Blue light high-temperature 3D-DIC video extensometer based on polarization imaging and measurement method
By adopting polarization imaging technology and blue light illumination in high-temperature DIC systems, the problems of optical noise interference and low measurement accuracy in high-temperature environments are solved, and high-precision three-dimensional displacement and strain measurement are achieved, reducing system costs.
Patent Information
- Application Number
- CN202411554697.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-04
- Publication Date
- 2024-11-29
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing high-temperature DIC systems face ambient optical noise interference and limited high-temperature imaging technology under high-temperature environments, resulting in low measurement accuracy and high cost.
Using a blue light high-temperature 3D-DIC video extensometer based on polarization imaging, data is captured from different viewing angles through two polarization cameras to reduce background light interference, and using narrowband pass filters and blue polarization LED light sources to improve image clarity and contrast. Accurate measurements are made by computer-generated three-dimensional models.
It realizes providing ultra-high-definition images and ultra-high-precision measurement results in high-temperature environments, avoiding the external force influence of traditional contact measurements and measurement blind spots caused by single-view occlusion, and reducing costs.
Smart Images

Figure CN119043876A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of in-situ testing instruments for mechanical properties of materials, and in particular relates to a blue light high temperature 3D-DIC video extensometer based on polarization imaging and a measuring method. Background Art
[0002] With the continuous development of modern industrial technology, the mechanical properties test of materials under high temperature conditions has gradually become the focus of scientific research and industry. Especially in the fields of aerospace, automobile manufacturing and energy industry, the mechanical properties test of high temperature materials is crucial to ensure the safety, reliability and durability of materials.
[0003] The functions of the video extensometer are divided into two measurement methods: contact and non-contact. The non-contact extensometer has significant advantages in the mechanical properties test of high-temperature materials. First, it can be measured in harsh environments such as high temperature and corrosion without direct contact with the object being measured, thus avoiding the damage of high temperature to the equipment and the influence of high temperature on the measurement results. Secondly, the non-contact extensometer uses optical means to perform long-distance measurement without applying any external force to the test piece, ensuring that the natural deformation and strain behavior of the material can be truly reflected. In addition, it can achieve high-resolution measurement from millimeters to microns, adapt to complex geometric structures, and is widely used in the measurement needs of various materials and shapes. Since there is no need for physical contact, the non-contact extensometer can also achieve fast data acquisition, which is suitable for real-time monitoring of dynamic deformation and strain, greatly improving the test efficiency and data accuracy. In summary, the non-contact extensometer provides higher flexibility and measurement accuracy in a variety of application scenarios and is an important tool in modern material mechanics testing.
[0004] Traditional displacement and strain measurement technologies often face many challenges under high temperature conditions. High temperature environments can cause conventional sensors to fail, and changes in the optical properties of the material surface can also affect the measurement accuracy. In order to meet these challenges, Digital Image Correlation (DIC) technology has been widely used in recent years. DIC technology can achieve non-contact, high-precision measurement of material surface displacement and strain through image capture and analysis. However, the application of existing DIC systems under high temperature conditions still has the following shortcomings: 1) Environmental optical noise interference: In a high temperature environment, the optical noise on the material surface will increase due to thermal radiation and high temperature scattering. Ordinary polarization camera solutions cannot normally capture clear images of the test samples, and subsequent data processing cannot guarantee measurement accuracy.
[0005] 2) Limited high temperature imaging technology: Currently, DIC systems used in high temperature environments mostly use infrared imaging or ultraviolet imaging, but the resolution and contrast of infrared imaging often cannot meet the requirements of fine measurement, and the cost of using ultraviolet polarization cameras is also greatly increased. Summary of the invention
[0006] The present invention provides a blue light high temperature 3D-DIC video extensometer based on polarization imaging on the one hand, and provides a measurement method of the blue light high temperature 3D-DIC video extensometer based on polarization imaging on the other hand. The present invention reduces the influence of background light in a high temperature furnace by using a polarization camera, and adopts two perspectives to capture data, thereby reducing the measurement blind area caused by occlusion of a single perspective, better distinguishing the characteristics of the target object, improving the clarity and contrast of the image, and ensuring the provision of ultra-high definition images and ultra-high precision measurement results in an ultra-high temperature environment.
