A method and system for motion analysis of images in automobile crash tests

Abstract

This invention relates to the field of automotive crash test technology, specifically to a method and system for image motion analysis in automotive crash tests. The method includes: pre-test preparation, including: selecting a high-speed camera; synchronously calibrating the camera using a synchronous detector; calibrating the camera's internal orientation elements; selecting the type and size of marker points and affixing them to key parts of the test object; calibrating control point coordinates; selecting an online or offline analysis method based on the changes in internal and external inverted position elements of the image measurement equipment and the correlation of the image measurement data channel; preparing for and initiating the crash test; data acquisition and preliminary processing to extract key information; and calculating the performance and accuracy of the image measurement data channel based on the key information, and evaluating whether the performance and accuracy indicators meet the requirements. This technical solution can improve the accuracy and testing efficiency of automotive crash tests.

Description

Technical Field

[0001] This invention relates to the field of automotive crash test technology, specifically to a method and system for automotive crash test image motion analysis. Background Technology

[0002] In the automotive industry, crash tests are crucial for evaluating vehicle safety performance. Precise analysis of the motion states during a crash provides key data support for improving vehicle safety design, thereby effectively reducing the risk of injury or death in traffic accidents. However, existing crash test image motion analysis techniques have many shortcomings and are insufficient to meet the ever-increasing demands of current automotive safety performance research.

[0003] Existing image motion analysis testing methods lack unified standards and specifications in their testing procedures, resulting in poor stability and repeatability of test results. Furthermore, existing methods do not comprehensively evaluate the performance and accuracy of image measurement data channels. They fail to consider the impact of multiple performance indices on positioning accuracy, making it impossible to fully understand the system's performance in different aspects. Simultaneously, the methods for calculating and evaluating accuracy indicators are imperfect, making it difficult to effectively determine whether test results meet requirements and ensuring the quality of test data. Summary of the Invention

[0004] The purpose of this invention is to propose a method for motion analysis of images in automobile crash tests, which can improve the accuracy and efficiency of automobile crash tests.

[0005] To achieve the above objectives, in a first aspect, the present invention provides a vehicle crash test image motion analysis testing system, comprising:

[0006] Preparing for the test includes:

[0007] Select a high-speed camera and set the lens aperture and camera exposure time;

[0008] Use a synchronization detector to calibrate the cameras to ensure that the synchronization accuracy between the cameras meets the requirements.

[0009] Complete the calibration of the camera's internal orientation elements, including the principal point, focal length, and distortion parameters;

[0010] Based on the pixel size of the marker, select the marker type and size, and paste the marker on the key parts of the object being tested, ensuring that the center position of the marker is accurate and that it is tightly attached to the surface of the object being tested.

[0011] Perform control point coordinate calibration by measuring the three-dimensional coordinates of the control points using measuring equipment;

[0012] Based on the changes in the internal and external reverse station elements of the image measurement equipment and the correlation of the image measurement data channel, select either an online analysis method or an offline analysis method.

[0013] Online analysis methods include:

[0014] Prepare for and initiate the collision test; collect and preliminarily process data to extract key information; based on the key information, calculate the performance and accuracy of the image measurement data channel, and evaluate whether the performance and accuracy indicators meet the requirements.

[0015] Offline analysis methods include:

[0016] Preliminary testing was conducted, collision test preparations were made and the collision test was initiated, the performance of the image measurement data channel was calculated, the performance indicators were evaluated to determine whether they met the requirements, and the imaging rate, light field environment and measurement space area parameters of the camera in subsequent tests were kept basically consistent.

[0017] Subsequent collision tests were conducted to calculate the accuracy of the image measurement data channel and assess whether the accuracy indicators met the requirements.

[0018] Beneficial effects of the basic solution: This technical solution standardizes test preparation procedures and improves test accuracy. The selection of a high-speed camera suitable for automotive crash tests, with its high resolution and high frame rate, captures details of rapidly moving objects at the moment of impact. A fixed-focus lens combined with an appropriate aperture value ensures sufficient light intake while controlling depth of field, guaranteeing clear imaging of the target. Reasonable exposure time settings reduce motion blur, providing a clear and stable image foundation for subsequent accurate analysis.

[0019] Synchronization calibration of the cameras using a synchronization detector ensures that the camera synchronization accuracy meets requirements, guaranteeing the synchronicity of shooting by each camera and avoiding data errors caused by camera asynchrony. Accurate calibration of the camera's internal orientation elements, obtaining principal points, focal lengths, and distortion parameters, helps improve the accuracy of image measurements and provides a guarantee for subsequent precise analysis.

[0020] Selecting the appropriate type and specifications based on the pixel size of the marker points and accurately attaching them to key parts of the object under test ensures stable motion information during collision. Control point coordinate calibration utilizes high-precision measuring equipment to measure three-dimensional coordinates, with high calibration accuracy requirements, providing a precise reference benchmark for image measurement and improving the accuracy and reliability of the measurement.

