A method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement

Abstract

This invention relates to a method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement, comprising the following steps: camera calibration of the high-speed camera to obtain camera parameters and scale factor; stereo matching of the initial static frames of the binocular images obtained by the high-speed camera under static conditions; identification of occluded targets in the left and right image sequences using an affine transformation-based occlusion target detection algorithm; image sequence matching based on the occlusion target detection results, performing binocular stereo-sequence dual matching for unoccluded speckles, and monocular sequence matching for speckles occluded only in one camera view; and, based on the image sequence matching results, sequentially performing coordinate calculation, coordinate system correction, information fusion, displacement and acceleration calculation, and impact response spectrum calculation to obtain the impact response spectrum. Compared with existing technologies, this invention has advantages such as accurate impact response spectrum calculation results under occlusion conditions.

Description

Technical Field

[0001] This invention relates to the field of high-speed video measurement technology, and in particular to a method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement. Background Technology

[0002] Shock environments are widespread in both military and civilian product fields, such as armor penetration and interception during missile attacks, stage separation during rocket launches, bird strikes during aircraft flight, drop tests of electronic products, and crash tests of automobiles. Shock response spectrum (SRS) testing is a crucial method for evaluating the performance and reliability of products under impact. Traditional measurement methods use contact sensors to collect impact signals and obtain the shock response spectrum to assess the degree of damage. However, shock environments are characterized by high frequency, transients, and high amplitudes, which can easily damage contact sensors. Furthermore, contact sensors suffer from installation difficulties and single-point measurement limitations. Additionally, when meeting the needs of large-scale measurements, installing numerous contact sensors can alter the structural characteristics of the target object, introducing sensor mass effects. High-speed video measurement technology offers advantages such as non-contact operation, high frame rate, three-dimensional measurement, and large-scale monitoring. It utilizes high-speed cameras to record the transient changes of the target object and accurately measures the three-dimensional coordinates of target feature points using computer vision and photogrammetric analytical methods, thereby obtaining the dynamic parameters of the object and providing a new means for shock environment testing.

[0003] However, the application of high-speed video measurement in impact testing also faces certain challenges. High-impact environments often generate obstructions, such as spatter in non-pyrotechnic impact tests and debris from explosions in pyrotechnic impact tests. When these obstructions appear in the camera's field of view, they significantly impact the performance and accuracy of image processing algorithms, leading to substantial deviations in the impact response spectrum measurement results. Summary of the Invention

[0004] The purpose of this invention is to address the problem of measuring the impact response spectrum of target points under occlusion conditions. Considering the morphological differences that arise when observing the same target using multi-view imaging technology, this invention provides an impact response spectrum calculation method based on hybrid monocular and binocular high-speed video measurement. Since the morphological information of an occluded target cannot be fully captured at a specific moment or from a single viewpoint, integrating observation information from different viewpoints can reveal the complete dynamic change process. Therefore, this invention employs a digital image correlation speckle matching method to match targets from different viewpoints in time and space. Then, combined with the occluded target extracted by a target detection algorithm, a hybrid monocular and binocular measurement strategy is proposed. Furthermore, a scale correction method based on coordinate system transformation is used to reconstruct the temporal displacement response of the measured target point under occlusion conditions, thereby obtaining a reliable impact response spectrum.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement includes the following steps:

[0007] Camera calibration: The high-speed camera is calibrated using Zhang Zhengyou's planar target calibration method to obtain camera parameters and scale factor;

[0008] Stereo matching of static images: Stereo matching of initial static frames of binocular images obtained by a high-speed camera is performed under static conditions using the digital image correlation speckle matching method.

[0009] Occlusion target detection: An occlusion target detection algorithm combining affine transformation is used to identify occlusion targets in the left and right image sequences;

[0010] Image sequence matching: Based on the occluded target detection results, image sequence matching is performed through an occlusion condition discrimination strategy. For unoccluded speckle patterns, the digital image correlation speckle matching method is used to perform stereo-sequence dual matching in the stereo sequence images, combined with camera parameters. For speckle patterns that are occluded only in one camera view, the digital image correlation speckle matching method is used to perform monocular sequence matching, combined with a scale factor.

[0011] Impact response spectrum calculation: Based on the image sequence matching results, the physical coordinates of the measurement points are reconstructed, and the coordinate system of the images taken from multiple perspectives is corrected. The corrected coordinates are then fused. The displacement of the speckle is determined based on the coordinate changes in the fusion results, and the acceleration is calculated. Based on the calculated acceleration, the impact response spectrum is calculated using the impact response spectrum calculation algorithm.

[0012] The digital image correlation speckle matching method is specifically as follows:

[0013] Digital image correlation algorithm is used to match speckles in images acquired by different high-speed cameras to obtain the correspondence of the same speckle in different acquisition times and imaging spaces. During the matching process, the first image of a selected camera is always used as the reference image. The region of interest is defined in the reference image, and the size and step of the matching window are set with a certain pixel coordinate as the center. Coarse matching and fine matching are used to match the image sequence in turn to determine the corresponding points.