[0007] The present invention is implemented as follows: a blue light high temperature 3D-DIC video extensometer based on polarization imaging, A blue light high temperature 3D-DIC video extensometer based on polarization imaging, comprising: High temperature test machine, two polarization cameras, narrow bandpass filters, blue polarized LED light source and computer; including: The high temperature testing machine comprises a high temperature furnace arranged in a shell, a support platform and a load applying device are arranged inside the furnace body of the high temperature furnace, the test piece is placed between the support platform and the load applying device, the load applying device comprises a pressure rod passing through the top of the furnace body and then connected to a counterweight block; a quartz glass window is opened on the shell; two polarization cameras, capturing images of the test piece from different viewing angles, with polarizers installed in front of the zoom polarization lenses of the polarization cameras; Narrow bandpass filters, placed before each polarizer; The blue polarized LED light source is arranged in front of the quartz glass window, and is used to emit blue light and illuminate the test piece through the quartz glass window; The computer receives two sets of images from two polarization cameras, matches the two sets of images, generates a disparity map, converts the disparity map into three-dimensional point cloud data, reconstructs a three-dimensional model of the test piece, and calculates the three-dimensional displacement field and three-dimensional strain field of the test piece based on the three-dimensional model.
[0008] Furthermore, the computer searches for the best match of a certain pixel point in an image taken by one of the polarization cameras at a corresponding moment in a window around the certain pixel point in an image taken by another polarization camera by using a block matching algorithm in a stereo matching algorithm, and after completing the stereo matching, calculates the disparity value of each pixel point to form a disparity map; The disparity map is converted into three-dimensional point cloud data through triangulation method to reconstruct the three-dimensional model of the object, including: using the disparity value in the disparity map as well as the focal length and the baseline length between the two polarization cameras to calculate the depth of each point in the captured scene, generate a depth map, create and operate three-dimensional point cloud data, calculate the three-dimensional coordinates using the depth map, and reconstruct the three-dimensional model.
[0009] Further, creating and manipulating three-dimensional point cloud data includes: performing point cloud denoising and filtering on the three-dimensional point cloud data; To align the point clouds, the iterative closest point algorithm is used to iteratively adjust the transformation matrix between the two point clouds to minimize the distance between the point clouds and achieve alignment; A smooth three-dimensional surface is generated by point cloud smoothing and interpolation, and the surface reconstruction method Poisso is used in combination with the depth map to generate a complete three-dimensional model.
[0010] Furthermore, calculating the three-dimensional displacement field and three-dimensional strain field of the test piece includes: A plurality of image correlation blocks are defined on the three-dimensional point cloud data, wherein the image correlation blocks are composed of a plurality of pixels and represent a local area on the three-dimensional surface, and the local area is used as a sub-area, and each sub-area is used as a basic unit of DIC analysis to track the displacement and deformation of the sub-area at different time points; In the three-dimensional point cloud data of different time frames, the matching position of each relevant block is calculated by the 3D-DIC algorithm, and the displacement is calculated with sub-pixel accuracy in combination with the cubic spline interpolation algorithm to obtain the displacement data; Calculate the displacement vector of each image-related block based on the displacement data; According to the displacement vector, a three-dimensional displacement field of the entire test piece is generated; Perform differential operation on the three-dimensional displacement field to calculate the three-dimensional strain field of the local area, and use the three-dimensional strain field to analyze the deformation behavior of the test piece under external force or thermal load; According to the three-dimensional strain field, the strain data is mapped to the three-dimensional model to generate an intuitive strain distribution diagram.
[0011] Furthermore, the strain data is mapped using colors to indicate the magnitude and direction of the strain.
