A method and system for measuring chest compression of automobile collision dummy
By posting multiple marking points on the car collision test dummy, using a binocular camera to collect and process images, obtaining the multi-position compression displacement of the dummy's chest, the problem of single measurement and large error in the prior art is solved, and the problem of "single measurement" and "large error" is realized. Accurate measurement of the amount of chest compression of dummies and the judgment of damage type in automobile collision tests.
Patent Information
- Application Number
- CN202410941578.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-07-15
AI Technical Summary
Existing automotive collision test dummies can only measure the compression displacement of a single position on the chest, and cannot effectively detect the compression at different positions and impact types. The sensor is limited by internal installation and cannot obtain data on slight impact and surface damage.
Post multiple marking points at the position to be measured on the dummy's chest, use a binocular camera to collect the images before and after the collision, obtain the compressed displacement of the marking points through three-dimensional reconstruction and image processing, and process the data of the occlusion points through inference and weight assignment, and divide the area into regions Calculate the average compression displacement to determine the type of chest deformation and the degree of injury.
Accurate measurement of the multiple positions and surface compression of dummy chests in automobile collision tests, solves the problems of single measurement and large error in the prior art, can obtain more detailed damage information, and improves the accuracy of measurement. and practicality.
Smart Images

Figure CN118464476B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of automobile collision experiments, and in particular to a method and system for measuring chest compression of an automobile collision dummy. Background Art
[0002] With the rapid development of my country's automobile industry and road traffic, cars have gradually become an important means of transportation for ordinary families. While cars have brought great convenience to people, they have also caused a series of traffic accidents. Human chest injury is an important cause of death or serious injury. According to literature statistics, among the deaths caused by traffic accidents, the number of deaths caused by chest injuries accounts for 30%. At the same time, human chest injury ranks second among all fatal injuries and serious injuries to passengers caused by car collisions, and its incidence is only lower than head injuries. Chest injuries mainly come from the collision between the human chest and the steering wheel, seat belts, interior and other parts when the car collides. Therefore, the study of the compression of the human chest during the collision is of great significance to the exploration of chest injuries.
[0003] Collision test is an important research method in the study of automobile passive safety. It includes real car collision test and simulated collision test. The simulated collision test is to use a trolley to test by simulating the collision deceleration waveform of the real car. Due to the authenticity and intuitiveness of the collision test, the collision test has an irreplaceable role and position in the study of automobile passive safety. The automobile collision test is a highly dangerous test, so it is impossible to use real people in the test, so the automobile collision test dummy was born. In the existing dummies, there is only one displacement sensor on the chest. In the collision test, only the compression displacement of a single position of the chest can be measured, and it is only for the frontal collision situation. This displacement sensor can only reflect the compression of the dummy's chest after being hit on the fixed position of the displacement sensor. The data obtained by this sensor cannot explore information such as the type of impact and the impact position. In fact, the compression of different positions of the dummy's chest is not the same under the action of the restraint system such as the seat belt or the difference in the impact position. The built-in sensor on the dummy's chest cannot reflect the type of damage to the dummy's exterior. Secondly, because the sensor is placed inside the chest cavity of the dummy, it can only generate displacement when a large impact causes deformation of rigid structures such as ribs, and cannot obtain impact data for minor impacts. At the same time, due to the cushioning of the dummy's skin, the compression of the dummy's chest surface cannot be obtained. This part of the damage in an actual collision can also cause serious damage to human skin and other tissues. In addition, existing traditional measuring devices such as displacement meters are affected by factors such as the measurement range and difficulty in fixing, and their application is limited. Summary of the invention
[0004] In order to solve the above technical problems, the present invention provides a method and system for measuring chest compression of an automobile collision dummy.
[0005] According to one aspect of the present invention, a method for measuring chest compression of a car collision dummy is provided, the method comprising the following steps:
[0006] Post multiple marking points on the chest of the dummy to be measured, conduct a collision test on the dummy, and use a calibrated camera to collect images of the dummy's chest before and after the collision;
[0007] The images before and after the collision are processed to obtain the compression displacement of multiple marking points after the collision; among them, for the marking points that are blocked after the collision, the corresponding compression displacement is obtained by inferring the compression displacement corresponding to other unblocked marking points.
[0008] In one possible implementation, the processing of the images before and after the collision to obtain the compression displacement of multiple markers after the collision includes: performing three-dimensional reconstruction based on the images collected by the dual cameras to obtain the three-dimensional coordinates of the markers before and after the collision; recording the coordinates of the markers before the collision as , the coordinates of the marked points after collision are , then the compression displacement of the marking point after collision is - .