[0014] The coarse matching uses normalized cross-correlation to obtain integer pixel shifts, as shown in the following formula:

[0015]

[0016] Among them, f m and g mf(x,y) and f(x,y) represent the average gray values ​​of the reference subset and the target subset, respectively. Representing coordinates (x, y) and The gray value at the specified location, M represents the half-step size of the matching window, and C represents the correlation coefficient; the reference subset and the target subset are determined according to the execution object of the digital image correlation speckle matching method;

[0017] Select the region with the highest correlation coefficient and use the corresponding pixel offset as the initial deformation parameter;

[0018] The fine matching process uses a first-order deformation function to describe the deformation of the reference region relative to the target region, i.e., sub-pixel deformation, based on a preset matching window size. The first-order deformation function is shown in the following equation:

[0019]

[0020] Where u and v are the displacement components of the center point P(x0,y0) in the x and y directions, respectively, and Δx and Δy are the displacements from the center point P to point P. i The distance u between (x, y) x u y v x and v y These represent displacement gradients, Here are the two-dimensional coordinates corresponding to the center point P;

[0021] The optimal correlation coefficient is obtained through iterative optimization in order to solve for the fine sub-pixel deformation parameters p = [u, u] x ,u y ,v,v x ,v y ].

[0022] The optimal correlation coefficient is obtained using the iterative normalized least squares method, and the criteria for the iterative normalized least squares method are as follows:

[0023]

[0024] The occlusion target detection specifically includes:

[0025] When extracting the pixel coordinates of occluded targets in an image sequence, the first frame of each camera is used as the reference image.

[0026] Unobstructed speckle patterns are selected in the reference image. The corresponding points of the selected speckle patterns in each frame of the image sequence are obtained using the digital image correlation speckle matching method. The transformation relationship between the image sequence and the reference image is established based on the affine transformation model of the following formula:

[0027] x'=Ax+By+t x

[0028] y'=Cx+Dy+t y

[0029] Where (x,y) are the coordinates of the speckle in the reference image, (x',y') are the coordinates of the corresponding point in the image sequence, A, B, C, and D are parameters controlling rotation, scaling, and cropping, and t x and t y It is a translation amount, and its affine transformation relationship is obtained through at least 3 pairs of coordinate points;

[0030] The image sequence is subjected to affine transformation using an affine transformation model to generate an affine transformed image, and the difference image between the affine transformed image and the original image sequence is obtained.

[0031] Occlusions in the difference image are extracted using the independent multimodal BGS algorithm from the BGS library, forming a set of occluded pixels.

[0032] The image sequence matching includes the following steps:

[0033] Select a search window centered on the target point to be measured, the size of which is the same as the size of the matching window;

[0034] An occlusion condition discrimination strategy is implemented to determine the occlusion status in the image at each time step. When discrimination is completed at all time steps or the early termination condition is met, the corresponding matching is performed based on the discrimination results. Specifically:

[0035] If the point to be measured is occluded in the search window of the image sequence captured by different cameras at a certain time, the early termination condition is met, the identifier of the point to be measured is set to 0, and the data of the point to be measured is discarded.

[0036] If the point to be measured is not occluded in the search window of the image sequence captured by different cameras at all times, then the identifier of the point to be measured is set to 1, and a stereo-sequence double matching is performed on the binocular measurement result of the point to be measured. The stereo-sequence double matching is as follows: first, sequence matching is performed on the monocular image sequence of a single camera, and after the matching is completed, stereo matching is performed on the binocular image sequences of different cameras.

[0037] If the point to be measured is not occluded in the search window of the image sequence taken by one camera at all times, but is occluded in the search window of the image sequence taken by another camera, then monocular sequence matching is performed based on the monocular measurement results of the camera that is not occluded. If the camera that is not occluded is the left camera, then the identifier of the point to be measured is set to 2; if it is the right camera, then the identifier of the point to be measured is set to 3.

[0038] If none of the above conditions are met, and at any given moment in the entire image sequence of the point to be measured, there is occlusion in the search window of at most one image sequence taken by a camera, then the identifier of the point to be measured is set to 4. For moments when there is no occlusion in the search windows of image sequences taken by different cameras, the binocular measurement results are used for binocular stereo-sequence matching. For moments when there is occlusion in the search window of an image sequence taken by one camera, the unoccluded monocular measurement results are used for monocular sequence matching.

[0039] The physical coordinates of the reconstructed measurement points are specifically as follows:

[0040] Based on the image sequence matching results, the actual physical coordinates of the measurement points in the world coordinate system are reconstructed using a spatial mapping operator. The spatial mapping operator includes three-dimensional reconstruction and a scale factor, which is expressed as the ratio of the three-dimensional world coordinates to the two-dimensional image coordinates within the measurement area on the object surface.

[0041] For speckle with marker 1, its three-dimensional coordinates are obtained through three-dimensional reconstruction technology. For speckle with markers 2, 3, and 4, the coordinates in the main impact direction are obtained by calculating the scale factor in the reference image.