[0012] A measurement method of a blue light high temperature 3D-DIC video extensometer based on polarization imaging, the method comprising: The S1 test machine is set up with a high temperature test environment, which meets the required temperature and load conditions; S2 adjusts the focal length and angle of the two polarization cameras so that the distance between the two polarization cameras is 10-20 times the size of the test piece, adjusts the viewing angle difference of the two polarization cameras to between 15-30 degrees, and aligns the optical axes of the two polarization cameras with the center of the test piece; S3 starts the blue polarized LED light source and installs a narrow bandpass filter in front of the polarizer; S4 calibrates the tilt angles of the two polarization cameras using a calibration plate to remove image distortion caused by tilt; The two polarization cameras of S5 start shooting at the same time, capturing images of the surface of the test piece; S6 receives two sets of images from two polarization cameras, matches the two sets of images, generates a disparity map, converts the disparity map into three-dimensional point cloud data, reconstructs the three-dimensional model of the test piece, and calculates the three-dimensional displacement field and three-dimensional strain field of the test piece.
[0013] Furthermore, S6 includes: Through the block matching algorithm in the stereo matching algorithm, the best match of the same window in an image taken by another polarization camera at the corresponding moment is found in the window around a certain pixel point in an image taken by one of the polarization cameras. After completing the stereo matching, the disparity value of each pixel point is calculated to form a disparity map. The disparity map is converted into three-dimensional point cloud data through triangulation method to reconstruct the three-dimensional model of the object, including: using the disparity value in the disparity map as well as the focal length and the baseline length between the two polarization cameras to calculate the depth of each point in the captured scene, generate a depth map, create and operate three-dimensional point cloud data, calculate the three-dimensional coordinates using the depth map, and reconstruct the three-dimensional model.
[0014] Further, it includes: performing point cloud denoising and filtering on the three-dimensional point cloud data; To align the point clouds, the iterative closest point algorithm is used to iteratively adjust the transformation matrix between the two point clouds to minimize the distance between the point clouds and achieve alignment; A smooth three-dimensional surface is generated by point cloud smoothing and interpolation, and the surface reconstruction method Poisso is used in combination with the depth map to generate a complete three-dimensional model.
[0015] Furthermore, calculating the three-dimensional displacement field and three-dimensional strain field of the test piece includes: A plurality of image correlation blocks are defined on the three-dimensional point cloud data, wherein the image correlation blocks are composed of a plurality of pixels and represent a local area on the three-dimensional surface, and the local area is used as a sub-area, and each sub-area is used as a basic unit of DIC analysis to track the displacement and deformation of the sub-area at different time points; In the three-dimensional point cloud data of different time frames, the matching position of each relevant block is calculated by the 3D-DIC algorithm, and the displacement is calculated with sub-pixel accuracy in combination with the cubic spline interpolation algorithm to obtain the displacement data; Calculate the displacement vector of each image-related block based on the displacement data; According to the displacement vector, a three-dimensional displacement field of the entire test piece is generated; Perform differential operation on the three-dimensional displacement field to calculate the three-dimensional strain field of the local area, and use the three-dimensional strain field to analyze the deformation behavior of the test piece under external force or thermal load; According to the three-dimensional strain field, the strain data is mapped to the three-dimensional model to generate an intuitive strain distribution diagram.
[0016] Furthermore, the strain data is mapped using colors to indicate the magnitude and direction of the strain.
[0017] Compared with the prior art, the present invention has the following beneficial effects: Traditional contact measurement methods easily impose additional loads on the test piece, affecting the natural deformation behavior of the material, and thus affecting the accuracy of the measurement results. The present invention adopts a non-contact optical measurement method, which does not apply any external force to the test piece, ensuring that the natural state of the material is preserved and the measurement results are more realistic.
[0018] In a high temperature environment, the surface of the material will emit a large amount of thermal radiation and background light. Traditional imaging technology is easily affected by these interferences, resulting in low image contrast, loss of details, and affecting measurement accuracy. The present invention uses a polarization camera and a polarized light source to receive the polarization information reflected by the test piece and image it, filter out non-polarized scattered light and radiated light, suppress background light interference in the high-temperature furnace, and improve image contrast and clarity. In addition, under polarized light irradiation, the reflected light of different materials carries different polarization information, which can better distinguish the characteristics of the target object.