[0009] In one possible implementation, for the marked point blocked after the collision, obtaining the corresponding compression displacement according to the compression displacements corresponding to other unblocked marked points includes:
[0010] For the marker point H that is blocked after the collision, the distance from point H to the marker point farthest from point H is Draw a circle with radius ; set each unobstructed marker point The distance from point H is , mark point The corresponding compression displacement is ;
[0011] At the marked point On the straight line established by the two points and the marked point H, the compression displacement is regarded as a linear change. The distance between the straight line and the marked point H is The compression at the point is the limit state, then for the marked point , the corresponding compression displacement at the marked point H for:
[0012] ;
[0013] For each compression displacement Assign a weight, then the compression displacement corresponding to the occluded marker point H for:
[0014] ;
[0015] In the formula, n Indicates the number of unoccluded markers.
[0016] In one possible implementation, after obtaining the compression displacements of multiple marking points after collision, predicting and judging the damage to the chest of the dummy according to the compression displacements specifically includes:
[0017] The position to be measured on the dummy's chest is divided into five areas: upper left, lower left, middle, upper right, and lower right. The average number of posting points in each area is taken to calculate the average compression displacement of each area.
[0018] The average compression displacement of each region is normalized respectively, and the normalization is the average compression displacement divided by the characteristic depth of the corresponding region; wherein the characteristic depth is the body thickness of the corresponding region of the dummy chest cavity;
[0019] Determine the chest deformation type of the dummy's chest according to the obtained normalized average compression displacement of each area; wherein the chest deformation type includes band constraint and uniform constraint;
[0020] The prediction models based on strip constraint and uniform constraint are used to predict the injury degree of the dummy chest.
[0021] In one possible implementation, the step of determining the chest deformation type of the dummy's chest according to the obtained normalized average compression displacement of each region includes:
[0022] Calculate the constraint discriminant value according to the following formula :
[0023] ;
[0024] In the formula, α, β, and γ are fixed constants.
[0025] ,
[0026] ,
[0027] ,
[0028] represents the normalized average compression displacement of the lower left area, represents the normalized average compression displacement of the lower right area, represents the normalized average compression displacement of the upper left area, represents the normalized average compression displacement of the upper right area, represents the normalized average compressive displacement of the middle area;
[0029] Constrain the discriminant value Compared with the preset critical threshold, if the constraint judgment value If it is greater than the preset critical threshold, the chest deformation type is uniform constraint; if it is less than the preset critical threshold, the chest deformation type is band constraint.
[0030] In one possible implementation, predicting the degree of injury of the dummy chest by using the prediction models based on the band constraint and the uniform constraint respectively includes:
[0031] The prediction model of the band constraint is:
[0032] ;
[0033] In the formula, δ, ε, ζ are fixed constants.
[0034] ;
[0035] ,
[0036] are the differential values of the average compression displacement in the upper left, middle, and upper right regions, respectively;
[0037] The prediction model for uniform constraints is:
[0038] ;
[0039] In the formula, η and θ are fixed constants;
[0040] when or When the values of are respectively greater than the corresponding preset critical thresholds, the injury level is severe.
[0041] In one possible implementation, after obtaining the compression displacement of multiple marking points after collision, the impact type of the dummy's chest is determined according to the compression displacement, and the impact type includes: frontal impact, side impact, and impact with multiple impact points.
[0042] According to another aspect of the present invention, a system for measuring chest compression of a car collision dummy is provided, the system comprising:
[0043] An image acquisition module is configured to post a plurality of marking points at positions to be measured on the dummy's chest, perform a collision test on the dummy, and use a calibrated camera to acquire images of the dummy's chest before and after the collision;
[0044] The image processing module is configured to process the images before and after the collision to obtain the compression displacement of multiple marking points after the collision; wherein, for the marking points that are blocked after the collision, the corresponding compression displacement is obtained by inferring the compression displacement corresponding to other unblocked marking points.
[0045] In one possible implementation, for a marked point blocked after the collision, the image processing module infers and obtains the corresponding compression displacement according to the compression displacements corresponding to other unblocked marked points, including:
[0046] For the marker point H that is blocked after the collision, the distance from point H to the marker point farthest from point H is Draw a circle with radius ; set each unobstructed marker point The distance from point H is , mark point The corresponding compression displacement is ;
[0047] At the marked point On the straight line established by the two points and the marked point H, the compression displacement is regarded as a linear change. The distance between the straight line and the marked point H is The compression at the point is the limit state, then for the marked point , the corresponding compression displacement at the marked point H for:
[0048] ;
[0049] For each compression displacement Assign a weight, then the compression displacement corresponding to the occluded marker point H for:
[0050] ;
[0051] In the formula, n Indicates the number of unoccluded markers.
[0052] In one possible implementation, the system further includes a damage prediction module, which is configured to predict and identify damage to the chest of the dummy according to the compression displacement after the collision of the multiple marker points after the collision, specifically including:
[0053] The position to be measured on the dummy's chest is divided into five areas: upper left, lower left, middle, upper right, and lower right. The average number of posting points in each area is taken to calculate the average compression displacement of each area.