[0042] The coordinate system correction of images captured from multiple perspectives specifically involves adjusting the world coordinate system O. w Perform correction:

[0043] The world coordinate system O w The X and Y axes are parallel to the planar target, and the Z axis is perpendicular to the XOY plane. Assuming the X-axis approximates the main impact direction, the world coordinate system O... w The correction consists of the following two steps:

[0044] Step 1 calibration: Rotate the XOZ plane around the Y-axis so that the X-axis is parallel to the object surface. By selecting two three-dimensional points (X1,Y1,Z1) and (X2,Y2,Z2) within the measurement area, calculate the angle θ between the vectors (X1,Z1) and (X2,Z2) on the XOZ plane.

[0045] The second step of correction is to rotate the XOY plane around the Z-axis to align the X-axis with the main impact direction. A straight line extraction algorithm is then used to obtain horizontal or vertical straight lines on the object's structure. Two three-dimensional points (X3, Y3, Z3) and (X4, Y4, Z4) are selected within the measurement area. The straight line formed by these two points is parallel to the extracted straight line, thus obtaining the angle β between the vectors (X3, Y3) and (X4, Y4) on the XOY plane. The corrected coordinates (X', Y', Z') are then represented as follows:

[0046]

[0047] Where (X,Y,Z) are the coordinates before correction.

[0048] The shock response spectrum calculation algorithm performs the following steps:

[0049] According to Newton's laws, the equations of motion for the mass block in a single-degree-of-freedom system are as follows:

[0050]

[0051] Where m, c, and k are the mass, damping coefficient, and stiffness of the SDOF, respectively, and f n Let x and y be the response displacement of the mass block and the displacement of the base, respectively, given its natural frequency. These are the velocity response of the mass block and the velocity of the base, respectively. The acceleration response at time t under this impact signal;

[0052] Expressing the relative displacement of the mass block with respect to the base as z = xy, we can derive:

[0053]

[0054] in, To input the impact excitation acceleration signal for this single-degree-of-freedom system, the damping coefficient is c = 2mξω. n stiffness The equation relating the impact response spectrum is then derived:

[0055]

[0056] Where ξ is the damping ratio, ω n The natural angular frequency is represented by ω. n =2πf n f n It is the natural frequency;

[0057] The relationship between the natural frequency and the peak value of the impact response of the single-degree-of-freedom system is obtained by solving the impact response spectrum equation using the digital recursive filtering method, thereby obtaining the impact response spectrum of the complex real physical system.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] This invention addresses the problem of unreliable impact response spectra when the target point is obscured by a high-magnitude impact during impact testing, leading to unreliable measured impact response spectra. It proposes a hybrid single / binocular high-speed video measurement method to calculate the impact response spectrum. A digital image correlation speckle matching method is used to obtain the spatiotemporal correspondence of the measurement point in different camera imaging views. Then, obscured target detection combined with affine transformation is used to obtain the obscured pixel set. Finally, by employing a hybrid single / binocular high-speed video measurement strategy and a scale correction method based on coordinate system transformation, the information of the target point from multiple viewing views is fused to reconstruct the dynamic change process of the obscured target point and obtain accurate impact response spectrum results. The reliability of this method is verified by comparing it with impact response spectrum results obtained based on accelerometers. This invention significantly improves the feasibility of applying high-speed video measurement technology in high-magnitude impact tests under obstructed conditions and has significant practical value. Attached Figure Description

[0060] Figure 1 This is a flowchart of the method of the present invention;

[0061] Figure 2 This is a schematic diagram of the standard impact response spectrum;

[0062] Figure 3 This is a schematic diagram of the left and right camera image sequence in one embodiment;

[0063] Figure 4 This is a schematic diagram of the occlusion target detection result in one embodiment;

[0064] Figure 5 This is a comparison chart of the impact response spectrum calculation results of different measurement methods when the speckle pattern is blocked in one embodiment. Detailed Implementation

[0065] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0066] Accurate impact response analysis of structures under complex impact environments is crucial. Contact sensor methods suffer from drawbacks such as installation difficulties, single-point measurement, and susceptibility to damage during impact. Non-contact high-speed video measurement methods offer advantages such as high frame rate, three-dimensional measurement, and large-area monitoring, making them highly promising for impact testing. However, high-volume impact tests are often accompanied by obstructions such as debris, significantly impacting the accuracy of video measurements. To address the issue of missing target dynamic information caused by obstructions during testing, this embodiment proposes an impact response spectrum calculation method based on hybrid mono / binocular high-speed video measurement, based on the concept of multi-viewpoint sequential image information transmission. It primarily utilizes a digital image correlation-based matching method to achieve spatiotemporal information matching of the target, then employs a target detection algorithm to determine the obstruction status of the target point, and finally combines a hybrid mono / binocular high-speed video measurement strategy and a scale correction method to reconstruct the dynamic response process of the target point under obstruction conditions. The results of the air gun impact test show that, compared with the impact response spectrum reference value obtained based on the accelerometer, the impact response spectrum obtained by the proposed method under the condition that the target point is blocked has good consistency with the reference result, with an average acceleration error of 12.32%, which verifies the feasibility and effectiveness of the proposed method.

