Metal internal characteristic selection area extraction method

Abstract

The invention discloses a metal internal characteristic selective extraction method, which comprises the following steps of: (1) scanning a block-shaped metal sample by adopting high-resolution X-ray CT (Computed Tomography), reconstructing characteristics including defect distribution, microstructure, holes and cracks in metal through a three-dimensional reconstruction algorithm, and generating a high-precision three-dimensional model; (2) selecting a required area, calculating voxel resolution according to the distance SDD between a radiation source and a detector and the distance SOD between a sample and the radiation source, displaying a reconstructed three-dimensional image in three-dimensional visual analysis software, and performing threshold segmentation through a computer vision algorithm, and automatically identifying and extracting required features for observation; the whole sample is placed in a square frame in an externally tangent form, and a Cartesian rectangular coordinate system is established. According to the invention, after the internal structure is observed by using the X-ray microscopic imaging technology, an interested area is selected for extraction by using linear cutting according to the three-dimensional structure information, and the internal defect of the sample can be efficiently and accurately positioned and extracted.

Description

Technical Field

[0001] The present invention relates to the field of material analysis and preparation, and specifically to a method for extracting internal metal features, which is suitable for accurately identifying and extracting structural features such as internal defects and microstructures of metal materials. Background Art

[0002] X-ray industrial CT is one of the most promising nondestructive testing methods. Leveraging the penetrating power of X-rays and computer image reconstruction algorithms, this technology enables three-dimensional visualization of the interior of the object being tested. Furthermore, X-ray penetration is sensitive to structure and material composition, enabling grayscale threshold segmentation of the reconstructed image to provide detailed measurements of material structure dimensions, internal elements, secondary phases, and defect distribution, providing significant value for material research and development, quality control, and performance optimization. Consequently, X-ray industrial CT is recognized as one of the most effective three-dimensional structural analysis methods in the field of metal materials.

[0003] In metal material research, in addition to macrostructure, microstructure, such as secondary phase distribution, texture variations, and compositional segregation zones, also significantly influences service performance. However, because the internal structure of metals is difficult to characterize, the relationship between structural changes and microstructure is also challenging to study. This requires extracting the region of interest after deformation and re-sampling.

[0004] Currently, existing technologies often require various processing methods to extract areas of interest inside materials. Existing methods, such as random cutting and mechanical processing extraction, cannot accurately extract the target area, resulting in low efficiency, poor precision, and serious material waste. How to use appropriate means to work together while maintaining efficiency and accuracy has always been a problem that technicians need to solve. Therefore, developing a set of fast, reliable, and highly feasible methods for extracting selected areas inside metals is of practical significance for understanding the deformation and fracture behavior of materials. This method will help solve the technical difficulties previously faced and provide more convenience and support for metal research. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for extracting metal internal feature selection areas to improve the extraction efficiency and accuracy of metal internal selection areas and overcome the precision and positioning difficulties in the prior art.

[0006] To achieve the above-mentioned purpose, the present invention adopts a technical solution: a method for extracting metal internal feature selection, comprising the following steps: (1) Use high-resolution X-ray CT to scan bulk metal samples, and reconstruct the internal features of the metal including defect distribution, microstructure, pores, and cracks through a three-dimensional reconstruction algorithm to generate a high-precision three-dimensional model; (2) Select the required area, calculate the voxel resolution based on the distance SDD between the source and the detector and the distance SOD between the sample and the source, display the reconstructed three-dimensional image in the three-dimensional visualization analysis software, and automatically identify and extract the required features for observation through the threshold segmentation of the computer vision algorithm; place the entire sample in a square frame in an circumscribed form, and establish a Cartesian coordinate system to accurately record the three-dimensional vertex coordinates of the selected area;

[0007] (3) Based on the calculated coordinate position and the internal structural characteristics of the sample, high-precision wire cutting is used for rough extraction; (4) Further extract the defects of the sample to obtain the final required sample.

[0008] In the preferred technical solution, the CT scanning sample detection in step (1) has a maximum field of view of up to 30 mm × 60 mm and a resolution of 20 μm.

