locating a moving target with a large field of view is provided. The method is implemented based on a spatial positioning system for moving targets with a large field of view. The spatial positioning system comprises: a first sensor array, a second sensor array, a synchronous acquisition module connected to the first sensor array and the second sensor array, and a terminal connected to the synchronous acquisition module; the first sensor array and the second sensor array are both dynamic visual sensor arrays;

[0008] The spatial positioning method of the moving target includes:

[0009] Pre-calibrating the first sensor array and the second sensor array to obtain calibration information;

[0010] The synchronous acquisition module acquires a pulse sequence generated when the first sensor array and the second sensor array detect a moving target;

[0011] The terminal obtains the pulse sequence, and obtains the spatial three-dimensional coordinates of the moving target according to the calibration information and the pulse sequence.

[0012] In one implementation, pre-calibrating the first sensor array and the second sensor array to obtain calibration information includes:

[0013] Acquire a calibration image, and perform intrinsic parameter calibration on each sensor in the first sensor array and the second sensor array according to the calibration image to obtain calibrated intrinsic parameter information;

[0014] The installation azimuth angle of each sensor is calibrated by using a rotating table and marking points to obtain the calibrated installation azimuth angle information;

[0015] The center distance between the first sensor array and the second sensor array is measured by using a distance ruler or a distance meter to obtain the center distance.

[0016] In one implementation, acquiring a calibration image, and performing intrinsic parameter calibration on each sensor in the first sensor array and the second sensor array according to the calibration image to obtain calibrated intrinsic parameter information includes:

[0017] Capturing a plurality of checkerboard images from different angles for each sensor in the first sensor array and the second sensor array, and using the checkerboard images as calibration images;

[0018] The intrinsic parameters of each sensor are calibrated using Zhang's calibration method according to the calibration image to obtain calibrated intrinsic parameter information.

[0019] In one implementation, calibrating the installation azimuth of each sensor using a rotating stage and a marking point to obtain calibrated installation azimuth information includes:

[0020] The first sensor array and the second sensor array are pre-mounted on a turntable, and the center of each sensor's image is aligned with a preset mark point in sequence using the turntable;

[0021] The rotation azimuth of the turntable when aligning with the preset mark point is recorded, and the installation azimuth of each sensor relative to the center of each array is calibrated according to the rotation azimuth to obtain calibrated installation azimuth information.

[0022] In one implementation, the terminal obtains the pulse sequence, and obtains the spatial three-dimensional coordinates of the moving target according to the calibration information and the pulse sequence, including:

[0023] The terminal obtains the pulse sequence and corrects image distortion of the pulse sequence using the intrinsic parameter information;

[0024] Determine the first centroid coordinates of the moving target in the first sensor array and the second centroid coordinates in the second sensor array according to the corrected pulse sequence;

[0025] Determining a first azimuth angle of the moving target relative to a center of a first sensor array and a second azimuth angle of the moving target relative to a center of a second sensor array according to the first centroid coordinates, the second centroid coordinates, and the installation azimuth information;

[0026] The spatial three-dimensional coordinates of the moving target are calculated according to the first azimuth angle, the second azimuth angle and the center distance.

[0027] In one implementation, determining the first center-of-mass coordinates of the moving target in the first sensor array and the second center-of-mass coordinates of the moving target in the second sensor array according to the corrected pulse sequence includes:

[0028] Segmenting the pulse sequence according to a preset time window and reconstructing an image of the moving target;

[0029] Binarize the reconstructed image to obtain the target image;

[0030] The first centroid coordinates of the moving target in the first sensor array and the second centroid coordinates of the moving target in the second sensor array are calculated based on the target image.

[0031] In one implementation, the first sensor array and the second sensor array have the same structure and the same large field of view.

[0032] In one implementation, the calculation formula for the first azimuth angle of the moving target relative to the center of the first sensor array is:

[0033]

[0034] The calculation formula of the second azimuth angle of the moving target relative to the center of the second sensor array is:

[0035]

[0036] in, is the coordinate of the first centroid; is the coordinate of the second centroid; (w,h) represents the lateral and longitudinal resolutions of the sensor array; (W,H) represents the horizontal and vertical fields of view of the corresponding sensor; represents the calibrated installation azimuth of the corresponding sensor in the first sensor array; Indicates the calibrated installation azimuth of the corresponding sensor in the second sensor array.