[0067] Specifically, such as Figure 1 As shown, the method includes the following steps:

[0068] Step 1) Camera calibration: Use Zhang Zhengyou's planar target calibration method to calibrate the high-speed camera to obtain camera parameters (intrinsic parameters, extrinsic parameters and distortion parameters) and scale factor.

[0069] Step 2) Static image stereo matching: Under static conditions, the initial static frames of the binocular images obtained by the high-speed camera are stereo matched using the digital image correlation speckle matching method.

[0070] In this embodiment, there are at least two high-speed cameras. When performing image matching, if there are only two high-speed cameras, the images captured by the two cameras are matched directly. If there are more than two high-speed cameras, the images are matched in pairs, and then the matching results are fused.

[0071] The digital image correlation speckle matching method is as follows: a digital image correlation algorithm is used to match speckles in images acquired by different high-speed cameras to obtain the correspondence of the same speckle in different acquisition times and imaging spaces; during the matching process, the first image of a selected camera (assumed to be the left camera in this embodiment) is always used as the reference image, a region of interest (ROI) is defined in the reference image, and the size and step size of the matching window are set with a certain pixel coordinate as the center. Coarse matching and fine matching are used to match the image sequence in turn to determine the corresponding points.

[0072] In coarse matching, normalized cross-correlation (NCC) is used to obtain integer pixel displacements, as shown in the following formula:

[0073]

[0074] Among them, f m and g m f(x,y) and f(x,y) represent the average gray values ​​of the reference subset and the target subset, respectively. Representing coordinates (x, y) and The grayscale value at the specified location, M represents the half-step size of the matching window, and C represents the correlation coefficient. The reference subset and target subset are determined based on the execution object of the digital image correlation speckle matching method.

[0075] Then, select the region with the highest correlation coefficient and use the corresponding pixel offset as the initial deformation parameter.

[0076] In fine matching, based on the preset matching window size, a first-order deformation function is used to describe the deformation of the reference region relative to the target region, i.e., sub-pixel deformation. The first-order deformation function is shown in the following equation:

[0077]

[0078] Where u and v are the displacement components of the center point P(x0,y0) in the x and y directions, respectively, and Δx and Δy are the displacements from the center point P to point P. i The distance u between (x, y) x u y v x and v y These represent displacement gradients, Let P be the two-dimensional coordinates of the center point.

[0079] The optimal correlation coefficient is obtained through iterative optimization in order to solve for the fine sub-pixel deformation parameters p = [u, u] x ,u y ,v,v x ,vy ].

[0080] In this embodiment, the optimal correlation coefficient is obtained using the iterative normalized least squares method, wherein the criteria for the iterative normalized least squares method are as follows:

[0081]

[0082] The digital image correlation speckle matching method is the focus of this invention, and it is also used for image sequence matching in step 4). The difference lies in the determined reference subset and target subset: the matching method used in static image stereo matching and image sequence matching is the same, but the objects targeted are different. For example, static image stereo matching targets the initial static frame of the binocular image; in image sequence matching, if there is no occlusion, the object is the binocular image, and the sequence matching of the monocular image is performed first, followed by the stereo matching of the binocular image; if there is occlusion, only the sequence matching of the monocular image is performed. That is, different reference subsets and target subsets can be determined according to the different objects to be matched, so that the digital image correlation speckle matching method can be applied.

[0083] Step 3) Occlusion Target Detection: Use an occlusion target detection algorithm that combines affine transformation to identify occlusion targets in the left and right image sequences.

[0084] Step 31) When extracting the pixel coordinates of the occluded target in the image sequence, the first frame in each camera is used as the reference image.

[0085] Step 32) Select unobstructed speckle patterns in the reference image, and use the digital image correlation speckle matching method to obtain the corresponding points of the selected speckle patterns in each frame of the image sequence. Establish the transformation relationship between the image sequence and the reference image based on the affine transformation model of the following formula:

[0086] x'=Ax+By+t x

[0087] y'=Cx+Dy+t y

[0088] Where (x,y) are the coordinates of the speckle in the reference image, (x',y') are the coordinates of the corresponding point in the image sequence, A, B, C, and D are parameters controlling rotation, scaling, and cropping, and t x and t y It is a translation, and its affine transformation relationship is obtained through at least 3 pairs of coordinate points.

[0089] Step 33) Perform an affine transformation on the image sequence using an affine transformation model to generate an affine transformed image, and obtain the difference image between the affine transformed image and the original image sequence.

[0090] Step 34) Extract occlusions (manually marked) from the differential image using the independent multimodal BGS algorithm in the BGS library to form an occlusion pixel set.

[0091] Step 4) Image sequence matching: Based on the occlusion target detection results, image sequence matching is performed through the occlusion condition discrimination strategy. For unoccluded speckles, the digital image correlation speckle matching method is used to perform stereo-sequence dual matching in the stereo sequence images, combined with camera parameters. For speckles that are occluded only in one camera view, the digital image correlation speckle matching method is used to perform monocular sequence matching, combined with the scale factor.

[0092] After obtaining the set of occluded pixels in the image sequence, unreliable speckles in the ROI need to be removed before image sequence matching, and then image sequence matching is performed, which includes the following steps:

[0093] Step 41) Select a search window centered on the target point to be measured. The size of the search window is the same as the size of the matching window.

