Calibration method of electron back scattering diffraction pattern and medium

Abstract

The invention provides an electron back scattering diffraction pattern calibration method and a medium. The method comprises the following steps: creating a series of Kikuchi strip normal triangles and a series of reflecting surface normal triangles; a target reflecting surface normal triangle matched with the Kikuchi belt normal triangle is screened out; performing matching counting according to the normal triangle of the target reflecting surface and the normal triangle of the Kikuchi zone, endowing information of a target crystal plane family determined according to a matching counting result to the corresponding Kikuchi zone so as to determine the crystal plane family to which the Kikuchi zone belongs, and obtaining a target Kikuchi zone; calculating to obtain an orientation estimation value corresponding to each Kikuchi band combination; screening out an optimal Kikuchi zone combination from all Kikuchi zone combinations according to all orientation estimation values, and calculating a target orientation value according to the optimal Kikuchi zone combination; and calculating calibration precision according to the target orientation value, and if the calibration precision is lower than a preset precision threshold, determining that the electron back scattering diffraction pattern is successfully calibrated. According to the invention, the calibration precision can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of material microstructure characterization and crystal orientation calculation, and particularly relates to a method and medium for calibrating electron backscatter diffraction patterns. Background Art

[0002] An EBSD (Electron Back-scattered Diffraction) device is a complete set of microscopic analysis equipment, which can give statistical crystallographic information of materials in a large range, such as crystal structure, composition, texture, etc.

[0003] EBSD has advantages such as a large analysis area and strong statistics. Its emergence provides an important and reliable scientific basis for the improvement of material properties. A core function of EBSD is to determine crystal orientation, calculate orientation difference, analyze orientation relationship, etc.

[0004] However, the accuracy of determining crystal orientation, calculating orientation difference, and analyzing orientation relationship depends on the reliability of the selected electron backscatter diffraction pattern. How to calibrate the electron backscatter diffraction pattern is a problem to be solved. Summary of the Invention

[0005] Embodiments of the present application provide a method and medium for calibrating electron backscatter diffraction patterns, which improve the calibration accuracy of Kikuchi patterns of crystals to be measured.

[0006] In a first aspect, embodiments of the present application provide a method for calibrating electron backscatter diffraction patterns, including:

[0007] Creating a series of Kikuchi band normal triangles according to the Kikuchi bands of the electron backscatter diffraction pattern of the sample to be detected, where the Kikuchi band normal triangle is composed of three non-collinear actual normal vectors corresponding to the electron backscatter diffraction pattern;

[0008] Creating a series of reflecting plane normal triangles according to a pre-constructed theoretical crystal, where the reflecting plane normal triangle is composed of three non-collinear theoretical normal vectors corresponding to the theoretical crystal;

[0009] Selecting a target reflecting plane normal triangle that matches the Kikuchi band normal triangle;

[0010] Performing matching counting according to the target reflecting plane normal triangle and the Kikuchi band normal triangle, and assigning the information of the target crystal plane family determined according to the matching counting result to the corresponding Kikuchi band, so as to determine the crystal plane family to which the Kikuchi band belongs to obtain a target Kikuchi band;

[0011] Calculating an orientation estimation value corresponding to each Kikuchi band combination, where the Kikuchi band combination includes two different target Kikuchi bands;

[0012] Screen out the optimal Kikuchi band combination from all the Kikuchi band combinations according to all the orientation estimation values, and calculate the target orientation value according to the optimal Kikuchi band combination; the optimal Kikuchi band combination is the Kikuchi band combination that maximizes the number of successfully calibrated target Kikuchi bands.

[0013] Calculate the calibration accuracy according to the target orientation value. If the calibration accuracy is lower than the preset accuracy threshold, determine that the electron backscatter diffraction pattern calibration is successful.

[0014] In a second aspect, an embodiment of the present application further provides a computer storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the calibration method of the electron backscatter diffraction pattern as described in the first aspect.