[0037] In one implementation, calculating the spatial three-dimensional coordinates of the moving target according to the first azimuth angle, the second azimuth angle, and the center distance includes:

[0038] Acquire a first straight line from the center point of the first sensor array to the moving target, and a second straight line from the center point of the second sensor array to the moving target;

[0039] The midpoint of the perpendicular bisector between the first straight line and the second straight line is calculated to obtain the three-dimensional spatial coordinates of the moving target.

[0040] In one implementation, the calculation formula for the spatial three-dimensional coordinates of the moving target is:

[0041]

[0042] in,

[0043]

[0044]

[0045]

[0046]

[0047] P (x, y, z) is the three-dimensional coordinate of the moving target; L is the center point of the first sensor array, the O R is the center point of the second sensor array; B is L With O R The center distance of .

[0048] The present invention also discloses a spatial positioning system for a moving target with a large field of view, comprising:

[0049] The first sensor array and the second sensor array are used to simultaneously detect a moving target and generate a pulse sequence when a moving target is detected;

[0050] a synchronous acquisition module, connected to the first sensor array and the second sensor array, for acquiring a pulse sequence generated when the first sensor array and the second sensor array detect a moving target;

[0051] a terminal connected to the synchronous acquisition module, configured to pre-calibrate the first sensor array and the second sensor array to obtain calibration information, and obtain the pulse sequence, and obtain the three-dimensional spatial coordinates of the moving target based on the calibration information and the pulse sequence;

[0052] The first sensor array and the second sensor array are both dynamic vision sensor arrays.

[0053] The present invention provides a method and system for spatially locating a moving target with a large field of view. The method is implemented based on a spatial positioning system for a moving target with a large field of view. The spatial positioning system includes: a first sensor array, a second sensor array, a synchronous acquisition module connected to the first and second sensor arrays, and a terminal connected to the synchronous acquisition module. The first and second sensor arrays are both dynamic visual sensor arrays. The method includes: pre-calibrating the first and second sensor arrays to obtain calibration information; the synchronous acquisition module collects a pulse sequence generated when the first and second sensor arrays detect a moving target; the terminal obtains the pulse sequence and obtains the three-dimensional spatial coordinates of the moving target based on the calibration information and the pulse sequence. The present invention forms a large field of view by splicing two sensor arrays. The field of view size and array structure can be designed according to specific application requirements. By calibrating the sensor arrays, the three-dimensional spatial coordinates of the moving target are obtained based on the collected pulse sequence when detecting the moving target, meeting the requirements of large-field-of-view and high-speed spatial positioning of targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flow chart of a preferred embodiment of the method for spatial positioning of a moving target with a large field of view in the present invention.

[0055] Figure 2 It is a principle block diagram of a preferred embodiment of the spatial positioning method for a moving target with a large field of view in the present invention.

[0056] Figure 3 This is a specific flow chart of step S100 in a preferred embodiment of the method for spatial positioning of a moving target with a large field of view in the present invention.

[0057] Figure 4It is a specific flow chart of step S300 in a preferred embodiment of the method for spatial positioning of a moving target with a large field of view in the present invention.

[0058] Figure 5 This is a specific flow chart of step S320 in a preferred embodiment of the method for spatial positioning of a moving target with a large field of view in the present invention.

[0059] Figure 6 This is a specific flow chart of step S340 in a preferred embodiment of the method for spatial positioning of a moving target with a large field of view in the present invention.

[0060] Figure 7 This is a schematic diagram of the principle of three-dimensional coordinate calculation based on azimuth observation in a preferred embodiment of the spatial positioning method for a moving target with a large field of view in the present invention.

[0061] Figure 8 It is a functional principle block diagram of a preferred embodiment of the spatial positioning system for moving targets with a large field of view in the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solutions and advantages of the present invention more clear and distinct, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0063] To solve the problem of high-speed spatial positioning of targets with a large field of view, the present invention provides a method and system for high-speed spatial positioning of targets with a large field of view. The system uses two sets of dynamic visual sensor arrays to form a large observation field with the same coverage range, and realizes rapid spatial three-dimensional coordinate calculation by observing the target azimuth angle.

[0064] The present invention provides a method for spatially locating a moving target with a large field of view based on a spatial positioning system for moving targets with a large field of view. The system comprises a first sensor array, a second sensor array, a synchronous acquisition module connected to the first and second sensor arrays, and a terminal connected to the synchronous acquisition module. The first and second sensor arrays are both dynamic visual sensor arrays, and the terminal can be a computer.