[0019] Traditional single-view DIC systems are prone to occlusion when measuring complex geometric structures, which results in partial surfaces being unable to be observed, causing data loss and measurement blind spots. The present invention uses two polarization cameras to synchronously capture images from different perspectives, which can fully cover the surface of the test piece, remove image distortion caused by tilt, and avoid measurement blind spots caused by single-view occlusion.
[0020] Polarization imaging technology uses the polarization characteristics of light to effectively reduce the interference of optical noise and thermal radiation on the image, and improve the imaging contrast and resolution in high-temperature environments. Compared with infrared imaging, blue light imaging has higher optical resolution and imaging accuracy, and is suitable for precision measurement in high-temperature environments. In addition, polarization imaging is combined with digital image correlation technology to capture data from two perspectives, reduce the measurement blind area caused by occlusion of a single perspective, and generate a high-precision three-dimensional model. Finally, the cost of polarization cameras is low, which is suitable for the promotion and use of this extensometer in the industry. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] Figure 1 It is a schematic structural diagram of an extensometer provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0022] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0023] Example 1, see Figure 1 As shown, the present invention provides a blue light high temperature 3D-DIC video extensometer based on polarization imaging, comprising: a high temperature testing machine 8, two polarization cameras 2, a narrow bandpass filter 5, a blue polarized LED light source 10 and a computer 1; wherein: The high temperature testing machine 8 includes a high temperature furnace 9 arranged in a shell, a support platform and a load applying device are arranged inside the furnace body of the high temperature furnace 9, a test piece 7 is placed between the support platform and the load applying device, the load applying device includes a pressure rod passing through the top of the furnace body and connected to a counterweight block 6; a set pressure is applied to the test piece 7 through the load applying device, and a quartz glass window is opened on the shell; photography and observation can be carried out through the quartz glass window.
[0024] Install the test piece 7 on the high temperature test machine 8, making sure it is in the proper position and can withstand the high temperature and mechanical load. Then set the required temperature and load conditions to simulate the actual working environment of the material. Start the high temperature test machine 8 and make it reach the predetermined temperature, the maximum temperature of which can reach 2300°C.
[0025] Two polarization cameras 2 are used to capture images of the test piece 7 from different viewing angles. Polarizers 4 are installed in front of the zoom polarization lenses of the polarization cameras 2. A narrow bandpass filter 5 is arranged in front of each polarizer 4. A blue polarized LED light source 10 is arranged between the two polarization cameras 2 and the quartz glass window, which is used to emit blue light 11 and illuminate the test piece 7 through the quartz glass window; at this time, the test piece 7 reflects the blue light. The blue light passes through the narrow band pass filter 5 and the polarizer 4, and then filters out the light of other colors, ensuring that the polarization camera 2 captures the blue light.
[0026] A computer 1 is set up to receive two sets of images from two polarization cameras 2, match the two sets of images, generate a disparity map, convert the disparity map into three-dimensional point cloud data, reconstruct the three-dimensional model of the test piece 7, and calculate the three-dimensional displacement field and three-dimensional strain field of the test piece 7 based on the three-dimensional model.