[0021] This technical solution selects either an online or offline analysis method based on the changes in the internal and external orientation elements of the image measurement equipment and the channel-related information of the image measurement data. Online analysis is suitable for situations where the equipment's orientation elements change or there is no channel-related information, allowing for real-time data processing and analysis. Offline analysis is suitable for situations where the equipment remains unchanged or undergoes only minor changes; initial testing calculates performance indicators, while subsequent experiments only focus on accuracy. Both methods meet the needs of different scenarios, improving the applicability of the testing system.

[0022] During online analysis, not only is data collected and preliminarily processed, but the performance and accuracy of the image measurement data channel are also calculated. Multiple performance indices are used to evaluate system performance, comprehensively considering the system's impact on positioning accuracy in various aspects. Simultaneously, accuracy indicators, such as camera position calculation index and proportional index, are calculated and assessed to determine if requirements are met, enabling real-time monitoring and evaluation of the testing process.

[0023] Offline analysis calculates and evaluates performance metrics in the initial testing phase to ensure system performance meets requirements. It then calculates and evaluates accuracy in subsequent crash tests, guaranteeing accuracy throughout the entire testing process. This combination of initial performance evaluation and subsequent accuracy evaluation provides comprehensive control over test results, ensuring data quality and improving testing efficiency.

[0024] This technical solution provides crucial technical support and assurance for the research and application of automotive crash testing by improving testing accuracy and reliability, optimizing testing processes and efficiency, enhancing test repeatability and verifiability, and promoting technological innovation and R&D. It has a wide range of applications and can be used in various passive safety tests, including whole vehicle tests, sled tests, and pedestrian protection tests.

[0025] As a feasible and preferred solution, a synchronization detector is used to calibrate the cameras to ensure that the synchronization accuracy between the cameras meets the requirements, including the following:

[0026] Connect all cameras involved in the test using a synchronization detector. Connect the trigger signal output port of the synchronization detector to the trigger input port of each camera. Start the synchronization detector and camera equipment. Set the synchronization mode and parameters on the operation interface of the synchronization detector and perform synchronization calibration.

[0027] As a feasible and preferred solution in camera synchronization calibration and standardization, the synchronization detector sends a synchronization trigger signal to each camera. After receiving the signal, the cameras take pictures simultaneously. The synchronization detector automatically calculates and adjusts the synchronization error between the cameras by detecting the difference in the shooting time of the cameras, so that the synchronization accuracy reaches the requirement of not less than 10μs.

[0028] As a feasible and preferred approach, control point coordinate calibration includes the following:

[0029] Prepare a calibration board with high-precision markers and known size and position. Place the calibration board within the camera's field of view, adjust its position and angle, and take multiple sets of images at different positions and angles, with each set containing at least 8 different positions of the calibration board's posture. Import the captured calibration board images using camera calibration software and calculate the camera's interior orientation elements.

[0030] As a feasible preferred solution, the online analysis method is suitable for situations where the orientation elements inside and outside the image measurement equipment change, or where there is no image measurement data channel related information; the offline analysis method is suitable for situations where the equipment of the image measurement data channel remains unchanged or only undergoes minor changes.

[0031] As a feasible and preferred option, the performance indicators of the image measurement data channel include the performance index of a single image recording device, including: focal length index to evaluate the impact of focal length error on positioning accuracy; distortion index to evaluate the impact of distortion parameters on positioning accuracy; marker recognition index to evaluate the impact of marker recognition accuracy on positioning accuracy; motion blur index to evaluate the impact of motion blur on positioning accuracy; marker motion index to determine the required motion distance of the current marker between two frames of images according to test requirements; control point distribution index to determine the number of control points in different image regions i and the coverage ratio of the regions where the control points are located; time base index to evaluate the impact of time base accuracy on positioning accuracy; initial moment recognition index to evaluate the impact of initial moment recognition accuracy on positioning accuracy; camera placement index, applicable only to two-dimensional motion analysis, used to evaluate the orientation requirements of the camera relative to the plane; and motion plane scale parameter, applicable only to two-dimensional motion analysis, used to clarify the scale information requirements of each motion plane.

[0032] The performance metrics of the image measurement data channel also include the performance index between image recording devices, including the interaction index, which is used to evaluate the positioning accuracy requirements of the measured object in the depth direction; and the synchronization index, which is used to evaluate the impact of asynchrony between cameras on positioning accuracy.