[0065] See Figure 1 , Figure 1 FIG. 1 is a flow chart of the method for spatial positioning of a moving target with a large field of view in the present invention. Figure 1 As shown, the spatial positioning method for a moving target with a large field of view according to an embodiment of the present invention includes the following steps:

[0066] Step S100: calibrate the first sensor array and the second sensor array in advance to obtain calibration information.

[0067] Specifically, if Figure 2 As shown, the present invention includes two major steps: positioning system calibration and target positioning measurement. The present invention adopts a dynamic vision sensor, which is divided into a left sensor array and a right sensor array; the positioning system calibration is to calibrate the internal parameters of each dynamic vision sensor through a calibration image, calibrate the installation azimuth of each dynamic vision sensor through a rotating table and a marking point, and calibrate the center distance of the sensor array through a rangefinder or a rangefinder. The target positioning measurement is to first correct the pulse sequence distortion using the internal parameters of the sensor; then use the corrected pulse sequence to reconstruct the target image, binarize the reconstructed image, and calculate the target center coordinates; then determine the azimuth of the moving target relative to the center of the left and right sensor arrays through the target center coordinates and the calibrated sensor installation azimuth; finally, use the target azimuth and the calibrated sensor array center distance to calculate the spatial three-dimensional coordinates of the high-speed moving target.

[0068] In one embodiment, the first sensor array and the second sensor array have the same structure, and the two sensor arrays are spliced to form a large field of view. The field of view size and array structure (such as an annular structure, a hemispherical structure) can be designed according to specific application requirements, and the two sensor arrays are separated by a certain distance.

[0069] In one implementation, the calibration information includes: intrinsic parameter information, installation azimuth information, and the center distance between the first sensor array and the second sensor array. Figure 3 As shown, the step S100 specifically includes:

[0070] Step S110 : Acquire a calibration image, and perform intrinsic parameter calibration on each sensor in the first sensor array and the second sensor array according to the calibration image to obtain calibrated intrinsic parameter information.

[0071] In one embodiment, step S110 specifically includes: capturing multiple checkerboard images from different angles for each sensor in the first sensor array and the second sensor array, and using the checkerboard images as calibration images; calibrating the internal parameters of each sensor using the Zhang calibration method based on the calibration images to obtain calibrated internal parameter information. In other words, the internal parameters of the dynamic vision sensor are calibrated using the Zhang calibration method, and for each dynamic vision sensor, multiple checkerboard images are captured from different angles for calibration. If the dynamic vision sensor has a grayscale image output, the grayscale image can be directly captured for calibration; if the dynamic vision sensor only outputs asynchronous pulses, the image capture and calibration can be performed by displaying a checkerboard on an LCD screen. The calibration process is processed on the terminal.

[0072] The step S110 is followed by the following step S120: calibrating the installation azimuth of each sensor using a rotating platform and marking points to obtain calibrated installation azimuth information.

[0073] In one embodiment, step S120 specifically includes: pre-installing the first sensor array and the second sensor array on a turntable, and using the turntable to align the center of the image of each sensor with a preset mark point in turn; recording the rotation azimuth angle of the turntable when aligning with the preset mark point, and calibrating the installation azimuth angle of each sensor relative to the center of its respective array according to the rotation azimuth angle to obtain the calibrated installation azimuth angle information.

[0074] That is, the user manually installs the sensor array on a precision turntable and uses the turntable to align the center of each sensor image with a preset mark point, such as the corner point of a chessboard. By recording the turntable's rotation azimuth, the installation azimuth of each sensor relative to its own array center is calibrated. or

[0075] After step S120, the following occurs: step S130: measuring the center-to-center distance B between the first sensor array and the second sensor array using a distance ruler or a distance meter to obtain a center-to-center distance B. In other words, the center-to-center distance B between the first sensor array and the second sensor array is accurately measured using the distance ruler or the distance meter.

[0076] After step S100, the following occurs: step S200, the synchronous acquisition module acquires pulse sequences generated when the first sensor array and the second sensor array detect a moving target. In other words, the synchronous acquisition module implements synchronous triggering acquisition of data from the two sensor arrays.