[0027] The computer 1 uses a block matching algorithm in a stereo matching algorithm to find the best match of the same window in an image taken by another polarization camera 2 at a corresponding moment in a window around a certain pixel point in an image taken by one of the polarization cameras 2. After completing the stereo matching, the disparity value of each pixel point is calculated to form a disparity map. The disparity map is converted into three-dimensional point cloud data through triangulation method to reconstruct the three-dimensional model of the object, including: using the disparity value in the disparity map and the focal length and the baseline length between the two polarization cameras 2 to calculate the depth of each point in the captured scene and generate a depth map. The disparity map is a grayscale image. The brightness of each pixel in the image represents the disparity value of the point. The cv2.StereoBM() function in OpenCV is used to implement this algorithm and generate a disparity map. Create and manipulate three-dimensional point cloud data, use the depth map to calculate the three-dimensional coordinates, and reconstruct the three-dimensional model. Among them, the np.zeros_like() function is used to generate a depth map, and various functions in Open3D are used to create and manipulate point cloud data, and the three-dimensional coordinates are calculated using the depth map. Among them, creating and operating 3D point cloud data includes: performing point cloud denoising and filtering on 3D point cloud data; performing point cloud denoising and filtering is used to improve data quality and reduce computational complexity. Using voxel filtering, it reduces the point cloud density by dividing the point cloud into voxel grids and aggregating the points of each voxel grid into one point, and then statistically removing outliers, removing noise by calculating the number of neighboring points of each point and removing points whose number of neighboring points is lower than a specified threshold, and then performing radius outlier removal, removing outliers according to the number of neighboring points within a specified radius; To align the point clouds, the iterative closest point algorithm is used to iteratively adjust the transformation matrix between the two point clouds to minimize the distance between the point clouds and achieve alignment; A smooth three-dimensional surface is generated by point cloud smoothing and interpolation, and the surface reconstruction method Poisso is used in combination with the depth map to generate a complete three-dimensional model.
[0028] The three-dimensional displacement field and three-dimensional strain field of the test piece 7 calculated by the computer 1 include: A plurality of image correlation blocks are defined on the three-dimensional point cloud data, wherein the image correlation blocks are composed of a plurality of pixels and represent a local area on the three-dimensional surface, and the local area is used as a sub-area, and each sub-area is used as a basic unit of DIC analysis to track the displacement and deformation of the sub-area at different time points; In the three-dimensional point cloud data of different time frames, the matching position of each relevant block is calculated by the 3D-DIC algorithm, and the displacement is calculated with sub-pixel accuracy in combination with the cubic spline interpolation algorithm to obtain the displacement data; Calculate the displacement vector of each image-related block based on the displacement data; According to the displacement vector, a three-dimensional displacement field of the entire test piece 7 is generated; the three-dimensional displacement field will reflect the actual movement of the test piece in the three-dimensional space.
[0029] Perform differential calculation on the three-dimensional displacement field to calculate the three-dimensional strain field in the local area, and use the three-dimensional strain field to analyze the deformation behavior of the test piece under external force or thermal load; According to the three-dimensional strain field, the strain data is mapped to the three-dimensional model to generate an intuitive strain distribution diagram.
[0030] The strain data is mapped using colors to indicate the magnitude and direction of the strain.
[0031] Error detection. The 3D-DIC algorithm automatically detects and corrects errors introduced by factors such as noise and loss of markers, and optimizes through multiple iterations to ensure data accuracy and reliability; Based on the three-dimensional displacement field and three-dimensional strain field calculated by the 3D-DIC algorithm, a deformation analysis report of the test piece 7 is generated, including results such as strain distribution diagram, displacement curve, and stress-strain relationship.
[0032] Embodiment 2, a measurement method of a blue light high temperature 3D-DIC video extensometer based on polarization imaging, the method comprising: S1 Testing machine sets up a high temperature test environment, and the high temperature test environment reaches the required temperature and load conditions; install the test piece 7 on the high temperature testing machine 8, ensuring that it is in a proper position and can withstand high temperature and mechanical load. Then set the required temperature and load conditions to simulate the actual working environment of the material. Start the high temperature testing machine 8 to make it reach the predetermined temperature, which can reach up to 2300°C; S2 adjusts the focal length and angle of the two polarization cameras 2 so that the distance between the two polarization cameras 2 is 10-20 times the size of the test piece 7, adjusts the viewing angle difference of the two polarization cameras 2 to be between 15-30 degrees, and aligns the optical axes of the two polarization cameras 2 with the center of the test piece 7; S3 starts the blue polarized LED light source 10 and installs the narrow bandpass filter 5 in front of the polarizer 4; starts the blue polarized LED light source 10 to provide stable blue light illumination. Installing the narrow bandpass filter 5 in front of the polarizer 4 ensures that only blue light is allowed to enter the polarization camera 2, thereby reducing interference from other light sources in the high temperature environment.