[0033] As a feasible and preferred approach, the evaluation of whether the two-dimensional and three-dimensional performance values ​​meet the requirements includes the following:

[0034] Two-dimensional performance values ​​are used to describe the performance of a single image recording device. The two-dimensional performance value of image recording device i... The calculation formula is:

[0035]

[0036] in, The serial number of the image recording device; The number of two-dimensional performance indices; Image recording device The actual value of the two-dimensional performance index; The number of two-dimensional performance indices;

[0037] Three-dimensional performance values ​​are used to describe the performance of each image recording device in the image measurement data channel. The calculation formula is:

[0038]

[0039] in, This refers to the sequence number of the three-dimensional performance index. Three-dimensional performance index The actual value, This represents the number of three-dimensional performance indices.

[0040] As a feasible preferred solution, the accuracy of the image measurement data channel is calculated. The accuracy indicators of the image measurement data channel include the camera position calculation index and the scale index. The camera position calculation index determines the impact of the camera position calculation method on the position accuracy, and the scale parameter is used to evaluate the existence requirements of reference distances in different directions in the object space.

[0041] Calculate the error and accuracy of a single reference distance measurement, including:

[0042] If the camera position calculation index of all image recording devices All requirements are met, and the reference distance is determined by any single time step within the analysis time interval. Measurement error ;

[0043]

[0044] in, To analyze any single time step within the time interval, For reference distance The calibration length;

[0045] If there are any camera position calculation indicators for image recording devices If the requirements are not met, the length measurement error for each time step must be calculated within the analysis time interval. ;

[0046]

[0047] in, To determine the initial time of the analysis time interval, This represents the end time of the analysis time interval;

[0048] Reference distance precision value It is a length measurement error With calibration length The ratio between them:

[0049]

[0050] Calculate the measurement error and accuracy of the image measurement channel, including:

[0051] Measurement error of image measurement channel Take all reference distance measurement errors The maximum value in:

[0052]

[0053] Precision value of image measurement channel Take all reference distance accuracy values The maximum value in:

[0054] .

[0055] As a feasible and preferred option, the performance and accuracy evaluation requirements for the online analysis method are as follows:

[0056] All performance metrics involved must be no less than 0.5;

[0057] The performance value of the image measurement data channel should be greater than 0.7;

[0058] Image measurement data channel length measurement error Must be lower than position accuracy ;

[0059] Precision value of image measurement data channel It must be lower than the precision limit;

[0060] The accuracy evaluation requirements for the offline analysis method in the initial testing are as follows:

[0061] All performance metrics involved must be no less than 0.5;

[0062] The performance value of the image measurement data channel is greater than 0.8;

[0063] Image measurement data channel length measurement error Must be lower than position accuracy ;

[0064] Precision value of image measurement data channel It must be lower than the precision limit;

[0065] The accuracy assessment requirements for subsequent crash tests are as follows:

[0066] For 3D motion analysis, check synchronization and ensure the synchronization index. ≥ 1;

[0067] Image measurement data channel length measurement error Must be lower than position accuracy ;

[0068] Precision value of image measurement data channel It must be below the precision limit.

[0069] Secondly, the present invention also provides a vehicle crash test image motion analysis testing system, which utilizes the aforementioned vehicle crash test image motion analysis testing method. Attached Figure Description

[0070] Figure 1 This is a logical schematic diagram of a motion analysis test method for automobile crash test images.

[0071] Figure 2 This is a schematic diagram of the image region distribution;

[0072] Figure 3 A schematic diagram showing the image width of each image region;

[0073] Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0074] To make the technical solution and advantages of this application clearer, the technical solution of the present invention will be further described in detail below with reference to the accompanying drawings. It is understood that the specific embodiments described herein are only some embodiments of the present invention, and are only used to explain this application, not to limit it. It should be noted that the technical features or combinations of technical features described in the following embodiments should not be considered isolated; they can be combined with each other to achieve better technical effects. The same reference numerals appearing in the accompanying drawings of the following embodiments represent the same features or components, and can be applied to different embodiments.

[0075] Furthermore, unless otherwise defined, the technical or scientific terms used in this invention description shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains.

[0076] Reference numerals: Electronic device 500 includes processor 501, communication interface 502, memory 503, and bus 504.

[0077] The present invention will now be described in further detail with reference to the accompanying drawings:

[0078] Example 1

[0079] Reference Figure 1 A vehicle crash test image motion analysis testing system, comprising:

[0080] Step S100, Pre-test preparation, including:

[0081] Step S101, Equipment selection and setup: Select a high-speed camera suitable for car crash tests, ensuring it has high resolution and high frame rate to meet the requirements for capturing fast-moving objects at the moment of collision.

[0082] A fixed-focus lens is used, with an aperture value of at least F5.6, preferably F8, so that the aperture value can control the depth of field while ensuring a certain amount of light intake, so that the target object is clearly imaged in the image.

[0083] The exposure time for the high-speed camera is set to no more than 200 μs to reduce motion blur.