[0077] After step S200, the terminal obtains the pulse sequence and obtains the three-dimensional coordinates of the moving target based on the calibration information and the pulse sequence. In other words, the collected data is sent to a terminal (such as a computer) to calculate the three-dimensional coordinates of the high-speed moving target.

[0078] In one implementation, Figure 4 As shown, the step S300 specifically includes:

[0079] Step S310: The terminal obtains the pulse sequence and corrects image distortion of the pulse sequence using the intrinsic parameter information;

[0080] Step S320: determining the first centroid coordinates of the moving target in the first sensor array and the second centroid coordinates of the moving target in the second sensor array according to the corrected pulse sequence;

[0081] Step S330: determining a first azimuth angle of the moving target relative to the center of the first sensor array and a second azimuth angle relative to the center of the second sensor array according to the first centroid coordinates, the second centroid coordinates, and the installation azimuth information;

[0082] Step S340: Calculate the three-dimensional coordinates of the moving target according to the first azimuth angle, the second azimuth angle, and the center distance.

[0083] Specifically, after acquiring the synchronous pulse sequence of the left and right sensor arrays, the image distortion of the pulse sequence is corrected using the sensor internal parameters obtained through calibration; based on the corrected pulse sequence, as well as the calibrated installation azimuth information and center spacing, the spatial three-dimensional coordinates of the moving target are finally calculated.

[0084] In one embodiment, if Figure 5 As shown, the step S320 specifically includes:

[0085] Step S321, segmenting the pulse sequence according to a preset time window, and reconstructing an image of the moving target;

[0086] Step S322: binarize the reconstructed image to obtain a target image;

[0087] Step S323 : Calculate the first centroid coordinates of the moving target in the first sensor array and the second centroid coordinates of the moving target in the second sensor array according to the target image.

[0088] Specifically, the pulse sequence is segmented according to the time window T and the image I(u,v) of the dynamic target is reconstructed. Among them, the image reconstruction method can be implemented by selecting methods such as event accumulation method and reconstruction network method. The event accumulation method is to sum the number of pulses corresponding to each pixel in the time period T to form an image; the reconstruction network method is to use a trained neural network, the input of the neural network is the pulse sequence, and the output is the reconstructed image. Then the reconstructed image is binarized to obtain image I b (u,v); finally use I b (u, v) calculates the first centroid coordinates of the moving target in the first sensor array and the second centroid coordinates in the second sensor array

[0089] In one implementation, both the first and second sensor arrays are dynamic vision sensor arrays, and both have identical fields of view and array structures. Dynamic vision sensors are a new type of biomimetic vision sensor that outputs pulsed data streams only in response to changes in scene light intensity. They offer advantages such as a large dynamic range, high temporal resolution, and lack of motion blur, making them ideal for high-speed target perception.

[0090] In step S330, the calculation formula of the first azimuth angle of the moving target relative to the center of the first sensor array is:

[0091]

[0092] The calculation formula of the second azimuth angle of the moving target relative to the center of the second sensor array is:

[0093]

[0094] in, is the coordinate of the first centroid; is the coordinate of the second centroid; (w,h) represents the lateral and longitudinal resolutions of the sensor array; (W,H) represents the horizontal and vertical fields of view of the corresponding sensor; represents the calibrated installation azimuth of the corresponding sensor in the first sensor array; Indicates the calibrated installation azimuth of the corresponding sensor in the second sensor array.

[0095] In one implementation, Figure 6 As shown, the step S340 specifically includes:

[0096] Step S341: Acquire a first straight line from the center point of the first sensor array to the moving target, and a second straight line from the center point of the second sensor array to the moving target;

[0097] Step S342: Calculate the midpoint of the perpendicular bisector between the first straight line and the second straight line to obtain the three-dimensional spatial coordinates of the moving target.

[0098] See also Figure 7 , assuming that the center points of the first sensor array and the second sensor array are O L and O R , with O L O R The midpoint O of the connecting line is the origin of the world coordinate system. L O R The distance between them is B, the spatial position of the high-speed moving target is P(x,y,z), and the azimuth angles of P obtained by the left and right sensor arrays are (φ L,θ L ) and (φ R ,θ R ); According to the spatial geometric relationship, by calculating the straight line and The three-dimensional coordinates of the dynamic target P are estimated using the midpoint of the perpendicular bisector method.