[0015] A calibration method and medium for an electron backscatter diffraction pattern provided by an embodiment of the present application. The method process is to create a series of Kikuchi band normal triangles and a series of reflection plane normal triangles; screen out the target reflection plane normal triangles that match the Kikuchi band normal triangles; perform matching counting according to the target reflection plane normal triangles and the Kikuchi band normal triangles, and assign the information of the target crystal plane family determined according to the matching counting result to the corresponding Kikuchi band to determine the crystal plane family to which the Kikuchi band belongs to obtain the target Kikuchi band; construct a Kikuchi band combination composed of every two target Kikuchi bands from all the target Kikuchi bands, and calculate the orientation estimation value corresponding to each Kikuchi band combination; screen out the optimal Kikuchi band combination from all the Kikuchi band combinations according to all the orientation estimation values, and calculate the target orientation value according to the optimal Kikuchi band combination; calculate the calibration accuracy according to the target orientation value. If the calibration accuracy is lower than the preset accuracy threshold, determine that the electron backscatter diffraction pattern calibration is successful. The present application can improve the calibration accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those skilled in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0017] Figure 1 Shows a schematic flow chart of some exemplary calibration methods for electron backscatter diffraction patterns;

[0018] Figure 2 Shows a schematic diagram of the construction of some exemplary experimental normal vectors;

[0019] Figure 3 Shows a schematic diagram of the creation of some exemplary Kikuchi pole normal triangles;

[0020] Figure 4 Shows schematic diagrams of the creation of some exemplary reflection plane normal triangles;

[0021] Figure 5 Shows schematic diagrams of the matching process of some exemplary reflection plane normal triangles and Kikuchi pole normal triangles;

[0022] Figure 6 Shows schematic diagrams of the calibration demonstration of some exemplary cubic - YSZ Kikuchi patterns;

[0023] Figure 7 Shows schematic diagrams of the comparison between some exemplary Oxford orientation maps (a) and the orientation maps calibrated in this application (b). Detailed implementation manners

[0024] Next, in conjunction with the accompanying drawings, the technical solutions in some embodiments of this application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. Based on the embodiments provided in this application, all other embodiments obtained by those of ordinary skill in the art belong to the scope protected by this application.

[0025] In the description of this application, it should be understood that the orientation or positional relationships indicated by the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to this application.

[0026] Unless otherwise required by the context, throughout the specification and claims, the term "comprising" is interpreted in an open, inclusive sense, that is, "including, but not limited to". In the description of the specification, the terms "one embodiment", "some embodiments", "exemplary embodiments", "exemplarily" or "some examples", etc. are intended to indicate that the specific features, structures, materials or characteristics related to the embodiment or example are included in at least one embodiment or example of this application. The schematic representations of the above terms are not necessarily referring to the same embodiment or example. In addition, the specific features, structures, materials or characteristics described above can be included in any one or more embodiments or examples in any appropriate manner.

[0027] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present application, unless otherwise specified, the meaning of "a plurality" is two or more.

[0028] When describing some embodiments, the expressions "coupled" and "connected" and their derivatives may be used. For example, when describing some embodiments, the term "connected" may be used to indicate that two or more components have direct physical or electrical contact with each other. Also, for example, when describing some embodiments, the term "coupled" may be used to indicate that two or more components have direct physical or electrical contact. However, the term "coupled" may also mean that two or more components do not have direct contact with each other but still cooperate or interact with each other. The embodiments disclosed herein are not necessarily limited to the content herein.

[0029] "At least one of A, B, and C" has the same meaning as "at least one of A, B, or C" and includes the following combinations of A, B, and C: only A, only B, only C, the combination of A and B, the combination of A and C, the combination of B and C, and the combination of A, B, and C.

[0030] "A and / or B" includes the following three combinations: only A, only B, and the combination of A and B.

[0031] In addition, the use of "based on" means open and inclusive because a process, step, calculation, or other action "based on" one or more of the stated conditions or values may in practice be based on additional conditions or values beyond the stated ones.

[0032] As used in the present application, "about", "substantially", or "approximately" includes the stated value and the average value within an acceptable deviation range of the specific value, where the acceptable deviation range is determined by a person of ordinary skill in the art considering the measurement being discussed and the error associated with the measurement of the specific quantity (i.e., the limitations of the measurement system).