[0054] The average compression displacement of each region is normalized respectively, and the normalization is the average compression displacement divided by the characteristic depth of the corresponding region; wherein the characteristic depth is the body thickness of the corresponding region of the dummy chest cavity;
[0055] Determine the chest deformation type of the dummy's chest according to the obtained normalized average compression displacement of each area; wherein the chest deformation type includes band constraint and uniform constraint;
[0056] The prediction models based on strip constraint and uniform constraint are used to predict the injury degree of the dummy chest.
[0057] The embodiments of the present invention have the following technical effects:
[0058] In order to obtain more accurate changes in the compression of the chest of a collision dummy and better explore the damage to the human chest in a car collision, the present invention proposes a method and system for measuring the compression of the chest of a car collision dummy. First, a plurality of marking points are posted at the position to be measured on the chest of the dummy, a collision test is performed on the dummy, and a calibrated camera is used to collect images of the chest of the dummy before and after the collision; then, the images before and after the collision are processed to obtain the compression displacement of the plurality of marking points after the collision; wherein, for the marking points that are blocked after the collision, the corresponding compression displacement is obtained by inferring the compression displacement corresponding to other unblocked marking points. The present invention can solve the measurement problem of the single compression of the chest of the collision dummy existing in the prior art and the large error, and adopts a non-contact external acquisition device to avoid the influence of factors such as the load and deformation of the measured object itself, and can obtain the type of external chest injury of the dummy, which is of great significance for exploring human chest injuries. The present invention can be applied to various working conditions such as automobile trolley collision simulation test and dummy chest calibration test, and can collect the compression of different positions of the dummy chest and the skin surface, with high practicality and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0060] Figure 1 The present invention provides a flow chart of a method for measuring chest compression of a car collision dummy.
[0061] Figure 2 It is a schematic diagram of marking point posting in an embodiment of the present invention.
[0062] Figure 3It is a schematic diagram of the positions of various components in a dummy chest calibration test in an embodiment of the present invention.
[0063] Figure 4 Schematic diagram of camera calibration black and white chess grid in an embodiment of the present invention.
[0064] Figure 5 Schematic diagram of classification of marked areas on the chest of a dummy in an embodiment of the present invention.
[0065] Figure 6 Schematic diagram of the compression of the dummy's chest under belt-shaped restraint in an embodiment of the present invention.
[0066] Figure 7 Schematic diagram of the compression of the dummy chest under uniform constraint in an embodiment of the present invention.
[0067] Figure 8 The present invention is a schematic structural diagram of a system for measuring chest compression of an automobile collision dummy provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0068] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.
[0069] The present invention provides a method and system for measuring the compression of a car collision dummy chest, which is used to measure the change of the compression of the dummy chest. The changes of specific marking points before and after the collision are collected by a binocular camera, the three-dimensional spatial displacement of the marking points is explored or inferred, and the change of the compression of the dummy chest is explored, so as to obtain the deformation of the dummy chest contour and the impact type, and at the same time, a prediction model of the dummy chest damage level is given, which more intuitively shows the dummy chest damage.
[0070] The embodiment of the present invention provides a method for measuring the chest compression of a car collision dummy, such as Figure 1 As shown, the method comprises the following steps:
[0071] S110, posting a plurality of marking points at the positions to be measured on the chest of the dummy, performing a collision test on the dummy, and using a calibrated camera to collect images of the dummy's chest before and after the collision;
[0072] S120, processing the images before and after the collision to obtain the compression displacements of multiple marking points after the collision; wherein, for the marking points that are blocked after the collision, the corresponding compression displacements are obtained by inferring the compression displacements corresponding to other unblocked marking points.
[0073] The method starts at S110. In S110, a plurality of marking points are posted at positions to be measured on the chest of a dummy, a collision test is performed on the dummy, and images of the dummy's chest before and after the collision are collected using a calibrated camera.
[0074] According to an embodiment of the present invention, Figure 2 As shown, the "+"-shaped marking points are flatly posted on the dummy's chest skin at the required measurement position, ensuring that the marking points are evenly illuminated relative to the camera and have no reflections. At the same time, due to the uncertainty of the dummy's impact position, it cannot be guaranteed that the marking points posted on the dummy can be exposed to the camera's field of view during the impact. Therefore, as many "+"-shaped marking stickers as possible are posted to provide enough information for the dummy position measurement.
[0075] The acquisition device includes a high-speed camera and a tripod fixing frame, and the calibration device is a black-and-white chessboard calibration board. The process of using the camera to collect images of the dummy's chest before and after the collision is as follows.
[0076] First, fix the position and place the dummy on the chest calibration test bench according to the chest calibration requirements. Place two high-speed cameras on the left and right sides of the calibration test bench respectively. Adjust the angles and focal lengths of the two cameras so that the dummy's chest is clearly visible in the field of view of the two cameras. At the same time, fix the cameras on a tripod. To prevent the vibration of the calibration bench from affecting the shooting accuracy, the camera tripod should be placed on the ground independently of the calibration bench, such as Figure 3 shown. Figure 3 In the middle, 1 is a dummy chest calibration pendulum, 2-1 and 2-2 are the left and right high-speed cameras respectively, 3 is a dummy chest calibration test bench, and 4 is a car test collision dummy.