[0033] S4 performs tilt angle calibration on the two polarization cameras 2 through the calibration plate to remove the image distortion caused by the tilt; this step can be achieved by using conventional existing methods. The pictures taken by the two polarization cameras 2 are calibrated through the computer 1 vision module of Camera Calibration in Matlab. The captured images are uploaded to the software, the images with large errors are deleted, and the calibration can be started by selecting the size of the calibration plate. The calculation is repeated. The average calibration error is less than 0.3 pixels. The parameters of the polarization camera 2 are available, and the intrinsic and extrinsic parameters of the polarization camera 2 are obtained. The intrinsic parameters are affected by factors such as the optical design of the polarization camera 2, the lens manufacturing process, and the lens installation error. Using the obtained intrinsic parameters, the image is corrected using the 'cv2.calibrateCamera()' and 'cv2.undistort()' functions in the OpenCV software to complete the radial and tangential distortion correction of the image; The external parameters describe the attitude and position of the polarization camera 2 relative to the world coordinate system, including the rotation matrix and the translation vector. Using the obtained external parameters, the 'cv2.warpPerspective()' function in the OpenCV software is used to perform image geometric transformation, correct the geometric distortion in the image, and remove the image distortion caused by tilt; Considering the particularity of blue light imaging in high temperature environments, this rotation correction may require multiple iterations to achieve the best correction effect.
[0034] S5: the two polarization cameras 2 start shooting at the same time to capture the image of the surface of the test piece 7; S6 receives two sets of images from two polarization cameras 2, matches the two sets of images, generates a disparity map, converts the disparity map into three-dimensional point cloud data, reconstructs the three-dimensional model of the test piece 7, and calculates the three-dimensional displacement field and three-dimensional strain field of the test piece 7.
[0035] S6 includes: Using the block matching algorithm in the stereo matching algorithm, in a window around a certain pixel point in an image taken by one of the polarization cameras 2, the best match of the same window in an image taken by another polarization camera 2 at the corresponding moment is found. After completing the stereo matching, the disparity value of each pixel point is calculated to form a disparity map; The disparity map is converted into three-dimensional point cloud data through triangulation method to reconstruct the three-dimensional model of the object, including: using the disparity value in the disparity map and the focal length and the baseline length between the two polarization cameras 2 to calculate the depth of each point in the captured scene, generate a depth map, create and operate the three-dimensional point cloud data, calculate the three-dimensional coordinates using the depth map, and reconstruct the three-dimensional model.
[0036] Further, it includes: performing point cloud denoising and filtering on the three-dimensional point cloud data; To align the point clouds, the iterative closest point algorithm is used to iteratively adjust the transformation matrix between the two point clouds to minimize the distance between the point clouds and achieve alignment; A smooth three-dimensional surface is generated by point cloud smoothing and interpolation, and the surface reconstruction method Poisso is used in combination with the depth map to generate a complete three-dimensional model.
[0037] Furthermore, calculating the three-dimensional displacement field and the three-dimensional strain field of the test piece 7 includes: A plurality of image correlation blocks are defined on the three-dimensional point cloud data, wherein the image correlation blocks are composed of a plurality of pixels and represent a local area on the three-dimensional surface, and the local area is used as a sub-area, and each sub-area is used as a basic unit of DIC analysis to track the displacement and deformation of the sub-area at different time points; In the three-dimensional point cloud data of different time frames, the matching position of each relevant block is calculated by the 3D-DIC algorithm, and the displacement is calculated with sub-pixel accuracy in combination with the cubic spline interpolation algorithm to obtain the displacement data; Calculate the displacement vector of each image-related block based on the displacement data; Based on the displacement vector, a three-dimensional displacement field of the entire test piece 7 is generated; Perform differential operation on the three-dimensional displacement field to calculate the three-dimensional strain field of the local area, and use the three-dimensional strain field to analyze the deformation behavior of the test piece under external force or thermal load; According to the three-dimensional strain field, the strain data is mapped to the three-dimensional model to generate an intuitive strain distribution diagram.