[0094] Step 42) Execute the occlusion condition discrimination strategy. At each time step, discriminate the occlusion status in the image. When discrimination is completed at all time steps or the early termination condition is met, perform the corresponding matching based on the discrimination results. Specifically:

[0095] If the point to be measured is occluded in the search windows of image sequences captured by different cameras at a certain time, that is, the search windows (i.e., ROIs) of the point to be measured in the left and right cameras at any given time are all occluded. L and ROI R If there are occluded pixels in the test point, the early termination condition is met, the identifier of the test point is set to 0, and the data of the test point is discarded.

[0096] If the point to be measured is not occluded in the search window of the image sequences taken by different cameras at all times, then the identifier of the point to be measured is set to 1, and stereo-sequence double matching is performed on the binocular measurement results of the point to be measured. The stereo-sequence double matching is as follows: first, sequence matching is performed on the monocular image sequence of a single camera, and after the matching is completed, stereo matching is performed on the binocular image sequences of different cameras.

[0097] If the point to be measured is not obstructed in the search window of the image sequence taken by one camera at all times, but is obstructed in the search window of the image sequence taken by another camera, then monocular sequence matching is performed based on the monocular measurement results of the camera that is not obstructed. If the camera that is not obstructed is the left camera, the identifier of the point to be measured is set to 2; if it is the right camera, the identifier of the point to be measured is set to 3.

[0098] If none of the above conditions are met, and at any given moment in the entire image sequence of the point to be measured, there is occlusion in the search window of at most one image sequence taken by a camera, then the identifier of the point to be measured is set to 4. For moments when there is no occlusion in the search windows of image sequences taken by different cameras, the binocular measurement results are used for binocular stereo-sequence matching. For moments when there is occlusion in the search window of an image sequence taken by one camera, the unoccluded monocular measurement results are used for monocular sequence matching.

[0099] The above occlusion condition discrimination strategy can be expressed as:

[0100]

[0101] in, Indicates the point to be measured, P. n identifier, and This represents the search window for the point to be measured across the entire image sequence in both the left and right images, where O is the set of occluded pixels. and This represents the search window for the i-th frame of the left and right images where the point to be measured is located.

[0102] Step 5) Impact response spectrum calculation: Based on the image sequence matching results, reconstruct the physical coordinates of the measurement points, and perform coordinate system correction on the images taken from multiple perspectives, and then perform information fusion on the corrected coordinates; determine the displacement of the speckle according to the coordinate changes in the fusion results, and calculate the acceleration; based on the calculated acceleration, calculate the impact response spectrum using the impact response spectrum calculation algorithm.

[0103] Step 51) Reconstruct the physical coordinates of the measurement points

[0104] Based on the image sequence matching results, the actual physical coordinates of the measurement points in the world coordinate system are reconstructed using a spatial mapping operator. This spatial mapping operator includes three-dimensional reconstruction and a scale factor, where the scale factor is expressed as the ratio of the three-dimensional world coordinates to the two-dimensional image coordinates within the measurement area of ​​the object surface.

[0105] For speckle with marker 1, its three-dimensional coordinates are obtained through three-dimensional reconstruction technology. For speckle with markers 2, 3, and 4, the coordinates in the main impact direction are obtained by calculating the scale factor in the reference image.

[0106] Step 52) Coordinate system correction

[0107] Because the world coordinate system O is established based on the planar target w The coordinate axes may not be consistent with the main impact direction, therefore it is necessary to adjust the world coordinate system O. w Perform corrections.

[0108] During the usual calibration process, the world coordinate system O w The X and Y axes are parallel to the planar target, and the Z axis is perpendicular to the XOY plane. Assuming the X-axis approximates the main impact direction, the world coordinate system O... w The correction consists of the following two steps:

[0109] Step 1 calibration: Rotate the XOZ plane around the Y-axis so that the X-axis is parallel to the object surface. By selecting two three-dimensional points (X1,Y1,Z1) and (X2,Y2,Z2) within the measurement area, calculate the angle θ between the vectors (X1,Z1) and (X2,Z2) on the XOZ plane.

[0110] The second step of correction is to rotate the XOY plane around the Z-axis to align the X-axis with the main impact direction. A straight line extraction algorithm is then used to obtain horizontal or vertical straight lines on the object's structure. Two three-dimensional points (X3, Y3, Z3) and (X4, Y4, Z4) are selected within the measurement area. The straight line formed by these two points is parallel to the extracted straight line, thus obtaining the angle β between the vectors (X3, Y3) and (X4, Y4) on the XOY plane. The corrected coordinates (X', Y', Z') are then represented as follows:

[0111]

[0112] Where (X,Y,Z) are the coordinates before correction.

[0113] Step 53) Information Fusion

[0114] The three-dimensional coordinates of each measurement point after coordinate system correction are fused to obtain the overall coordinate information of the surface of the object to be measured.

[0115] Step 54) Displacement and acceleration calculation

[0116] Based on the results of the fusion in step 53), the existing displacement and acceleration calculation methods are used to calculate the displacement and acceleration. This embodiment will not elaborate on this calculation process.