[0077] In the image measurement process and machine vision applications, in order to determine the relationship between the three-dimensional geometric position of a point on the surface of a spatial object and its corresponding point in the image, a geometric model of camera imaging must be established. These geometric model parameters are the camera parameters. Under most conditions, these parameters must be obtained through experiments and calculations. The process of solving the parameters is called camera calibration (or camera calibration). The camera is calibrated using Zhang Zhengyou's black and white chess calibration method. The calibration board is as follows: Figure 4 Assume that 50 calibration plate images are taken, and the calibration plate is located on the plane of W=0 in the world coordinate system, and U and V are parallel to the two sides of the calibration plate. According to the width of each chessboard grid, the world coordinates of each corner point can be obtained (U i , V i , 0). After imaging the current calibration plate with a camera, the pixel coordinates of each black and white contact corner point in the image can be obtained using the corner detection algorithm. Because these corner points are one-to-one corresponding, multiple sets of equations from the world coordinate system to the pixel coordinate system can be established.
[0078] The unknown coefficients are solved through the following solution process to obtain the camera's intrinsic and extrinsic parameter matrix and distortion coefficients, completing the camera calibration process.
[0079] 1) Solve the product H of the internal and external parameters: A (R1R2 T ) is denoted as matrix H, where A is the internal parameter matrix, R1 and R2 are the first two columns of the rotation matrix R. The three columns of matrix H are denoted as (H1H2H3), and Z is the scale factor, then
[0080]
[0081] Using the above formula, eliminating the scale factor Z, we can get:
[0082]
[0083]
[0084] Among them, (u, v) is the pixel coordinate obtained by the camera, (U, V, 0) is the artificially defined world coordinate system, and both are known quantities, so the H matrix can be obtained.
[0085] 2) Solve the intrinsic parameter matrix: For simplicity, the camera intrinsic parameter matrix A is recorded as:
[0086]
[0087] Let B be B=A -T A -1 , and B is a symmetric matrix.
[0088]
[0089] Because B is a symmetric matrix and B=A -T A -1 , according to the calibration relationship provided by the calibration plate photos, each calibration relationship provides two constraint equations. When the number of calibration plate images is greater than 3 (generally 15-20 calibration plate images are required), the least squares fitting best vector result can be used to obtain matrix B. According to the corresponding relationship between the elements of matrix B and the camera intrinsic parameters α, β, γ, u0, v0, the camera's intrinsic parameter matrix A can be obtained.
[0090] 3) Solve the extrinsic parameter matrix: The intrinsic parameters of the camera have been obtained in the above process, so in this step, the position relationship between the two cameras is obtained, that is, the position transformation relationship R and T of one camera to the other camera. For the same camera, the intrinsic parameter matrix of the camera depends on the internal parameters of the camera. The extrinsic parameter matrix depends on the position relationship between the calibration plate and the camera. In the relationship A (R1R2T) = H, the matrix H and the matrix A have been solved. Through this relationship, the extrinsic parameter matrix corresponding to each picture can be obtained. Assuming that 50 calibration images are taken, 50 extrinsic parameter matrices are obtained, and the extrinsic parameter matrices of the left and right cameras are obtained. According to the following solution process, the R and T values of each matrix are obtained, and each R and T are approximately equal. Finally, the final camera group extrinsic parameter matrix is obtained by taking the average value.
[0091] 4) Solve the camera distortion correction parameters: the pixel coordinates after distortion ( , ) and the ideal undistorted pixel coordinates (u, v) are related by:
[0092]
[0093] Among them, r is the distance from the image pixel to the center of the image, is the constraint equation coefficient matrix D.
[0094] For each corner point, as long as the pixel coordinates after distortion are known ( , ), the ideal undistorted pixel coordinates (u, v), we can construct the two equations above. In the case of m images, each image has n calibration plate corner points, it is possible to get m*n unknowns k=[k1, k2] T The constraint equation is denoted by D, and the non-homogeneous term on the right side of the equation is denoted by d. It can be recorded in matrix form: Dk=d, and then the least squares method is used:
[0095]
[0096] The ideal, undistorted pixel coordinates (u, v) can be obtained.
[0097] Through calibration, for a single camera, assuming that N images are taken, N external parameter matrices of the camera can be obtained; for a dual-camera system, the external parameter matrices of the left and right cameras can be obtained, which are , , , , where i=1, 2…, N. For any point in the i-th image, assume that its coordinates in the world coordinate system, the left camera coordinate system, and the right camera coordinate system are 、 and , then:
[0098]
[0099] Eliminate in the above formula , then at each calibration position, the position transformation relationship between the two cameras is , It can be expressed as:
[0100]
[0101] Since the positions of the two cameras in the dual-camera system are relatively fixed, the calculated , are approximately equal, and the final camera group extrinsic parameter matrix can be obtained by taking the average value, and the calibration process is completed.