[0038] The strain data is mapped using colors to indicate the magnitude and direction of the strain.
[0039] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A blue light high temperature 3D-DIC video extensometer based on polarization imaging, It is characterized in that include: High temperature test machine, two polarization cameras, narrow bandpass filters, blue polarized LED light source and computer; including: The high temperature testing machine comprises a high temperature furnace arranged in a shell, a support platform and a load applying device are arranged inside the furnace body of the high temperature furnace, the test piece is placed between the support platform and the load applying device, the load applying device comprises a pressure rod passing through the top of the furnace body and then connected to a counterweight block; a quartz glass window is opened on the shell; two polarization cameras, capturing images of the test piece from different viewing angles, with polarizers installed in front of the zoom polarization lenses of the polarization cameras; Narrow bandpass filters, placed before each polarizer; The blue polarized LED light source is arranged in front of the quartz glass window, and is used to emit blue light and illuminate the test piece through the quartz glass window; The computer receives two sets of images from two polarization cameras, matches the two sets of images, generates a disparity map, converts the disparity map into three-dimensional point cloud data, reconstructs a three-dimensional model of the test piece, and calculates the three-dimensional displacement field and three-dimensional strain field of the test piece based on the three-dimensional model.
2. A blue light high temperature 3D-DIC video extensometer based on polarization imaging according to claim 1, It is characterized in that The computer uses a block matching algorithm in a stereo matching algorithm to find the best match of the same window in an image taken by another polarization camera at a corresponding moment in a window around a certain pixel point in an image taken by one of the polarization cameras, and after completing the stereo matching, calculates the disparity value of each pixel point to form a disparity map; The disparity map is converted into three-dimensional point cloud data through triangulation method to reconstruct the three-dimensional model of the object, including: using the disparity value in the disparity map as well as the focal length and the baseline length between the two polarization cameras to calculate the depth of each point in the captured scene, generate a depth map, create and operate three-dimensional point cloud data, calculate the three-dimensional coordinates using the depth map, and reconstruct the three-dimensional model.
3. A blue light high temperature 3D-DIC video extensometer based on polarization imaging according to claim 2, It is characterized in that Create and manipulate 3D point cloud data, including: performing point cloud denoising and filtering on 3D point cloud data; To align the point clouds, the iterative closest point algorithm is used to iteratively adjust the transformation matrix between the two point clouds to minimize the distance between the point clouds and achieve alignment; A smooth three-dimensional surface is generated by point cloud smoothing and interpolation, and the surface reconstruction method Poisso is used in combination with the depth map to generate a complete three-dimensional model.
4. The blue light high temperature 3D-DIC video extensometer based on polarization imaging according to claim 3, It is characterized in that Calculation of the three-dimensional displacement field and three-dimensional strain field of the test piece includes: A plurality of image correlation blocks are defined on the three-dimensional point cloud data, wherein the image correlation blocks are composed of a plurality of pixels and represent a local area on the three-dimensional surface, and the local area is used as a sub-area, and each sub-area is used as a basic unit of DIC analysis to track the displacement and deformation of the sub-area at different time points; In the three-dimensional point cloud data of different time frames, the matching position of each related block is calculated by the 3D-DIC algorithm, and the displacement is calculated with sub-pixel accuracy in combination with the cubic spline interpolation algorithm to obtain the displacement data; Calculate the displacement vector of each image-related block based on the displacement data; According to the displacement vector, a three-dimensional displacement field of the entire test piece is generated; Perform differential operation on the three-dimensional displacement field to calculate the three-dimensional strain field of the local area, and use the three-dimensional strain field to analyze the deformation behavior of the test piece under external force or thermal load; According to the three-dimensional strain field, the strain data is mapped to the three-dimensional model to generate an intuitive strain distribution diagram.
5. The blue light high temperature 3D-DIC video extensometer based on polarization imaging according to claim 4, It is characterized in that The strain data is mapped using colors to indicate the magnitude and direction of the strain.