[0117] Step 55) Calculation of impact response spectrum

[0118] The shock response spectrum is a calculation function based on acceleration time history. Its basic theory is to apply a time-domain acceleration shock excitation to a real physical system to calculate the shock response caused by the shock input. By decomposing the complex real physical system into a series of single-degree-of-freedom (SDOF) systems with different natural frequencies, each SDOF system has a unique time history response to a given basic shock input. The maximum value of the response for each system is obtained, and the results for each SDOF are plotted to obtain the shock response spectrum. Therefore, the shock response spectrum is the absolute peak acceleration response of each SDOF system to a time-domain load input, with the horizontal axis representing the natural frequency of each SDOF and the vertical axis representing its corresponding absolute peak acceleration response value. The shock response spectrum can be used to assess the damage boundary of a given component and the maximum dynamic load that a device can withstand. A standard shock response spectrum is shown below. Figure 2 As shown, it consists of three elements: low-frequency slope, inflection point frequency, and high-frequency amplitude.

[0119] Taking a certain SDOF as an example, according to Newton's laws, the equation of motion for the mass block in a single-degree-of-freedom system is:

[0120]

[0121] Where m, c, and k are the mass, damping coefficient, and stiffness of the SDOF, respectively, and f n Let x and y be the response displacement of the mass block and the displacement of the base, respectively, given its natural frequency. These are the velocity response of the mass block and the velocity of the base, respectively. Let be the acceleration response at time t under this impact signal (the point represents the derivative with respect to time).

[0122] Expressing the relative displacement of the mass block with respect to the base as z = xy, we can derive:

[0123]

[0124] in, To input the impact excitation acceleration signal for this single-degree-of-freedom system, the damping coefficient is c = 2mξω. n stiffness The equation relating the impact response spectrum is then derived:

[0125]

[0126] Where ξ is the damping ratio (usually set to 0.05), ω n The natural angular frequency is represented by ω. n =2πf n f n This is the natural frequency.

[0127] By solving the impulse response spectrum relation equation, the relationship between the natural frequency and the peak value of the impulse response of the single-degree-of-freedom system can be obtained, thereby acquiring the impulse response spectrum of a complex real physical system.

[0128] The impulse response spectrum equation is typically solved using the digital recursive filtering method proposed by Smallwood. The Smallwood method uses a filter to simulate the SDOF system, treating the output signal after passing the time-domain signal through the filter as a single-degree-of-freedom impulse response. It has the advantages of simplicity, clarity, and high computational accuracy. The final expression of the Smallwood method is shown below:

[0129]

[0130] p0=1-θsin(ω d Δt) / (ω d Δt)

[0131] p1=2θ(sin(ω d Δt)-cos(ω d Δt))

[0132] p2=θ(θ-sin(ω d Δt) / (ω d Δt))

[0133] q1=2θcos(ω d Δt)

[0134]

[0135] Where Δt represents the unit time interval. Let be the acceleration response at time t. Let t be the impact excitation acceleration signal input to the single-degree-of-freedom system, where p0, p1, p2, q1, and q2 are coefficient terms, and ω is the acceleration signal. n Let ξ represent the natural angular frequency, and ξ be the damping ratio.

[0136] This embodiment verifies the usability of the proposed shock response spectrum calculation method based on hybrid single / binocular high-speed video measurement under obstruction conditions using an impact test conducted on an air gun impact test bench. The measurement object is a T-shaped tooling plate of uniform material, approximately 20 cm × 20 cm in size. A speckle pattern and circular markers are arranged on the surface of the T-shaped plate, and an accelerometer with a sampling frequency of 100 kHz is placed there. The high-speed camera used is an Acuteye AE-1-M-3500 with a maximum resolution of 1280 × 860 pixels. In the experiment, the two cameras were positioned approximately 2 meters apart, with a resolution set to 320 × 320 pixels. Each camera was equipped with a 35 mm fixed-focus lens, and the frame rate was set to 9000 Hz. During the experiment, the high-magnitude impact caused the circular markers on the surface of the T-shaped plate to detach, obstructing the target point being measured.

[0137] The reliability of the invention was verified by measuring the speckle pattern under occlusion conditions and comparing the impact response spectrum obtained by the present invention with the impact response spectrum obtained based on the accelerometer.

[0138] like Figure 3 As shown, (a), (b), and (d) are the first frame (static), 1120th frame, and 1650th frame of the left image sequence, respectively; (c) and (e) are the occluded target detection results corresponding to (b) and (d); and (f)-(g) are the corresponding right image sequence images. 3(a) shows the first frame (static) of the image sequence acquired by the left camera; (b) and (d) show the 1120th and 1650th frames of the left image sequence, reflecting the speckle occlusion caused by the fall of artificial markers. It can be seen that targets occluded in the left camera may not be occluded in the right camera. Figure 4 As shown, (a) is the set of occluded pixels in the left image sequence, (b) is the set of occluded pixels in the right image sequence, and (c) is the speckle information statistics in the ROI. In the figure, magenta indicates that the measurement was performed using a stereo camera, green and blue indicate that the measurement was performed using only the left or right camera, cyan indicates that the measurement was performed using a combination of mono / stereo, and yellow indicates that the target does not need to be measured. Figure 4 (a) and (b) show the motion of occluded targets in the entire image sequence extracted using an occluded target detection algorithm. A hybrid monocular / binocular high-speed video measurement strategy is employed to identify each target point to be measured, and a corresponding identifier is assigned to each target point, such as... Figure 4 As shown in (c).