[0084] Step S102, perform camera synchronization calibration and standardization, including:

[0085] Use a synchronization tester to calibrate the cameras, ensuring a synchronization accuracy of at least 10 μs between them. Specifically, connect the synchronization tester to all cameras participating in the test. Connect the trigger signal output port of the synchronization tester to the trigger input port of each camera, ensuring a secure connection. Start the synchronization tester and camera equipment, and set the synchronization mode and parameters on the synchronization tester's interface. Perform the synchronization calibration operation; the synchronization tester sends synchronization trigger signals to each camera, and the cameras simultaneously capture images upon receiving the signals. By detecting differences in camera shooting times, the synchronization tester automatically calculates and adjusts the synchronization error between cameras, ensuring a synchronization accuracy of at least 10 μs.

[0086] Step S103: Complete the in-camera orientation element calibration, including principal point, focal length, and distortion parameters. Specifically, prepare a calibration board for in-camera orientation element calibration. The calibration board has high-precision marker points with known sizes and positions. Place the calibration board within the camera's field of view, and adjust its position and angle so that the camera can capture all the marker points on the calibration board. Take multiple sets of images of the calibration board at different positions and angles, with each set containing at least 8 different orientations of the calibration board.

[0087] Import the captured calibration board image using professional camera calibration software. The software identifies the markers on the calibration board and uses algorithms to calculate the camera's interior orientation elements.

[0088] Step S104: Select an appropriate marker type and size based on the marker pixel size. For example, in a collision test, the camera resolution is 4096×2160, and the pixel size is 5μm. According to empirical formulas, a circular marker with a diameter of 20 pixels should be selected. This marker has sufficient recognizability in the image, while not being too large and affecting the measurement accuracy.

[0089] Choose high-contrast, easily identifiable marker materials, such as reflective stickers or fluorescent materials. Affix the markers to critical areas of the object being tested, such as the vehicle body or component connection points, ensuring the markers will not detach or be damaged during a collision. When affixing the markers, ensure they are accurately centered and adhere tightly to the surface of the object being tested.

[0090] Step S105: Control point coordinate calibration. The calibration accuracy of control points should be at least 10 times that of general marker point positioning. Specifically, determine the location of control points. Control points should be distributed within the camera's field of view and have high accuracy requirements. Use high-precision measuring equipment, such as a total station or laser tracker, to measure the three-dimensional coordinates of the control points.

[0091] Step S200: Based on the changes in the internal and external orientation elements of the image measurement equipment and the relevant information of the image measurement data channel, an online analysis method or an offline analysis method is adopted. The online analysis method is suitable for situations where the internal and external orientation elements of the image measurement equipment change, or where there is no relevant information about the image measurement data channel; the offline analysis method is suitable for situations where the equipment of the image measurement data channel remains unchanged or only undergoes minor changes.

[0092] Step S300, online analysis method, including:

[0093] Step S301: Prepare for the crash test by placing the test object (such as a vehicle or trolley) with the markings installed on it on the crash test stand and adjusting its position and orientation to meet the test requirements. Ensure that the test object is within the camera's field of view and that each camera can clearly capture the markings and control points on the test object.

[0094] Check the connections and parameter settings of all cameras, lenses, synchronization devices, and other related equipment to ensure they are functioning correctly. Double-check that the camera's exposure time, aperture value, synchronization accuracy, and interior orientation elements meet the test requirements.

[0095] Step S302: Start the collision test equipment and trigger the collision action. During the collision, all cameras, under the control of the synchronous detector, simultaneously capture motion images of the test object at a set frame rate.

[0096] In step S303, the image data captured by the camera is transmitted to the data acquisition module in real time. The data acquisition module stores and performs preliminary processing on the image data, extracting key information such as the location of marker points and motion trajectory.

[0097] Step S304 involves real-time analysis of the extracted image data, including:

[0098] Step S304-1: Calculate the performance of the image measurement data channel. The performance of the image measurement data channel is composed of different indices, as shown in Table 1. The selection of the indices depends on the application scenario (two-dimensional / three-dimensional).

[0099] Table 1 Performance Index

[0100]

[0101] Focal length index ( This is used to evaluate the impact of focal length error on positioning accuracy. The calculation formula is as follows:

[0102]

[0103] in, For theoretical positioning accuracy, For the first actual positioning accuracy, The calculation formula is:

[0104]

[0105] in, For object distance, Focal length For focal length accuracy.

[0106] Distortion index ( This is used to evaluate the impact of distortion parameters on positioning accuracy. The calculation formula is as follows:

[0107]

[0108] in, The second actual positioning accuracy is calculated using the following formula:

[0109]

[0110] in, For distortion accuracy, This refers to the pixel size.

[0111] Marker Recognition Index ( This is used to evaluate the impact of marker recognition accuracy on positioning accuracy. The calculation formula is:

[0112]

[0113] in, The actual diameter of the marker point and the theoretical diameter of the marker point are calculated using the following formulas:

[0114]

[0115] in, The diameter of the desired marker point.