[0009] In the preferred technical solution, the calculation formula for voxel resolution in step (2) is voxel size = detector size / SDD / SOD, where SOD represents the distance from the ray source to the object, and SDD represents the distance from the ray source to the detector. The resolution can increase as the field of view decreases.

[0010] In the preferred technical solution, the dimensions extracted by high-precision wire cutting in step (3) reserve a surface thickness consumed by grinding and polishing.

[0011] In the preferred technical solution, the precision of the wire cutting in step (3) is within 0.02 mm, meeting the requirements of high-precision sample preparation.

[0012] In the preferred technical solution, the surface grinding and polishing in step (4) includes the following steps: using 400-grit sandpaper to grind the cut sample to remove the oxide film, and then using 1200-grit, 2000-grit, and 3000-grit sandpaper to grind in sequence by rotating 90° so that the next layer of sandpaper can cover the scratches of the previous layer of sandpaper, thereby obtaining a bright sample surface with fine and uniform scratches distributed in a single direction; at the same time, cooling water is continuously passed through the grinding and polishing process for cooling.

[0013] In a preferred technical solution, the surface treatment in step (4) adopts one or more of electrolytic polishing, vibration polishing, and argon ion polishing.

[0014] In the preferred technical solution, the further extraction of defects from the sample in step (4) includes: if mechanical property testing is required, continuing cutting; if tissue characterization is required, performing surface grinding and polishing and surface treatment on the cut sample.

[0015] In a preferred technical solution, in step (2), the three-dimensional visualization analysis software is Avizo software.

[0016] In the technical solution of this application, the characteristic structure inside the sample needs to have a significant X-ray absorption contrast difference with the matrix. The sample thickness will affect the X-ray transmittance and thus the imaging quality of the material. The appropriate maximum sample test thickness needs to be selected according to different material densities.

[0017] In the technical solution of the present application, for samples with extremely high precision requirements, the cut samples can be scanned at the maximum resolution, polished, and cut using a focused ion beam, so that characteristic structures of up to nanometer size can be extracted.

[0018] Due to the application of the above technical solution, the present invention has the following advantages compared with the prior art: 1. This invention combines X-ray microscopy with high-precision cutting technology to achieve rapid and accurate extraction of regions of interest within metal samples, while ensuring continuity between observation and sample preparation, significantly improving sample preparation efficiency and quality.

[0019] 2. The present invention is not only suitable for the extraction of tissue around internal cracks, but also for other scenarios such as the extraction of porosity defects in 3D printed samples, the extraction of holes in powder metallurgy samples, and the extraction of component segregation areas inside cast metal samples.

[0020] 3. It provides a traceable and highly repeatable sample preparation process, providing high-quality sample support for further research on the internal structure evolution, service behavior and damage mechanism of materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is the flow chart of metal internal selection extraction; Figure 2 It is a schematic diagram of CT test; Figure 3 It is a schematic diagram of the coordinate calibration of the selected area; Figure 4 It is a schematic diagram of wire cutting; Figure 5 It is a comparison of the surface morphology of the sample after grinding and polishing with the pre-observation morphology. DETAILED DESCRIPTION

[0022] The present invention will be further described below with reference to the accompanying drawings and embodiments: Example 1

[0023] See also Figure 1 As shown, a method for extracting metal internal feature selection includes the following steps: (1) Use high-resolution X-ray CT to scan bulk metal samples, and reconstruct the internal features of the metal including defect distribution, microstructure, pores, and cracks through a three-dimensional reconstruction algorithm to generate a high-precision three-dimensional model; (2) The reconstructed three-dimensional image is displayed in Avizo software. The threshold segmentation is performed by computer vision algorithm to automatically identify and extract the required features for observation. The required area is selected and the voxel resolution is calculated based on the distance SDD between the source and the detector and the distance SOD between the sample and the source. The sample is placed in a square frame in an external form and a Cartesian coordinate system is established to accurately record the three-dimensional vertex coordinates of the selected area, such as Figure 3 As shown; (3) Based on the calculated coordinate position and the internal structural characteristics of the sample, high-precision wire cutting is used for rough extraction; (4) Further extract the defects of the sample to obtain the final required sample.