[0102]
[0103] in, , T = .
[0104] After calibration, the calibrated camera is used to collect images of the dummy's chest before and after the collision. The image captured by the camera can be regarded as a two-dimensional matrix containing grayscale value information. A single image can only provide two-dimensional plane information of an object, and its depth information can only be obtained with the help of a camera image from another angle.
[0105] Then, S120 is executed, in which the images before and after the collision are processed to obtain the compression displacements of multiple marking points after the collision; wherein, for the marking points that are blocked after the collision, the corresponding compression displacements are obtained by inferring the compression displacements corresponding to other unblocked marking points.
[0106] According to the embodiment of the present invention, three-dimensional reconstruction is performed using images collected by dual cameras to obtain three-dimensional coordinates of the marking points before and after the collision.
[0107] Specifically, the three-dimensional coordinates of the marker points before and after the collision are calculated as follows: Taking the left camera coordinate system as the world coordinate system, the imaging models of the two cameras are:
[0108]
[0109]
[0110] The equation is:
[0111]
[0112] in:
[0113]
[0114] , The parameters in the calibration process have been obtained. , )( , ) is the pixel coordinate of the marker point on the left and right cameras, and the solution is . That is the three-dimensional coordinate of the required mark point.
[0115] Collect images of the marker points before and after the collision to obtain the three-dimensional coordinates before and after the collision. The coordinates of the marker points before the collision are recorded as , the coordinates of the marked points after collision are , then the compression displacement of the marking point before and after the collision is - .
[0116] If the required measurement point can be exposed in the camera's field of view during a collision, the compression displacement of the measurement point can be directly calculated using the above calculation method; if the required measurement point is blocked during a collision, such as by the presence of a seat belt, airbag, or steering wheel, the marked point may be blocked during a collision and cannot be captured by the camera. Therefore, it is necessary to use information from other measurement points for inference.
[0117] For the marked points that are blocked after the collision, the compressed displacement corresponding to the blocked marked points is inferred based on the compressed displacement corresponding to other unblocked marked points, including:
[0118] Take the required measurement point H as the base point and the distance between the point A0 on the dummy's chest skin and the farthest point H. Draw a circle with radius A. i The distance from point H to (i=1, 2…n) is The compression displacement is According to the above information, i On the straight line established by point A and point H, the compression displacement of the dummy is considered to change linearly, and the compression at the point D0 away from point H on the straight line is the limit state (no compression). i Point, compression displacement at point H for:
[0119]
[0120] In the formula, , n represents the number of unoccluded markers.
[0121] Since the distances of the observable measuring points from point H vary, the closer the distance to point H, the greater the credibility and correlation. Therefore, when calculating the compression displacement of point H from the compression displacements corresponding to n measuring points, a higher weight is given to the displacement information calculated from the closer points, and a lower weight is given to the displacement information calculated from the farther points. The compression displacement of the obscured measuring point H is calculated by the following formula: :
[0122]
[0123] Furthermore, the method further includes S130, after obtaining the compression displacements of the plurality of marking points after the collision, predicting and judging the damage to the chest of the dummy according to the compression displacements, specifically including:
[0124] The position to be measured on the dummy's chest is divided into five areas: upper left, lower left, middle, upper right, and lower right. The average number of posting points in each area is taken to calculate the average compression displacement of each area.
[0125] The average compression displacement of each region is normalized respectively, and the normalization is the average compression displacement divided by the characteristic depth of the corresponding region; wherein the characteristic depth is the body thickness of the corresponding region of the dummy chest cavity;
[0126] Determine the chest deformation type of the dummy's chest according to the obtained normalized average compression displacement of each area; wherein the chest deformation type includes band constraint and uniform constraint;
[0127] The prediction models based on strip constraint and uniform constraint are used to predict the injury degree of the dummy chest.
[0128] According to an embodiment of the present invention, the dummy chest marker posting area is roughly divided into five parts according to the posting position: upper left (UL), upper right (UR), middle (CT), lower left (LL), and lower right (LR). Figure 5 As shown. The average compression displacement of each area is calculated by taking the average value according to the number of marked points in each area. When the number of marked points in the area is greater than 5, the maximum and minimum points of the compression displacement are not calculated; when the number of marked points in the area is less than or equal to 5, the average value is directly calculated. The average compression displacement corresponding to the upper left area is calculated as follows:
[0129]
[0130] In the formula, N is the number of marked points in the upper left area UL; other areas are calculated in the same way as above , , , .