[0139] The occlusion caused by speckles leads to significant errors in image processing algorithms, which in turn affects the calculation results of SRS. Figure 5The results show the impact response spectrum of a target point occluded in the left and right cameras at different times, with the occlusion time being longer in the right camera. It can be seen that using the left camera resulted in a large error in the high-frequency range. When using the right camera and the binocular camera, the overall SRS curve deviated significantly, indicating the significant impact of target occlusion on the SRS measurement results. However, the hybrid measurement method, through an image-matching-based hybrid strategy, integrates information from the speckle in different temporal and spatial dimensions, accurately recovering the dynamic changes of the speckle under occlusion conditions. Compared with the SRS reference value obtained based on accelerometers, the SRS measured by this invention shows good consistency with the reference result, with an average acceleration error of 12.32%, demonstrating the effectiveness of this invention in impact testing under occlusion conditions.

[0140] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement, characterized in that, Includes the following steps: Camera calibration: The high-speed camera was calibrated using Zhang Zhengyou's planar target calibration method to obtain camera parameters and scale factor; Stereo matching of static images: Stereo matching of initial static frames of binocular images obtained by a high-speed camera is performed under static conditions using the digital image correlation speckle matching method. Occlusion target detection: An occlusion target detection algorithm combining affine transformation is used to identify occlusion targets in the left and right image sequences; Image sequence matching: Based on the occluded target detection results, image sequence matching is performed through an occlusion condition discrimination strategy. For unoccluded speckle patterns, the digital image correlation speckle matching method is used to perform stereo-sequence dual matching in the stereo sequence images, combined with camera parameters. For speckle patterns that are occluded only in one camera view, the digital image correlation speckle matching method is used to perform monocular sequence matching, combined with a scale factor. Impact response spectrum calculation: Based on the image sequence matching results, the physical coordinates of the measurement points are reconstructed, and the coordinate system of the images taken from multiple perspectives is corrected. The corrected coordinates are then fused. The displacement of the speckle is determined based on the coordinate changes in the fusion results, and the acceleration is calculated. Based on the calculated acceleration, the impact response spectrum is calculated using the impact response spectrum calculation algorithm.

2. The method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement according to claim 1, characterized in that, The digital image correlation speckle matching method is specifically as follows: Digital image correlation algorithm is used to match speckles in images acquired by different high-speed cameras to obtain the correspondence of the same speckle in different acquisition times and imaging spaces. During the matching process, the first image of a selected camera is always used as the reference image. The region of interest is defined in the reference image, and the size and step of the matching window are set with a certain pixel coordinate as the center. Coarse matching and fine matching are used to match the image sequence in turn to determine the corresponding points.

3. The method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement according to claim 2, characterized in that, The coarse matching uses normalized cross-correlation to obtain integer pixel shifts, as shown in the following formula: in, and These represent the average gray values ​​of the reference subset and the target subset, respectively. and Representing coordinates ( x , y )and grayscale value at that location M This represents the half-step size of the matching window. C The correlation coefficient is represented; the reference subset and target subset are determined according to the execution object of the digital image correlation speckle matching method; Select the region with the highest correlation coefficient and use the corresponding pixel offset as the initial deformation parameter; The fine matching process uses a first-order deformation function to describe the deformation of the reference region relative to the target region, i.e., sub-pixel deformation, based on a preset matching window size. The first-order deformation function is shown in the following equation: in, u and v The center point is respectively P ( , )exist x and displacement components in the y direction, and It is the center point P Time P i ( x , y The distance, , , and They represent the displacement gradients, ( , (with ) as the center point P The corresponding two-dimensional coordinates; The optimal correlation coefficient is obtained through iterative optimization in order to solve for fine sub-pixel deformation parameters. .

4. The method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement according to claim 3, characterized in that, The optimal correlation coefficient is obtained using the iterative normalized least squares method, and the criteria for the iterative normalized least squares method are as follows: 。 5. The method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement according to claim 1, characterized in that, The occlusion target detection includes: When extracting the pixel coordinates of occluded targets in an image sequence, the first frame of each camera is used as the reference image. Unobstructed speckle patterns are selected in the reference image. The corresponding points of the selected speckle patterns in each frame of the image sequence are obtained using the digital image correlation speckle matching method. The transformation relationship between the image sequence and the reference image is established based on the affine transformation model of the following formula: in,( x , y ) are the coordinates of the speckle in the reference image, ( x ', y ') represents the coordinates of the corresponding point in the image sequence. A , B , C and D These are parameters that control rotation, scaling, and shearing. t x and t y It is a translation amount, and its affine transformation relationship is obtained through at least 3 pairs of coordinate points; The image sequence is subjected to affine transformation using an affine transformation model to generate an affine transformed image, and the difference image between the affine transformed image and the original image sequence is obtained. Occlusions in the difference image are extracted using the independent multimodal BGS algorithm from the BGS library, forming a set of occluded pixels.