[0024] In this embodiment, the size of the sample extracted by high-precision wire cutting in step (3) has a surface thickness reserved for grinding and polishing.

[0025] In this embodiment, the surface polishing in step (4) includes the following steps: using 400-grit sandpaper to polish the cut sample to remove the oxide film, and then using 1200-grit, 2000-grit, and 3000-grit sandpaper to polish in sequence by rotating 90° so that the next layer of sandpaper can cover the scratches of the previous layer of sandpaper, thereby obtaining a bright sample surface with fine and uniform scratches distributed in a single direction; at the same time, cooling water is continuously passed through the polishing process for cooling.

[0026] In this embodiment, the surface treatment in step (4) adopts one or more of electrolytic polishing, vibration polishing, and argon ion polishing.

[0027] In this embodiment, the CT scanning sample detection in step (1) has a maximum field of view of up to 30 mm × 60 mm and a resolution of 20 μm.

[0028] In this embodiment, the calculation formula of the voxel resolution in step (2) is voxel size = detector size / SDD / SOD, where SOD represents the distance from the ray source to the object, and SDD represents the distance from the ray source to the detector.

[0029] In this embodiment, the precision of the wire cutting in step (3) is within 0.02 mm.

[0030] In this embodiment, if mechanical property testing is required, cutting is continued; if tissue characterization is required, surface grinding and polishing and surface treatment are performed on the cut sample to obtain the final required sample.

[0031] After obtaining the final required sample, proceed to the next step of testing. Example 2

[0032] This embodiment provides a sample preparation method for extracting a certain pore distribution characteristic surface inside a 3D-printed Al-Mg alloy based on a method for extracting selected areas of metal internal features, which includes obtaining the three-dimensional structure of the alloy sample, mechanical polishing, cleaning, and sample preservation.

[0033] a. Acquisition of three-dimensional structure of Al-Mg alloy sample This embodiment uses the X-Lab domestic CT tester. First, select a suitable sample preparation device according to the sample, fix the sample on the sample preparation device, and make the central axis of the sample and the sample preparation device on the same straight line. Figure 2 As shown. Select the required voltage, current, and exposure time based on sample density and thickness. Note that optimal sample imaging quality is achieved when the X-ray transmittance is 40%. For Al-Mg alloy, the optimal parameters are 80 kV and 50 µA, respectively. After loading the sample, perform a rough alignment at a low magnification. With the stage angle at 0°, adjust the Z and X axes for vertical and horizontal alignment. With the stage angle at 90°, adjust the X90 and Z axes for horizontal and vertical alignment until the sample's area of interest appears in the center of the crosshairs on the detector screen. After rough alignment, move the X-ray source closer to the sample and the detector further away to increase magnification. Recentering the sample after magnification is still necessary, requiring only minor adjustments. As with coarse alignment, when the stage angle is 0°, adjust the Z and X axes for vertical and horizontal alignment. When the stage angle is 90°, adjust the X90 and Z axes for horizontal and vertical alignment until the sample's region of interest appears at the center of the crosshairs on the detector display. Rotate 360° to ensure the sample's region of interest remains within the detector display during testing. After alignment, perform data acquisition to obtain the three-dimensional feature distribution of the Al-Mg alloy sample.

[0034] b. Region selection The acquired two-dimensional slice data were imported into the commercial software Avizo for three-dimensional structure reconstruction and observation of the internal pore distribution.

[0035] c. Sample cutting The reconstructed three-dimensional image is displayed in Avizo software, and the required features are automatically identified and extracted for observation through threshold segmentation by computer vision algorithm. The required area is selected, and the voxel resolution at this time is calculated by combining the SDD and SOD values during the test with the equipment parameters. The Cartesian coordinate system of the sample is established with the vertices of the sample's circumscribed cube as the origin coordinates, and the three-dimensional vertex coordinates of the selected area are recorded at this time. The quadrilateral thin plate is clamped on the wire cutting machine, and the preset origin coordinates are set as the origin in the wire cutting system. Based on the calculated coordinate position, combined with the internal structural characteristics of the sample, such as Figure 4 As shown, the line cutting is moved to the coordinates of the selection cube using a digital control system to perform rough extraction.