6. A measurement method using the blue light high temperature 3D-DIC video extensometer based on polarization imaging as claimed in claim 1, It is characterized in that The method includes: The S1 test machine is set up with a high temperature test environment, which meets the required temperature and load conditions; S2 adjusts the focal length and angle of the two polarization cameras so that the distance between the two polarization cameras is 10-20 times the size of the test piece, adjusts the viewing angle difference of the two polarization cameras to between 15-30 degrees, and aligns the optical axes of the two polarization cameras with the center of the test piece; S3 starts the blue polarized LED light source and installs a narrow bandpass filter in front of the polarizer; S4 calibrates the tilt angles of the two polarization cameras using a calibration plate to remove image distortion caused by tilt; The two polarization cameras of S5 start shooting at the same time, capturing images of the surface of the test piece; S6 receives two sets of images from two polarization cameras, matches the two sets of images, generates a disparity map, converts the disparity map into three-dimensional point cloud data, reconstructs the three-dimensional model of the test piece, and calculates the three-dimensional displacement field and three-dimensional strain field of the test piece.
7. The measuring method of the blue light high temperature 3D-DIC video extensometer based on polarization imaging according to claim 6, It is characterized in that S6 includes: Through the block matching algorithm in the stereo matching algorithm, the best match of the same window in an image taken by another polarization camera at the corresponding moment is found in the window around a certain pixel point in an image taken by one of the polarization cameras. After completing the stereo matching, the disparity value of each pixel point is calculated to form a disparity map. The disparity map is converted into three-dimensional point cloud data through triangulation method to reconstruct the three-dimensional model of the object, including: using the disparity value in the disparity map as well as the focal length and the baseline length between the two polarization cameras to calculate the depth of each point in the captured scene, generate a depth map, create and operate three-dimensional point cloud data, calculate the three-dimensional coordinates using the depth map, and reconstruct the three-dimensional model.
8. The measuring method according to claim 7, It is characterized in that Create and manipulate 3D point cloud data, including: performing point cloud denoising and filtering on 3D point cloud data; To align the point clouds, the iterative closest point algorithm is used to iteratively adjust the transformation matrix between the two point clouds to minimize the distance between the point clouds and achieve alignment; A smooth three-dimensional surface is generated by point cloud smoothing and interpolation, and the surface reconstruction method Poisso is used in combination with the depth map to generate a complete three-dimensional model.
9. The measuring method according to claim 8, It is characterized in that Calculation of the three-dimensional displacement field and three-dimensional strain field of the test piece includes: A plurality of image correlation blocks are defined on the three-dimensional point cloud data, wherein the image correlation blocks are composed of a plurality of pixels and represent a local area on the three-dimensional surface, and the local area is used as a sub-area, and each sub-area is used as a basic unit of DIC analysis to track the displacement and deformation of the sub-area at different time points; In the three-dimensional point cloud data of different time frames, the matching position of each relevant block is calculated by the 3D-DIC algorithm, and the displacement is calculated with sub-pixel accuracy in combination with the cubic spline interpolation algorithm to obtain the displacement data; Calculate the displacement vector of each image-related block based on the displacement data; According to the displacement vector, a three-dimensional displacement field of the entire test piece is generated; Perform differential operation on the three-dimensional displacement field to calculate the three-dimensional strain field of the local area, and use the three-dimensional strain field to analyze the deformation behavior of the test piece under external force or thermal load; According to the three-dimensional strain field, the strain data is mapped to the three-dimensional model to generate an intuitive strain distribution diagram.
10. The measuring method according to claim 9, It is characterized in that The strain data is mapped using colors to indicate the magnitude and direction of the strain.
Citation Information
Patent Citations
Apparatus and method for liquid expanding deformation test of geomembrane based on three-dimensional DIC
CN108956310A
Synchronous measurement system and measurement methodof temperature field and displacement field based on 3D-DIC (three-dimensional digital image correlation)
CN110057399A
Polarization imaging-based three-dimensional measurement method of ice shape
CN110645921A
Crack segmentation method and system based on strain prior constraint
CN116416269A
Multi-class variable input nonlinear finite element model updating hybrid test method based on video recognition
CN116484681A