[0116] Motion blur index ( The formula used to evaluate the impact of motion fuzziness on positioning accuracy is as follows:

[0117]

[0118] in, This represents the actual motion blur value. The calculation formula is:

[0119]

[0120] in, For the speed of movement, This refers to the exposure time.

[0121] Marker Motion Index (MMI) This is used to determine the required motion distance between two frames of the current marker point based on test requirements. The calculation formula is as follows:

[0122]

[0123] in, This represents the theoretical distance traveled. This represents the current distance traveled. The calculation formula is:

[0124]

[0125] in, Image rate.

[0126] Reference Figure 2 and Figure 3 Control point distribution index ( The control point distribution index determines the number of control points in different image regions i (i=1,2,3,4) and the coverage ratio of the regions where the control points are located. In 2D motion analysis where the camera is perpendicular to the motion plane, the control point distribution index is 1. Otherwise, the control point distribution index must be calculated using the following formula:

[0127]

[0128]

[0129] when hour,

[0130] when 10 % hour, 0

[0131]

[0132] in, The state of control point existence in image region i (i=1,2,3,4, when there is at least one control point in the region). (otherwise it is 0); For image width, For image height, The area of ​​the region formed by the control points; For control point distribution parameters, This refers to the area parameters of the control points.

[0133] Time base index ( This is used to evaluate the impact of time reference accuracy on positioning accuracy; the calculation formula is as follows:

[0134]

[0135] in, The third actual positioning accuracy is calculated using the following formula:

[0136]

[0137] in, This represents the total time offset. For the speed of movement, The calculation formula is:

[0138]

[0139] in, For each step size, the time offset is... For time intervals, For image resolution, The calculation formula is:

[0140]

[0141] in, For image resolution accuracy.

[0142] Initial time identification index ( This is used to evaluate the impact of initial identification accuracy on positioning accuracy, and the calculation formula is:

[0143]

[0144] in, The fourth actual positioning accuracy is calculated using the following formula:

[0145]

[0146] in, This is the time difference between the initial frame and the T0 signal.

[0147] Camera placement index ( This is only applicable to 2D motion analysis, used to evaluate the camera's orientation requirements (perpendicular or non-perpendicular) relative to a plane. When placed vertically... =1), the camera must be precisely perpendicular to the plane of motion. A non-perpendicular camera placement relative to the plane of motion is only permitted when all measured objects lie within the same plane. (When not perpendicular...) =0), the position and orientation of the camera relative to the plane of motion must be determined, and the measured values ​​must be corrected. At the same time, the requirements of the control point distribution index and the focal length index must be met.

[0148]

[0149] in, The orientation of the camera relative to the moving plane; The focal length index; This is the control point distribution index.

[0150] Motion plane scale parameters ( This applies only to 2D motion analysis and is used to define the required scale information for each motion plane. If the object being measured lies across multiple motion planes, scale information is needed for each plane. There are two ways to obtain scale information: using a reference distance within the plane, or using the precise distance between the motion plane and the reference plane.

[0151]

[0152]

[0153] in, The number of moving planes, The reference distance is for the motion plane i; Let i be the distance between the motion plane and the reference plane; For the proportional information of the motion plane i (when or hour, ;when and hour, ); Provides proportional information for all moving planes.

[0154] Interaction Index ( This is used to evaluate the required positioning accuracy of the measured object in the depth direction, and the worst triangulation configuration is generally selected. If this worst combination consists of three or more cameras, then the two best-performing cameras should be used. The calculation formula is as follows:

[0155]

[0156]

[0157]

[0158]

[0159] in, The distance from the camera base to the object being measured; The length of the camera base; For target recognition accuracy; and Indicates the distance between the target and the camera; and Indicates camera focal length, and Camera pixel size; This represents the theoretical positioning accuracy ratio of the Z-axis (depth of field) relative to the X / Y-axis. This represents the theoretical positioning accuracy in the X / Y directions (field of view). This represents the ratio of the actual positioning accuracy in the Z-axis (depth of field) to that in the X / Y-axis. Indicates the actual positioning accuracy in the X / Y directions; Indicates the actual positioning accuracy (Z-axis). Indicates the theoretical positioning accuracy (Z-axis).

[0160] synchronicity index ( This is used to evaluate the impact of camera asynchrony on positioning accuracy. Image coordinate values ​​can also be corrected by interpolation using known asynchrony levels, while the synchronization index uses the current asynchrony accuracy. The worst pair of cameras should be used to determine the synchronization index, calculated as follows:

[0161]

[0162]

[0163]

[0164]

[0165]

[0166]

[0167] in, Let v be the velocity of motion; Indicates the synchronization difference between cameras; This indicates the distance from the center of the camera base to the object being measured; The length of the camera base; For target recognition accuracy; and Indicates the distance between the target and the camera; and Indicates camera focal length, and Camera pixel size; This represents the theoretical positioning accuracy ratio of the Z-axis (depth of field) relative to the X / Y-axis. This represents the theoretical positioning accuracy in the X / Y directions (field of view). This represents the ratio of the actual positioning accuracy in the Z-axis (depth of field) to that in the X / Y-axis. Indicates the actual positioning accuracy in the X / Y directions; Indicates the actual positioning accuracy (Z-axis). Indicates the theoretical positioning accuracy (Z-axis).