[0033] Electron backscatter diffraction patterns include Kikuchi bands (Kikuchi lines) and Kikuchi poles. Kikuchi lines are bright and dark paired parallel lines that appear in the electron diffraction pattern, reflecting the orientation and deviation of the crystal. In the case of symmetric incidence, that is, when the diffracting crystal plane is parallel to the incident beam direction, Kikuchi bands often appear as dark bands or bright bands, which are called Kikuchi bands. Among them, different Kikuchi bands correspond to crystal planes with different indices, and the intersection points of Kikuchi bands, namely Kikuchi poles, correspond to the zone axes of the crystal. By analyzing the position and direction of Kikuchi bands, the normal vector of the corresponding diffracting crystal plane can be inferred, thereby obtaining the orientation information of the crystal. The direction of the normal vector of the diffracting crystal plane can be determined by the crystal plane index, which can help determine the direction of the crystal plane in three-dimensional space. Crystal plane index is a set of symbols used in crystallography to represent the orientation of crystal planes, usually represented by Miller indices. It is marked by the reciprocal of the intercepts of the crystal plane on the three crystal axes and the relatively prime integer ratio, reflecting the orientation of the crystal plane normal in the lattice, usually represented as (hkl).

[0034] EBSD equipment usually needs to be installed in a scanning electron microscope, including an EBSD detector, an image processing system, and a computer system. The detector consists of a phosphor screen (such as a phosphorous screen) on the outer surface and a high-sensitivity CCD (Charge Coupled Device) camera behind the screen. The phosphor screen is mainly used to detect Kikuchi lines, while the CCD camera is mainly used to record Kikuchi lines. The computer system is mainly used to analyze the collected Kikuchi lines to obtain various information such as crystal orientation, orientation difference, orientation relationship, and crystal structure. The EBSD equipment is installed in the scanning electron microscope and is excited by the electron beam of the scanning electron microscope. When the electron beam interacts with the sample to be inspected (tilted at a certain angle, such as 60-70°), some backscattered electrons with relatively small energy loss will undergo Bragg diffraction with the crystal planes on the surface of the sample to be inspected, thereby forming two conical diffraction beams, which are approximately parallel to the crystal planes where Bragg diffraction occurs. The two conical diffraction beams intersect with the phosphor screen placed in front of the sample to be inspected to form Kikuchi bands. Different Kikuchi bands correspond to crystal planes with different indices, and the intersection points of Kikuchi bands, namely Kikuchi poles, correspond to the zone axes of the crystal. Diffraction information, background noise, and other signals together constitute an electron backscatter diffraction pattern. The screen of the EBSD equipment is a phosphor screen, which is used to receive the backscattered electron signal generated by the sample under the irradiation of the electron beam and convert it into a visible light image including the electron backscatter diffraction pattern. The crystallographic information given by EBSD comes from the measurement and analysis of the electron backscatter diffraction pattern. The diffraction pattern is photographed and stored by the CCD camera and then viewed on the computer.

[0035] The core of EBSD is the acquisition of electron backscatter diffraction patterns and the measurement and calibration of electron backscatter diffraction patterns. Pattern calibration refers to determining the crystal plane indices of Kikuchi bands, that is, calibrating which crystal plane each detected Kikuchi band corresponds to, and then calculating the orientation. The crystal orientation is defined as the relative orientation of the three crystal axes of the crystal (such as crystal axes

[100] ,

[010] ,

[001] ) in the sample coordinate system (for example, RD rolling direction, TD transverse direction, ND normal direction). Regardless of how the crystal orientation changes, the crystal plane angle remains unchanged. Measuring the crystal plane angle corresponding to the Kikuchi band and matching it with the theoretical crystal plane angle in the database is a conventional EBSD pattern calibration method. Pattern calibration is carried out by the so-called "voting method". This method measures the Kikuchi band angle and compares it with the crystal standard database within a certain angular tolerance range, so as to calibrate the crystal plane indices corresponding to the Kikuchi band and further determine other crystallographic information. The voting method is fast and simple. However, the same crystal plane angle may correspond to different crystal plane indices, which easily leads to non-unique calibration results. For example, the calibration effect of EBSD at grain boundaries is worse, which is usually due to pattern overlap at grain boundaries. When the problem of pattern overlap exists, Kikuchi bands belonging to different grains can be detected from a single pattern. These bands are mixed together, which causes the calibration algorithm to be unable to find a calibration solution that makes all bands self-consistent. Therefore, how to improve the accuracy of pattern calibration is a technical problem that needs to be solved urgently.

[0036] This application proposes a calibration method for electron backscatter diffraction patterns. Figure 1 The flow diagrams of some exemplary calibration methods for electron backscatter diffraction patterns are shown. As Figure 1 shown in the flow, it includes the following steps:

[0037] S100. Create a series of Kikuchi band normal triangles KNT according to the Kikuchi bands of the electron backscatter diffraction pattern of the sample to be detected. The Kikuchi band normal triangle KNT is composed of three non-collinear actual normal vectors corresponding to the electron backscatter diffraction pattern.