6. The method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement according to claim 1, characterized in that, The image sequence matching includes the following steps: Select a search window centered on the target point to be measured, the size of which is the same as the size of the matching window; An occlusion condition discrimination strategy is implemented to determine the occlusion status in the image at each time step. When discrimination is completed at all time steps or the early termination condition is met, the corresponding matching is performed based on the discrimination results. Specifically: If the point to be measured is occluded in the search window of the image sequence captured by different cameras at a certain time, the early termination condition is met, the identifier of the point to be measured is set to 0, and the data of the point to be measured is discarded. If the point to be measured is not occluded in the search window of the image sequence captured by different cameras at all times, then the identifier of the point to be measured is set to 1, and a stereo-sequence double matching is performed on the binocular measurement result of the point to be measured. The stereo-sequence double matching is as follows: first, sequence matching is performed on the monocular image sequence of a single camera, and after the matching is completed, stereo matching is performed on the binocular image sequences of different cameras. If the point to be measured is not occluded in the search window of the image sequence taken by one camera at all times, but is occluded in the search window of the image sequence taken by another camera, then monocular sequence matching is performed based on the monocular measurement results of the camera that is not occluded. If the camera that is not occluded is the left camera, then the identifier of the point to be measured is set to 2; if it is the right camera, then the identifier of the point to be measured is set to 3. If none of the above conditions are met, and at any given moment in the entire image sequence of the point to be measured, there is occlusion in the search window of at most one image sequence taken by a camera, then the identifier of the point to be measured is set to 4. For moments when there is no occlusion in the search windows of image sequences taken by different cameras, the binocular measurement results are used for binocular stereo-sequence matching. For moments when there is occlusion in the search window of an image sequence taken by one camera, the unoccluded monocular measurement results are used for monocular sequence matching.

7. The method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement according to claim 6, characterized in that, The physical coordinates of the reconstructed measurement points are specifically as follows: Based on the image sequence matching results, the actual physical coordinates of the measurement points in the world coordinate system are reconstructed using a spatial mapping operator. The spatial mapping operator includes three-dimensional reconstruction and a scale factor, which is expressed as the ratio of the three-dimensional world coordinates to the two-dimensional image coordinates within the measurement area on the object surface.

8. The method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement according to claim 7, characterized in that, For speckle with marker 1, its three-dimensional coordinates are obtained through three-dimensional reconstruction technology. For speckle with markers 2, 3, and 4, the coordinates in the main impact direction are obtained by calculating the scale factor in the reference image.

9. The method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement according to claim 1, characterized in that, The coordinate system correction for images taken from multiple perspectives specifically involves adjusting the world coordinate system. O w Perform correction: The world coordinate system O w of X shaft and Y The axis is parallel to the plane target. Z Axis perpendicular to XOY Plane, assuming X The axis approximates the main impact direction, in the world coordinate system. O w The correction consists of the following two steps: Step 1 calibration: XOZ Plane around Y Rotation of the axis, causing X The axis is parallel to the object's surface, and two three-dimensional points are selected within the measurement area. X 1, Y 1, Z 1) and ( X 2, Y 2, Z 2) Calculate XOZ Vectors on a plane X 1, Z 1) and ( X 2, Z 2) The included angle between θ ; Second step of correction: XOY The plane rotates about the Z-axis, making X Aligning the axis with the main impact direction, a straight line extraction algorithm is used to obtain horizontal or vertical straight lines on the object's structure, and two three-dimensional points are selected within the measurement area. X 3, Y 3, Z 3) and ( X 4, Y 4, Z 4) The straight line formed by these two points is parallel to the extracted straight line, thus obtaining the vector ( X 3, Y 3) and ( X 4, Y 4) In XOY Angle on a plane β Then the corrected coordinates ( X' , Y' , Z' The following is represented: in,( X , Y , Z () represents the coordinates before correction.

10. The method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement according to claim 1, characterized in that, The shock response spectrum calculation algorithm performs the following steps: According to Newton's laws, the equations of motion for the mass block in a single-degree-of-freedom system are as follows: in, m , c and k These represent the mass, damping coefficient, and stiffness of the mass block in a single-degree-of-freedom system. x and y These are the response displacement of the mass block and the displacement of the base, respectively. , These are the velocity response of the mass block and the velocity of the base, respectively. For impact signal t The acceleration response at any given moment; The relative displacement of the mass block with respect to the base is expressed as z = x - y The derivation yields: in, To input the impact excitation acceleration signal of this single-degree-of-freedom system, the damping coefficient... stiffness Then, the equation relating the impact response spectrum is derived: in, For the damping ratio, The natural angular frequency is represented as . , It is the natural frequency; The relationship between the natural frequency and the peak value of the impact response of the single-degree-of-freedom system is obtained by solving the impact response spectrum equation using the digital recursive filtering method, thus obtaining the impact response spectrum of the real physical system.

Images