[0099] In one embodiment, the calculation formula for the spatial three-dimensional coordinates of the moving target is:

[0100]

[0101] in,

[0102]

[0103]

[0104]

[0105]

[0106] P (x, y, z) is the three-dimensional coordinate of the moving target; L is the center point of the first sensor array, the O R is the center point of the second sensor array; B is L With O R The center distance of .

[0107] The spatial positioning method for large-field-of-view moving targets proposed in the present invention provides a feasible way to locate high-speed moving targets with a large field of view. The dynamic visual sensor array used in the spatial positioning system for large-field-of-view moving targets only responds to moving targets, which not only avoids the disadvantage of blurred imaging of moving targets by ordinary cameras, but also greatly reduces the redundant data generated by static backgrounds. In practical applications, the amount of redundant data can be greatly reduced, saving computing resources and communication bandwidth; and three-dimensional positioning is achieved by independently determining the azimuth angles by two groups of sensor arrays, which not only avoids the disadvantages of a small binocular field of view and the need for image feature matching, but also has the characteristic of flexible and adjustable sensor array spacing.

[0108] Furthermore, if Figure 8 As shown, based on the above-mentioned method for spatial positioning of a moving target with a large field of view, the present invention also provides a spatial positioning system for a moving target with a large field of view, comprising:

[0109] The first sensor array 10 and the second sensor array 20 are used to simultaneously detect a moving target and generate a pulse sequence when a moving target is detected;

[0110] a synchronous acquisition module 30 connected to the first sensor array 10 and the second sensor array 20 and configured to acquire pulse sequences generated when the first sensor array 10 and the second sensor array 20 detect a moving target;

[0111] a terminal 40 connected to the synchronous acquisition module 30, configured to pre-calibrate the first sensor array 10 and the second sensor array 20 to obtain calibration information and the pulse sequence, and obtain the three-dimensional spatial coordinates of the moving target based on the calibration information and the pulse sequence;

[0112] The first sensor array and the second sensor array are both dynamic vision sensor arrays.

[0113] In one implementation, the first sensor array and the second sensor array are calibrated in advance to obtain calibration information, specifically including: acquiring a calibration image, and calibrating the intrinsic parameters of each sensor in the first sensor array and the second sensor array based on the calibration image to obtain calibrated intrinsic parameter information; calibrating the installation azimuth of each sensor using a rotating stage and marking points to obtain calibrated installation azimuth information; and measuring the center-to-center distance between the first sensor array and the second sensor array using a range ruler or a rangefinder to obtain the center-to-center distance.

[0114] In one embodiment, acquiring a calibration image, calibrating the intrinsic parameters of each sensor in the first sensor array and the second sensor array based on the calibration image, and obtaining calibrated intrinsic parameter information includes: capturing a plurality of checkerboard images of each sensor in the first sensor array and the second sensor array from different angles, and using the checkerboard images as calibration images; and calibrating the intrinsic parameters of each sensor using the Zhang calibration method based on the calibration images to obtain calibrated intrinsic parameter information.

[0115] In one embodiment, the installation azimuth of each sensor is calibrated by a rotating table and a marking point to obtain calibrated installation azimuth information, including: pre-installing the first sensor array and the second sensor array on a turntable, and using the turntable to align the center of the image of each sensor with the preset marking point in turn; recording the rotation azimuth of the turntable when aligning with the preset marking point, and calibrating the installation azimuth of each sensor relative to the center of its own array according to the rotation azimuth to obtain calibrated installation azimuth information.

[0116] In one embodiment, the terminal is also used to obtain the pulse sequence, and use the internal parameter information to correct the image distortion of the pulse sequence; determine the first center of mass coordinates of the moving target in the first sensor array and the second center of mass coordinates in the second sensor array based on the corrected pulse sequence; determine the first azimuth angle of the moving target relative to the center of the first sensor array and the second azimuth angle relative to the center of the second sensor array based on the first center of mass coordinates, the second center of mass coordinates and the installation azimuth information; and calculate the spatial three-dimensional coordinates of the moving target based on the first azimuth angle, the second azimuth angle and the center distance.

[0117] In one embodiment, the terminal is also used to segment the pulse sequence according to a preset time window and reconstruct an image of the moving target; binarize the reconstructed image to obtain a target image; and calculate the first center of mass coordinates of the moving target in the first sensor array and the second center of mass coordinates in the second sensor array based on the target image.