[0168] Evaluation of two-dimensional and three-dimensional performance values ​​includes:

[0169] Two-dimensional performance values ​​are used to describe the performance of a single image recording device. The two-dimensional performance value of image recording device i... The calculation formula is:

[0170]

[0171] in, The serial number of the image recording device; The number of two-dimensional performance indices; Image recording device The actual value of the two-dimensional performance index; The number of two-dimensional performance indices (for 2D image motion analysis, =11; For 3D image motion analysis ).

[0172] Three-dimensional performance values ​​are used to describe the performance of each image recording device in the image measurement data channel. The calculation formula is:

[0173]

[0174] in, This refers to the sequence number of the three-dimensional performance index. Three-dimensional performance index The actual value, The number of three-dimensional performance indices (default value) =2).

[0175] Step S304-2: Calculate the accuracy of the image measurement data channel. The accuracy indicators of the image measurement data channel include the camera position calculation index and the scaling index, including:

[0176] Camera position calculation index ( The method for calculating the camera position determines the impact on position accuracy. The calculation rules are as follows:

[0177] If the camera position needs to be determined dynamically, then , ;

[0178] If the camera position only needs to be determined from a single image or does not need to be determined at all (2D analysis), then:

[0179]

[0180]

[0181]

[0182] in, Method for determining camera position; The maximum displacement of a fixed point in image space; For pixel size cs, The distance between the fixed point and the camera; Focal length; Theoretical positioning accuracy; This represents the maximum displacement of a fixed point in the object's space.

[0183] proportional parameters ( It is used to evaluate the requirement for the existence of reference distances in different directions in object space.

[0184] =1 indicates that there is a reference distance in the i direction; =0 indicates that there is no reference distance in the i direction;

[0185]

[0186] in, This indicates the existence of the reference distance in the i-th direction.

[0187] Calculate the error and accuracy of a single reference distance measurement, including:

[0188] If the camera position calculation index of all image recording devices If all requirements are met (≥1), the reference distance can be determined using any single time step within the analysis time interval. Measurement error .

[0189]

[0190] in, To analyze any single time step within the time interval, For reference distance The calibration length.

[0191] If there are any camera position calculation indicators for image recording devices If the requirement is not met (<1), then the length measurement error for each time step must be calculated within the analysis time interval. .

[0192]

[0193] in, To determine the initial time of the analysis time interval, This is the end time of the analysis time interval.

[0194] Reference distance precision value It is a length measurement error With calibration length The ratio between them:

[0195]

[0196] Calculate the measurement error and accuracy of the image measurement channel, including:

[0197] Measurement error of image measurement channel Take all reference distance measurement errors The maximum value in:

[0198]

[0199] Precision value of image measurement channel Take all reference distance accuracy values The maximum value in:

[0200] .

[0201] Step S304-3: Evaluate whether the performance and accuracy indicators meet the requirements. Specific requirements are as follows:

[0202] All performance metrics involved must be no less than 0.5;

[0203] The performance value of the image measurement data channel should be greater than 0.7;

[0204] Image measurement data channel length measurement error Must be lower than position accuracy ;

[0205] Precision value of image measurement data channel It must be below the precision limit.

[0206] Step S400, offline analysis method. In the offline analysis method, the performance index is calculated only once in the early stage of testing. In the subsequent collision test, only the accuracy needs to be calculated and the overall result of the image measurement data channel is evaluated.

[0207] Step S401: Following the same requirements as the online analysis method, set the parameters for the camera and lens, including exposure time and aperture value. In this embodiment, the high-speed camera exposure time is set to 150μs, and the lens aperture value is set to F8; use a synchronization detector to complete camera synchronization calibration and ensure synchronization accuracy is not less than 10μs. Perform in-camera orientation element calibration and record the calibration results.

[0208] Step S402: Prepare for the collision test by setting up a scenario similar to the subsequent collision test, including placing the simulated test object, attaching markers to its surface, and setting control points. Ensure that the imaging rate of the camera, the light field environment, and the measurement space area remain basically consistent with the subsequent test. The light field environment simulates the lighting conditions of the collision test site through lighting arrangements, and the size and shape of the measurement space area are the same as the actual test site. Start the camera to capture multiple sets of image data.