[0036] d. Sample grinding and polishing If tissue characterization is required, the cut samples are ground and polished. First, the surface oxide scale is polished with 400-mesh sandpaper, and then the surface is ground and polished with 1000-mesh, 1500-mesh, 2000-mesh, and 3000-mesh sandpaper respectively. The grinding and polishing time is matched with the grinding and polishing force.

[0037] e. Compare the slice results in CT with the surface thrown out by grinding to confirm that it is the desired observation surface. The sample preparation is completed and the final required sample is obtained. Figure 5 The following is a comparison of a cross-section of the Al-Mg alloy interior and the surface after grinding and polishing. Figures 5-(a) and (b) show the cross-section of a specific surface within the sample and the scanning electron microscopic observation after grinding and polishing, demonstrating their consistency. Figures a1-a4 and b1-b4 are magnified comparisons of the two results, fully illustrating this finding.

[0038] This example uses 3D-printed Al-Mg alloy as an example. The printed samples often contain a large number of pores for a variety of reasons, such as the reaction of aluminum with hydrogen protons to generate hydrogen, the molten metal entrapping ambient gases and solidifying, and the powder itself containing pores. Therefore, when the 3D-printed block is cut and tensile tested, the pore distribution greatly affects the material properties. This method enables pre-structural characterization of the material, and then extraction and testing of material properties in selected areas with different pore distribution characteristics.

[0039] Example 3 This embodiment provides a method for extracting regions near microcracks after deformation based on a method for extracting regions with internal metal features.

[0040] In order to study the correlation between the initiation and propagation of internal cracks in steel and the microstructure, the microstructure at the initiation crack is extracted, and the deformation mechanism activated by the surrounding grains during crack propagation is studied in depth.

[0041] 1. First, use high-resolution X-ray CT to scan the deformed steel material to obtain a three-dimensional structure with higher resolution and more prominent details, from which the required cracks can be found.

[0042] 2. Import the acquired 2D slice data into the commercial software Avizo for 3D structural reconstruction. Use the 3D reconstruction algorithm to reconstruct the characteristics of the crack inside the metal. Select the crack at this time and select the 3D vertex coordinates of the selected area.

[0043] 3. Calculate the voxel resolution at this point using the SDD and SOD values during the test, combined with the equipment parameters. Establish a Cartesian coordinate system with the vertices of the sample's circumscribed cube as the origin coordinates, and record the vertex coordinates of the selected cube at this point. Clamp the material and the quadrilateral sheet onto the wire cutting machine, and set the preset origin coordinates as the origin in the wire cutting system. Use the digital control system to move the wire cutting machine to the coordinates of the selected cube for cutting.

[0044] 4. At this time, a crack with a width of 10 microns is selected for detailed discussion. However, conventional wire cutting and polishing processes will introduce operational errors during manual operation, making it difficult to achieve the accuracy required by this embodiment and very likely to damage the sample.

[0045] Therefore, in this step, the sample is finely polished to a certain thickness, and polishing is stopped when the observation surface reaches 100 microns. Focused ion beam cutting technology is used to cut the sample layer by layer in more detail to accurately extract the microstructure around the crack at this time. The final sample is obtained, and the preparation is completed.

[0046] It is well known that material fracture is caused by crack initiation and propagation. However, crack initiation is often not limited to surface cracks; internal crack initiation is also a decisive factor in material fracture. Understanding the mechanisms of internal crack initiation and propagation during material deformation is crucial for understanding material fracture behavior. In this example, by extracting deformation information around internal cracks in the sample, we investigated the correlation between crack initiation and propagation and the grain orientation or the activated deformation mechanism.

[0047] Example 4 This embodiment provides a method for extracting different second phase densities in metals based on a method for extracting selected regions of metal internal features.