[0118] In one embodiment, the terminal is also used to obtain a first straight line from the center point of the first sensor array to the moving target, and a second straight line from the center point of the second sensor array to the moving target; and calculate the midpoint of the perpendicular bisector between the first straight line and the second straight line to obtain the spatial three-dimensional coordinates of the moving target.

[0119] The present invention also provides a computer-readable storage medium storing a computer program, wherein the computer program can be executed to implement the steps of the method for spatial positioning of a moving target with a large field of view as described above.

[0120] In summary, the present invention discloses a method and system for spatial positioning of a moving target with a large field of view. The method is implemented based on a spatial positioning system for a moving target with a large field of view. The spatial positioning system for a moving target with a large field of view includes: a first sensor array, a second sensor array, a synchronous acquisition module connected to the first sensor array and the second sensor array, and a terminal connected to the synchronous acquisition module; the first sensor array and the second sensor array are both dynamic visual sensor arrays; the method for spatial positioning of a moving target with a large field of view includes: pre-calibrating the first sensor array and the second sensor array to obtain calibration information; the synchronous acquisition module collects the pulse sequence generated when the first sensor array and the second sensor array detect a moving target; the terminal obtains the pulse sequence and obtains the spatial three-dimensional coordinates of the moving target based on the calibration information and the pulse sequence. The present invention forms a large field of view by splicing two sensor arrays. The field of view size and array structure can be designed according to specific application requirements. By calibrating the sensor array, when detecting a moving target, the spatial three-dimensional coordinates of the moving target are obtained based on the collected pulse sequence, thereby meeting the requirements of spatial positioning of a large field of view and high-speed target.

[0121] It should be understood that the application of the present invention is not limited to the above examples. For those skilled in the art, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. A method for spatial positioning of a moving target with a large field of view, wherein the method is implemented based on a spatial positioning system for a moving target with a large field of view, and is characterized in that: The spatial positioning system for a moving target with a large field of view comprises: a first sensor array, a second sensor array, a synchronous acquisition module connected to the first sensor array and the second sensor array, and a terminal connected to the synchronous acquisition module; the first sensor array and the second sensor array are both dynamic visual sensor arrays; The spatial positioning method of the moving target includes: Pre-calibrating the first sensor array and the second sensor array to obtain calibration information; The synchronous acquisition module acquires a pulse sequence generated when the first sensor array and the second sensor array detect a moving target; The terminal obtains the pulse sequence, and obtains the spatial three-dimensional coordinates of the moving target according to the calibration information and the pulse sequence; The pre-calibrating the first sensor array and the second sensor array to obtain calibration information includes: Acquire a calibration image, and perform intrinsic parameter calibration on each sensor in the first sensor array and the second sensor array according to the calibration image to obtain calibrated intrinsic parameter information; The installation azimuth angle of each sensor is calibrated by using a rotating table and marking points to obtain the calibrated installation azimuth angle information; Measuring the center distance between the first sensor array and the second sensor array by using a distance ruler or a distance meter to obtain the center distance; The terminal obtains the pulse sequence, and obtains the spatial three-dimensional coordinates of the moving target according to the calibration information and the pulse sequence, including: The terminal obtains the pulse sequence and corrects image distortion of the pulse sequence using the intrinsic parameter information; Determine the first centroid coordinates of the moving target in the first sensor array and the second centroid coordinates in the second sensor array according to the corrected pulse sequence; Determining a first azimuth angle of the moving target relative to a center of a first sensor array and a second azimuth angle of the moving target relative to a center of a second sensor array according to the first centroid coordinates, the second centroid coordinates, and the installation azimuth information; The spatial three-dimensional coordinates of the moving target are calculated according to the first azimuth angle, the second azimuth angle and the center distance.

2. The spatial positioning method for a moving target with a large field of view according to claim 1, characterized in that: The acquiring of the calibration image, and performing intrinsic parameter calibration on each sensor in the first sensor array and the second sensor array according to the calibration image to obtain calibrated intrinsic parameter information includes: Capturing a plurality of checkerboard images from different angles for each sensor in the first sensor array and the second sensor array, and using the checkerboard images as calibration images; The intrinsic parameters of each sensor are calibrated using Zhang's calibration method according to the calibration image to obtain calibrated intrinsic parameter information.