[0131] Then, due to the different thickness of the dummy skin and the structural differences at different positions, the average compression displacement of the five regions is normalized by dividing them by the characteristic depth L of the region where they are located. The characteristic depth L is the body thickness from front to back of the dummy chest region. For example, HybridⅢ50 th Characteristic depth of UL and UR areas of the dummy and The characteristic depth of the CT area is about 223 mm. The characteristic depth of the LL and LR regions is about 236 mm. and It is about 220mm. The normalization formula is as follows:
[0132]
[0133]
[0134]
[0135]
[0136]
[0137] Then, the constraint judgment is performed based on the normalized compression amount of each region, where the constraints are divided into band constraints and uniform constraints. Based on multiple test data and theoretical analysis, the constraint judgment formula is given, namely:
[0138]
[0139] in, The value range of is (2.6, 2.7), The value range of is (2.5, 2.6), The value range of is (7.7, 7.9), and the parameter value range is obtained by calculating multiple test data.
[0140]
[0141]
[0142]
[0143] represents the normalized average compression displacement of the lower left area, represents the normalized average compression displacement of the lower right area, represents the normalized average compression displacement of the upper left area, represents the normalized average compression displacement of the upper right area, represents the normalized average compressive displacement of the middle area; Indicates the symmetry of the lower chest deformation, and its value range is 1-2. For symmetrical chest deformation (such as uniform constraint), is closer to 2; for asymmetric chest deformation (such as belt restraint), The value of is closer to 1; Indicates the symmetry of the upper chest, and its value range is 1-3. For symmetrical chest deformation, its value is closer to 3; for asymmetrical chest deformation, its value is closer to 1. Based on the comparative analysis of multiple test results, the constraint discriminant value The critical threshold of is 9, that is, values greater than 9 are considered uniform constraints, and values less than 9 are considered band constraints.
[0144] Then, the prediction models based on the belt-shaped constraint and uniform constraint are used to predict the degree of injury to the dummy chest. For the belt-shaped constraint, the following prediction model can better predict the injury category:
[0145]
[0146] in, The value range is (10.2, 10.3), The value range is (13.1, 13.3), The value range of is (4.4, 4.5), and the parameter value range is obtained by calculating multiple test data.
[0147]
[0148]
[0149] In the formula, , , The average compression displacement , , Differentiate to get .
[0150] For band constraints, The critical threshold is -6. When AIS is >-6, the chest injury grade AIS is ≥3, otherwise AIS is <3.
[0151] Table 1 Abbreviated Injury Scale (AIS)
[0152]
[0153] For uniform constraints, the prediction model is:
[0154]
[0155] in, The value range is (10.0, 10.2), The value range of is (4.1, 4.3), and the parameter value range is obtained by calculating multiple test data.
[0156] For uniform constraints, The critical threshold is 6.5. When it is >6.5, the chest injury grade AIS≥3, otherwise AIS<3.
[0157] Furthermore, it also includes S140, after obtaining the compression displacement of multiple marking points before and after the collision, judging the impact type of the dummy's chest according to the compression displacement, and the impact type includes: frontal impact, side impact, and impact with multiple impact points.
[0158] According to an embodiment of the present invention, the maximum compression displacement P is obtained by comparing the displacement information of the n marking points measured. max , according to P max The marking point can determine its main impact displacement: if P max and its similar displacement points are mainly concentrated on the right side of the dummy's chest, it can be determined that the impact is a side impact; if P max and its similar displacement points are mainly concentrated in front of the dummy's chest, then it can be determined that the impact is a frontal impact; if P max If the points with similar displacement are distributed in different positions, it can be determined that the impact point is not unique. The above information helps to better judge the impact type and provide more valuable information for automobile safety design.
[0159] Furthermore, the damage to the dummy's chest can be judged based on the deformation of the dummy's chest contour. Figure 6 As shown in Figure 1, the T value is less than 9, which is a belt-shaped restraint. Combined with the change in compression displacement, it can be inferred that the injury may be caused by restraints such as seat belts; Figure 7 As shown, the T value is greater than 9, which indicates uniform restraint. Combined with the change in compression, it can be inferred that the damage may be caused by restraints such as airbags.
[0160] The present invention can quickly and accurately detect the compression displacement of the dummy's chest in the collision test, and solve the problem that the compression of the dummy's chest in the existing automobile collision test is obtained in a single and limited value. At the same time, the operation method is relatively simple. Based on the existing automobile collision dummy test, non-contact external acquisition is adopted, which is not affected by the load and deformation of the dummy itself, and the measurement range is large. The compression displacement of any position on the dummy's chest can be obtained through the present invention. At the same time, for the blocked parts, predictions can be made based on the non-blocked points. According to the obtained compression data, information such as the constraint type, impact position, predicted injury level, and cause of injury can be determined. Combined with the built-in sensor data of the dummy's chest, a more reference-worthy overall reference compression of the chest is given. The present invention improves the authenticity and accuracy of the dummy's chest compression data, can better meet the needs of researchers, and provide more valuable data information for research.