[0209] Step S403, Performance Index Calculation and Evaluation: The acquired image data is processed to calculate all performance indicators. The calculation method is the same as the online analysis method. However, it only needs to be calculated once during the initial testing phase to evaluate whether the performance indicators meet the following requirements, including:

[0210] All performance metrics involved must be no less than 0.5;

[0211] The performance value of the image measurement data channel is greater than 0.8;

[0212] Image measurement data channel length measurement error Must be lower than position accuracy ;

[0213] Precision value of image measurement data channel It must be below the precision limit.

[0214] Step S404: If the performance indicators meet the requirements, then in subsequent crash tests, only the accuracy needs to be calculated and the overall results evaluated. The accuracy evaluation requirements in subsequent crash tests are as follows:

[0215] For 3D motion analysis, check synchronization and ensure the synchronization index. ≥ 1;

[0216] Image measurement data channel length measurement error Must be lower than position accuracy ;

[0217] Precision value of image measurement data channel It must be below the precision limit.

[0218] In step S405, if the accuracy is found to be unsatisfactory, troubleshooting and repair are required, and the test must be repeated.

[0219] This disclosure also provides a vehicle crash test image motion analysis testing system, which employs a vehicle crash test image motion analysis testing method.

[0220] This disclosure also provides a storage medium storing a computer program. When the computer program is executed by a processor, it can implement all the steps of the above-described method for motion analysis of automobile crash test images.

[0221] Those skilled in the art will understand that implementing all or part of the process in a vehicle crash test image motion analysis testing method can be accomplished by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium. When executed, the program can include the processes of various embodiments of the vehicle crash test image motion analysis testing method. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0222] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the aforementioned method for motion analysis of automotive crash test images. In this application embodiment, the processor is the control center of the computer system; it can be a physical machine processor or a virtual machine processor.

[0223] Reference Figure 4 The electronic device 500 includes at least one processor 501, at least one communication interface 502, at least one memory 503, and at least one bus 504. The bus 504 is used for communication between these components, the communication interface 502 is used for signaling or data communication with other node devices, and the memory 503 stores machine-readable instructions executable by the processor 501. When the electronic device 500 is running, the processor 501 communicates with the memory 503 via the bus 504. When the machine-readable instructions are invoked by the processor 501, they execute the steps of the aforementioned automotive crash test image motion analysis test method.

[0224] The above content is merely an embodiment of the present invention. Commonly known structures and characteristics of the solutions are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are aware of all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can improve and implement this solution based on the guidance provided in this application and their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A method for motion analysis testing of images from automobile crash tests, characterized in that: include: Preparing for the test includes: Select a high-speed camera and set the lens aperture and camera exposure time; Use a synchronization detector to calibrate the cameras to ensure that the synchronization accuracy between the cameras meets the requirements. Complete the calibration of the camera's internal orientation elements, including the principal point, focal length, and distortion parameters; Based on the pixel size of the marker, select the marker type and size, and paste the marker on the key parts of the object being tested, ensuring that the center position of the marker is accurate and that it is tightly attached to the surface of the object being tested. Perform control point coordinate calibration by measuring the three-dimensional coordinates of the control points using measuring equipment; Based on the changes in the internal and external reverse station elements of the image measurement equipment and the correlation of the image measurement data channel, select either an online analysis method or an offline analysis method. Online analysis methods include: Prepare for and initiate the collision test; collect and preliminarily process data to extract key information; based on the key information, calculate the performance and accuracy of the image measurement data channel, and evaluate whether the performance and accuracy indicators meet the requirements. Offline analysis methods include: Preliminary testing was conducted, collision test preparations were made and the collision test was initiated, the performance of the image measurement data channel was calculated, the performance indicators were evaluated to determine whether they met the requirements, and the imaging rate, light field environment and measurement space area parameters of the camera in subsequent tests were kept basically consistent. Subsequent collision tests were conducted to calculate the accuracy of the image measurement data channel and assess whether the accuracy indicators met the requirements. The performance metrics for the image measurement data channel include the performance index of a single image recording device, including: focal length index (to evaluate the impact of focal length error on positioning accuracy); distortion index (to evaluate the impact of distortion parameters on positioning accuracy); marker recognition index (to evaluate the impact of marker recognition accuracy on positioning accuracy); motion blur index (to evaluate the impact of motion blur on positioning accuracy); marker motion index (to determine the required motion distance of the current marker between two frames based on test requirements); control point distribution index (to determine the number of control points in different image regions i and the coverage ratio of the regions where the control points are located); time base index (to evaluate the impact of time base accuracy on positioning accuracy); initial moment recognition index (to evaluate the impact of initial moment recognition accuracy on positioning accuracy); camera placement index (only applicable to two-dimensional motion analysis, used to evaluate the camera's orientation requirements relative to the plane); and motion plane scale parameter (only applicable to two-dimensional motion analysis, used to clarify the scale information requirements of each motion plane). The performance metrics of the image measurement data channel also include the performance index between image recording devices, including the interaction index, which is used to evaluate the positioning accuracy requirements of the measured object in the depth direction; and the synchronization index, which is used to evaluate the impact of asynchrony between cameras on positioning accuracy.