3. The spatial positioning method for a moving target with a large field of view according to claim 1, characterized in that: The installation azimuth of each sensor is calibrated by the rotating table and the marking point to obtain the calibrated installation azimuth information, including: The first sensor array and the second sensor array are pre-mounted on a turntable, and the center of each sensor's image is aligned with a preset mark point in sequence using the turntable; The rotation azimuth of the turntable when aligning with the preset mark point is recorded, and the installation azimuth of each sensor relative to the center of each array is calibrated according to the rotation azimuth to obtain calibrated installation azimuth information.

4. The spatial positioning method for a moving target with a large field of view according to claim 1, characterized in that: The step of determining the first centroid coordinates of the moving target in the first sensor array and the second centroid coordinates of the moving target in the second sensor array according to the corrected pulse sequence includes: Segmenting the pulse sequence according to a preset time window and reconstructing an image of the moving target; Binarize the reconstructed image to obtain the target image; The first centroid coordinates of the moving target in the first sensor array and the second centroid coordinates of the moving target in the second sensor array are calculated based on the target image.

5. The spatial positioning method for a moving target with a large field of view according to claim 1, characterized in that: The first sensor array and the second sensor array have the same structure and the same large field of view.

6. The spatial positioning method for a moving target with a large field of view according to claim 5, characterized in that: The calculation formula of the first azimuth angle of the moving target relative to the center of the first sensor array is: ； The calculation formula of the second azimuth angle of the moving target relative to the center of the second sensor array is: ； in, is the coordinate of the first centroid; is the coordinate of the second centroid; Indicates the lateral and longitudinal resolution of the sensor array; Indicates the horizontal and vertical fields of view of the corresponding sensor; represents the calibrated installation azimuth of the corresponding sensor in the first sensor array; Indicates the calibrated installation azimuth of the corresponding sensor in the second sensor array.

7. The method for spatial positioning of a moving target with a large field of view according to claim 6, characterized in that: Calculating the spatial three-dimensional coordinates of the moving target according to the first azimuth angle, the second azimuth angle, and the center distance includes: Acquire a first straight line from the center point of the first sensor array to the moving target, and a second straight line from the center point of the second sensor array to the moving target; The midpoint of the perpendicular bisector between the first straight line and the second straight line is calculated to obtain the three-dimensional spatial coordinates of the moving target.

8. The method for spatial positioning of a moving target according to claim 7, wherein: The calculation formula of the spatial three-dimensional coordinates of the moving target is: ; in, ; ; ； ； ； ； is the spatial three-dimensional coordinate of the moving target; is the center point of the first sensor array, is the center point of the second sensor array; for and The center distance of .

9. A spatial positioning system for a moving target with a large field of view, characterized in that: include: The first sensor array and the second sensor array are used to simultaneously detect a moving target and generate a pulse sequence when a moving target is detected; a synchronous acquisition module, connected to the first sensor array and the second sensor array, for acquiring a pulse sequence generated when the first sensor array and the second sensor array detect a moving target; a terminal connected to the synchronous acquisition module, configured to pre-calibrate the first sensor array and the second sensor array to obtain calibration information, and obtain the pulse sequence, and obtain the three-dimensional spatial coordinates of the moving target based on the calibration information and the pulse sequence; The first sensor array and the second sensor array are both dynamic vision sensor arrays; The pre-calibrating the first sensor array and the second sensor array to obtain calibration information includes: Acquire a calibration image, and perform intrinsic parameter calibration on each sensor in the first sensor array and the second sensor array according to the calibration image to obtain calibrated intrinsic parameter information; The installation azimuth angle of each sensor is calibrated by using a rotating table and marking points to obtain the calibrated installation azimuth angle information; Measuring the center distance between the first sensor array and the second sensor array by using a distance ruler or a distance meter to obtain the center distance; Obtaining the pulse sequence, and obtaining the spatial three-dimensional coordinates of the moving target according to the calibration information and the pulse sequence, including: The terminal obtains the pulse sequence and corrects image distortion of the pulse sequence using the intrinsic parameter information; Determine the first centroid coordinates of the moving target in the first sensor array and the second centroid coordinates in the second sensor array according to the corrected pulse sequence; Determining a first azimuth angle of the moving target relative to a center of a first sensor array and a second azimuth angle of the moving target relative to a center of a second sensor array according to the first centroid coordinates, the second centroid coordinates, and the installation azimuth information; The spatial three-dimensional coordinates of the moving target are calculated according to the first azimuth angle, the second azimuth angle and the center distance.

Images