[0161] Another embodiment of the present invention provides a system for measuring chest compression of a car collision dummy, such as Figure 8 As shown, the system includes:
[0162] An image acquisition module 810 is configured to post a plurality of marking points at the position to be measured on the dummy's chest, perform a collision test on the dummy, and use a calibrated camera to acquire images of the dummy's chest before and after the collision;
[0163] The image processing module 820 is configured to process the images before and after the collision to obtain the compression displacement of multiple marking points after the collision; wherein, for the marking points that are blocked after the collision, the corresponding compression displacement is obtained by inferring the compression displacement corresponding to other unblocked marking points.
[0164] In this embodiment, preferably, the image processing module 820 infers the corresponding compression displacement of the marked point blocked after the collision according to the compression displacement corresponding to other unblocked marked points, including:
[0165] For the marker point H that is blocked after the collision, the distance from point H to the marker point farthest from point H is Draw a circle with radius ; set each unobstructed marker point The distance from point H is , mark point The corresponding compression displacement is ;
[0166] At the marked point On the straight line established by the two points and the marked point H, the compression displacement is regarded as a linear change. The distance between the straight line and the marked point H is The compression at the point is the limit state, then for the marked point , the corresponding compression displacement at the marked point H for:
[0167]
[0168] For each compression displacement Assign a weight, then the compression displacement corresponding to the occluded marker point H for:
[0169]
[0170] In the formula, n Indicates the number of unoccluded markers.
[0171] In this embodiment, preferably, the system further includes a damage prediction module 830, which is configured to predict and judge the damage of the dummy's chest according to the compression displacement after obtaining the compression displacement of multiple marking points after collision, specifically including:
[0172] The position to be measured on the dummy's chest is divided into five areas: upper left, lower left, middle, upper right, and lower right. The average number of posting points in each area is taken to calculate the average compression displacement of each area.
[0173] The average compression displacement of each region is normalized respectively, and the normalization is the average compression displacement divided by the characteristic depth of the corresponding region; wherein the characteristic depth is the body thickness of the corresponding region of the dummy chest cavity;
[0174] Determine the chest deformation type of the dummy's chest according to the obtained normalized average compression displacement of each area; wherein the chest deformation type includes band constraint and uniform constraint;
[0175] The prediction models based on strip constraint and uniform constraint are used to predict the injury degree of the dummy chest.
[0176] For the undetailed parts of the system for measuring chest compression of a car collision dummy according to an embodiment of the present invention, please refer to the above detailed description of the method embodiment.
[0177] It should be noted that the terms used in the present invention are only for describing specific embodiments, rather than limiting the scope of the present application. As shown in the present specification, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of more restrictions, the elements defined by the sentence "include one..." do not exclude the presence of other identical elements in the process, method or device including the elements.
[0178] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.
Claims
1. A method for measuring chest compression of a car collision dummy, characterized in that: The following steps are involved: Post multiple marking points on the chest of the dummy to be measured, conduct a collision test on the dummy, and use a calibrated camera to collect images of the dummy's chest before and after the collision; Process the images before and after the collision to obtain the compression displacement of multiple marking points after the collision; for the marking points that are blocked after the collision, the corresponding compression displacement is obtained by inferring the compression displacement corresponding to other unblocked marking points; After obtaining the compression displacement of multiple marking points after collision, the damage to the chest of the dummy is predicted and judged according to the compression displacement, specifically including: The position to be measured on the dummy's chest is divided into five areas: upper left, lower left, middle, upper right, and lower right. The average number of posting points in each area is taken to calculate the average compression displacement of each area. The average compression displacement of each region is normalized respectively, and the normalization is the average compression displacement divided by the characteristic depth of the corresponding region; wherein the characteristic depth is the body thickness of the corresponding region of the dummy chest cavity; Determine the chest deformation type of the dummy's chest according to the obtained normalized average compression displacement of each area; wherein the chest deformation type includes band constraint and uniform constraint; The damage degree of the dummy chest is predicted based on the prediction models of band constraint and uniform constraint respectively; The method of determining the chest deformation type of the dummy's chest according to the obtained normalized average compression displacement of each area includes: Calculate the constraint discriminant value according to the following formula : ; In the formula, α, β, and γ are fixed constants. , , , represents the normalized average compression displacement of the lower left area, represents the normalized average compression displacement of the lower right area, represents the normalized average compression displacement of the upper left area, represents the normalized average compression displacement of the upper right area, represents the normalized average compressive displacement of the middle area; Indicates the symmetry of the lower chest deformation, Indicates symmetry of the upper chest; Constrain the discriminant value Compared with the preset critical threshold, if the constraint judgment value If it is greater than the preset critical threshold, the chest deformation type is uniform constraint; if it is less than the preset critical threshold, the chest deformation type is band constraint.