2. The method for motion analysis of automobile crash test images according to claim 1, characterized in that: Using a synchronization detector to calibrate the cameras ensures that the synchronization accuracy between cameras meets the requirements, including the following: Connect all cameras involved in the test using a synchronization detector. Connect the trigger signal output port of the synchronization detector to the trigger input port of each camera. Start the synchronization detector and camera equipment. Set the synchronization mode and parameters on the operation interface of the synchronization detector and perform synchronization calibration.

3. The method for motion analysis of automobile crash test images according to claim 2, characterized in that: In camera synchronization calibration and standardization, the synchronization detector sends a synchronization trigger signal to each camera. After receiving the signal, the cameras take pictures simultaneously. The synchronization detector automatically calculates and adjusts the synchronization error between the cameras by detecting the difference in the shooting time of the cameras, so that the synchronization accuracy reaches the requirement of not less than 10μs.

4. The method for motion analysis testing of automobile collision test images according to claim 1, characterized in that: Control point coordinate calibration includes the following: Prepare a calibration board with high-precision markers and known size and position. Place the calibration board within the camera's field of view, adjust its position and angle, and take multiple sets of images at different positions and angles, with each set containing at least 8 different positions of the calibration board's posture. Import the captured calibration board images using camera calibration software and calculate the camera's interior orientation elements.

5. The method for motion analysis of vehicle collision test images according to claim 1, characterized in that: The online analysis method is applicable when the orientation elements inside and outside the image measurement device change, or when there is no image measurement data channel related information; the offline analysis method is applicable when the image measurement data channel device remains unchanged or only undergoes minor changes.

6. The method for motion analysis of vehicle collision test images according to claim 1, characterized in that: Evaluate whether the two-dimensional and three-dimensional performance values ​​meet the requirements, including the following: Two-dimensional performance values ​​are used to describe the performance of a single image recording device. The two-dimensional performance value of image recording device i... The calculation formula is: in, The serial number of the image recording device; The number of two-dimensional performance indices; Image recording device The actual value of the two-dimensional performance index; The number of two-dimensional performance indices; Three-dimensional performance values ​​are used to describe the performance of each image recording device in the image measurement data channel. The calculation formula is: in, This refers to the sequence number of the three-dimensional performance index. Three-dimensional performance index The actual value, This represents the number of three-dimensional performance indices.

7. The method for motion analysis of automobile crash test images according to claim 6, characterized in that: The accuracy of the image measurement data channel is calculated. The accuracy indicators of the image measurement data channel include the camera position calculation index and the scale index. The camera position calculation index determines the impact of the camera position calculation method on the position accuracy, and the scale parameter is used to evaluate the existence requirements of reference distances in different directions in the object space. Calculate the error and accuracy of a single reference distance measurement, including: If the camera position calculation index of all image recording devices All requirements are met, and the reference distance is determined by any single time step within the analysis time interval. Measurement error ; in, To analyze any single time step within the time interval, For reference distance The calibration length; If there are any camera position calculation indicators for image recording devices If the requirements are not met, the length measurement error for each time step must be calculated within the analysis time interval. ; in, To determine the initial time of the analysis time interval, This represents the end time of the analysis time interval; Reference distance precision value It is a length measurement error With calibration length The ratio between them: Calculate the measurement error and accuracy of the image measurement channel, including: Measurement error of image measurement channel Take all reference distance measurement errors The maximum value in: Precision value of image measurement channel Take all reference distance accuracy values The maximum value in: 。 8. The method for motion analysis of vehicle collision test images according to claim 1, characterized in that: The evaluation requirements for the performance and accuracy indicators of the online analysis method are as follows: All performance metrics involved must be no less than 0.5; The performance value of the image measurement data channel should be greater than 0.7; Image measurement data channel length measurement error Must be lower than position accuracy ; Precision value of image measurement data channel It must be lower than the precision limit; The accuracy evaluation requirements for the offline analysis method in the initial testing are as follows: All performance metrics involved must be no less than 0.5; The performance value of the image measurement data channel is greater than 0.8; Image measurement data channel length measurement error Must be lower than position accuracy ; Precision value of image measurement data channel It must be lower than the precision limit; The accuracy assessment requirements for subsequent crash tests are as follows: For 3D motion analysis, check synchronization and ensure the synchronization index. ≥ 1; Image measurement data channel length measurement error Must be lower than position accuracy ; Precision value of image measurement data channel It must be below the precision limit.

9. A vehicle crash test image motion analysis testing system, characterized in that: The method for analyzing motion in images from a car crash test, as described in any one of claims 1-8, is employed.

Images