[0048] 1. Use high-resolution X-ray CT to scan bulk metal samples, obtain three-dimensional structural information through three-dimensional reconstruction algorithms, and generate high-precision three-dimensional models; 2. Automatically identify the target area and select the required area through computer vision algorithms, calculate the voxel resolution corresponding to the test phase, establish the Cartesian coordinate system of the sample, select different second phase density distribution areas for selection, and accurately record the three-dimensional vertex coordinates of the selected areas; 3. Based on the calculated coordinate position and the internal structural characteristics of the sample, high-precision wire cutting is used to extract the material selection area; 4. After extraction, multiple steps of precision cutting are performed to obtain the final required sample.

[0049] 5. Conduct mechanical property tests on materials with different second phase distributions.

[0050] In this example, the second phase, as the most prevalent strengthening mechanism in alloys, can pin dislocations and increase material strength. However, the distribution of the second phase in the alloy may not be uniform, which can lead to human error when testing the performance of material samples and even mislead research conclusions. Preliminary observation of the second phase distribution in samples is crucial for subsequent material performance testing.

Claims

1. A method for extracting metal internal feature selection, characterized in that: The steps include: (1) Use high-resolution X-ray CT to scan bulk metal samples, and reconstruct the internal features of the metal including defect distribution, microstructure, pores, and cracks through a three-dimensional reconstruction algorithm to generate a high-precision three-dimensional model; (2) The reconstructed three-dimensional image is displayed in a three-dimensional visualization analysis software. The threshold segmentation is performed by a computer vision algorithm to automatically identify and extract the required features for observation. The required area is selected and the voxel resolution is calculated based on the distance SDD between the source and the detector and the distance SOD between the sample and the source. The entire sample is placed in a square frame in an circumscribed form and a Cartesian coordinate system is established to accurately record the three-dimensional vertex coordinates of the selected area. (3) Based on the calculated coordinate position and the internal structural characteristics of the sample, high-precision wire cutting is used for rough extraction; (4) Further extract the defects of the sample to obtain the final required sample.

2. The method for extracting metal internal feature selection according to claim 1, characterized in that: In the step (1), the CT scanning sample is used for detection with a maximum field of view of up to 30 mm × 60 mm and a resolution of 20 μm.

3. The method for extracting metal internal feature selection according to claim 1, characterized in that: The calculation formula of the voxel resolution in step (2) is voxel size = detector size / SDD / SOD, where SOD represents the distance from the ray source to the object, and SDD represents the distance from the ray source to the detector. The resolution can increase as the field of view decreases.

4. The method for extracting metal internal feature selection according to claim 1, characterized in that: The size of the sample extracted by high-precision wire cutting in step (3) has a surface thickness reserved for grinding and polishing.

5. The method for extracting metal internal feature selection according to claim 1, characterized in that: The precision of the wire cutting in step (3) is within 0.02 mm, which meets the requirements of high-precision sample preparation.

6. The method for extracting metal internal feature selection according to claim 1, characterized in that: The surface polishing in step (4) includes the following steps: using 400-grit sandpaper to polish the cut sample to remove the oxide film, and then using 1200-grit, 2000-grit, and 3000-grit sandpaper to rotate 90 degrees in sequence for polishing, so that the next layer of sandpaper can cover the scratches of the previous layer of sandpaper, thereby obtaining a bright sample surface with fine and uniform scratches distributed in a single direction; at the same time, cooling water is continuously passed through the polishing process for cooling.

7. The method for extracting metal internal feature selection according to claim 1, characterized in that: The surface treatment in step (4) adopts one or more of electrolytic polishing, vibration polishing, and argon ion polishing.

8. The method for extracting metal internal feature selection according to claim 1, characterized in that: The further extraction of defects from the sample in step (4) includes: if mechanical property testing is required, continuing cutting; if tissue characterization is required, performing surface grinding and polishing and surface treatment on the cut sample.

9. The method for extracting metal internal feature selection according to claim 1, characterized in that: In step (2), the three-dimensional visualization analysis software is Avizo software.

Images