2. The method for measuring chest compression of a car collision dummy according to claim 1, characterized in that: The processing of the images before and after the collision to obtain the compression displacement of multiple marking points after the collision includes: performing three-dimensional reconstruction based on the images collected by the dual cameras to obtain the three-dimensional coordinates of the marking points before and after the collision; recording the coordinates of the marking points before the collision as , the coordinates of the marked points after collision are , then the compression displacement of the marking point after collision is - .
3. The method for measuring chest compression of a car collision dummy according to claim 2, characterized in that: For the marked point that is blocked after the collision, obtaining the corresponding compression displacement amount by inferring according to the compression displacement amounts corresponding to other unblocked marked points comprises: For the marker point H that is blocked after the collision, the distance from point H to the marker point farthest from point H is Draw a circle with radius ; set each unobstructed marker point The distance from point H is , mark point The corresponding compression displacement is ; At the marked point On the straight line established by the two points and the marked point H, the compression displacement is regarded as a linear change. The distance between the straight line and the marked point H is The compression at the point is the limit state, and the limit state is no compression. , the corresponding compression displacement at the marked point H for: ; For each compression displacement Assign a weight, then the compression displacement corresponding to the occluded marker point H for: ; In the formula, n Indicates the number of unoccluded markers.
4. The method for measuring chest compression of a car collision dummy according to claim 1, characterized in that: The prediction models based on the band constraint and the uniform constraint respectively predict the damage degree of the dummy chest, including: The prediction model of the band constraint is: ; In the formula, δ, ε, ζ are fixed constants. , are the differential values of the average compression displacement in the upper left, middle, and upper right regions, respectively; The prediction model for uniform constraints is: ; In the formula, η and θ are fixed constants; when or When the values of are respectively greater than the corresponding preset critical thresholds, the injury level is severe.
5. The method for measuring chest compression of a car collision dummy according to claim 1, characterized in that: After obtaining the compression displacements of the multiple marking points after collision, the impact type of the dummy's chest is determined according to the compression displacements, and the impact types include: frontal impact, side impact, and impact with multiple impact points.
6. A system for measuring chest compression of a car collision dummy, characterized in that: include: An image acquisition module is configured to post a plurality of marking points at positions to be measured on the dummy's chest, perform a collision test on the dummy, and use a calibrated camera to acquire images of the dummy's chest before and after the collision; An image processing module is configured to process the images before and after the collision to obtain the compression displacement of multiple marking points after the collision; wherein, for the marking points that are blocked after the collision, the corresponding compression displacement is obtained by inferring the compression displacement corresponding to other unblocked marking points; After obtaining the compression displacement of multiple marking points after collision, the damage to the chest of the dummy is predicted and judged according to the compression displacement, specifically including: The position to be measured on the dummy's chest is divided into five areas: upper left, lower left, middle, upper right, and lower right. The average number of posting points in each area is taken to calculate the average compression displacement of each area. The average compression displacement of each region is normalized respectively, and the normalization is the average compression displacement divided by the characteristic depth of the corresponding region; wherein the characteristic depth is the body thickness of the corresponding region of the dummy chest cavity; Determine the chest deformation type of the dummy's chest according to the obtained normalized average compression displacement of each area; wherein the chest deformation type includes band constraint and uniform constraint; The damage degree of the dummy chest is predicted based on the prediction models of band constraint and uniform constraint respectively; The method of determining the chest deformation type of the dummy's chest according to the obtained normalized average compression displacement of each area includes: Calculate the constraint discriminant value according to the following formula : ; In the formula, α, β, and γ are fixed constants. , , , represents the normalized average compression displacement of the lower left area, represents the normalized average compression displacement of the lower right area, represents the normalized average compression displacement of the upper left area, represents the normalized average compression displacement of the upper right area, represents the normalized average compressive displacement of the middle area; Indicates the symmetry of the lower chest deformation, Indicates symmetry of the upper chest; Constrain the discriminant value Compared with the preset critical threshold, if the constraint judgment value If it is greater than the preset critical threshold, the chest deformation type is uniform constraint; if it is less than the preset critical threshold, the chest deformation type is band constraint.
7. The automobile collision dummy chest compression measurement system according to claim 6, characterized in that: In the image processing module, for a marked point blocked after the collision, obtaining the corresponding compression displacement according to the compression displacement corresponding to other unblocked marked points includes: For the marker point H that is blocked after the collision, the distance from point H to the marker point farthest from point H is Draw a circle with radius ; set each unobstructed marker point The distance from point H is , mark point The corresponding compression displacement is ; At the marked point On the straight line established by the two points and the marked point H, the compression displacement is regarded as a linear change. The distance between the straight line and the marked point H is The compression at the point is the limit state, and the limit state is no compression. , the corresponding compression displacement at the marked point H for: ; For each compression displacement Assign a weight, then the compression displacement corresponding to the occluded marker point H for: ; In the formula, n Indicates the number of unoccluded markers.