[0038] S200. Create a series of reflection plane normal triangles RNT according to the pre-constructed theoretical crystal. The reflection plane normal triangle RNT is composed of three non-collinear theoretical normal vectors corresponding to the theoretical crystal.

[0039] S300. Screen out the target reflection plane normal triangle that matches the Kikuchi band normal triangle KNT.

[0040] S400. Perform matching counting according to the target reflection plane normal triangle and the Kikuchi band normal triangle KNT, and assign the information of the target crystal plane family determined according to the matching counting result to the corresponding Kikuchi band, so as to determine the crystal plane family to which the Kikuchi band belongs to obtain the target Kikuchi band.

[0041] S500. Calculate the orientation estimation value E corresponding to each Kikuchi band combination, where each Kikuchi band combination includes two different target Kikuchi bands.

[0042] S600. Screen out the optimal Kikuchi band combination from all the Kikuchi band combinations according to all the orientation estimation values E, and calculate the target orientation value M according to the optimal Kikuchi band combination; the optimal Kikuchi band combination is the Kikuchi band combination that enables the largest number of successfully calibrated target Kikuchi bands.

[0043] S700. Calculate the calibration accuracy MAD according to the target orientation value M. If the calibration accuracy MAD is lower than the preset accuracy threshold, it is determined that the electron backscatter diffraction pattern calibration is successful.

[0044] In some examples, the sample to be detected includes a sample prepared from any one or more of materials such as ceramics, metals, semiconductors, minerals, single crystals, and polycrystals. The diffraction pattern includes a number of Kikuchi bands. The electron backscatter diffraction pattern center (pattern center, PC) refers to the perpendicular line drawn from the electron beam incident point to the diffraction pattern on the fluorescent screen, and the foot position is the pattern center. According to the Kikuchi bands in the electron backscatter diffraction pattern of the sample to be detected, this application first extracts the actual normal vectors, and constructs a series of Kikuchi band normal triangles KNT based on every three non-collinear actual normal vectors. Subsequently, using the pre-constructed theoretical crystal model, extract its corresponding theoretical normal vectors, and construct a series of reflection plane normal triangles RNT based on every three non-collinear theoretical normal vectors. Screen out the target reflection plane normal triangle that matches the Kikuchi band normal triangle KNT through a matching algorithm. Next, based on the matching counting method, compare the target reflection plane normal triangle with the Kikuchi band normal triangle KNT to match each Kikuchi band with the corresponding target crystal plane family. The target crystal plane family is defined as the crystal plane family corresponding to the reflection plane normal triangle RNT with the highest matching frequency with the Kikuchi band. Construct Kikuchi band combinations composed of every two target Kikuchi bands from all the detected target Kikuchi bands. For each pair of different Kikuchi band combinations, calculate the corresponding orientation estimation value E. By analyzing all the orientation estimation values E, screen out the optimal Kikuchi band combination, which is defined as the combination that enables the largest number of successfully calibrated target Kikuchi bands. Based on the optimal Kikuchi band combination, calculate the final target orientation value M. Finally, use the target orientation value M to evaluate the calibration accuracy MAD of the successfully calibrated Kikuchi bands. If the calibration accuracy MAD reaches the preset accuracy threshold, it is determined that the electron backscatter diffraction pattern calibration is successful.

[0045] In this application, by creating a series of Kikuchi band normal triangles (KNT) and a series of reflection plane normal triangles (RNT) and performing matching, the actual normal vector and the theoretical normal vector can be accurately matched, and the crystal plane family corresponding to the Kikuchi band can be determined more accurately, thereby significantly improving the calibration accuracy MAD of the Kikuchi band. Moreover, by creating the Kikuchi band normal triangles (KNT) and the reflection plane normal triangles (RNT) and using the side length and angle information of the triangles for matching, the characteristics of the crystal structure can be considered more comprehensively, rather than relying solely on single crystal plane information, thus improving the adaptability to different crystal structures, being able to adapt to different crystal structures and electron backscatter diffraction patterns of different qualities, and having a good calibration effect for complex crystal structures and low-quality patterns. In addition, this application calculates the target orientation value M based on the matching count and the optimal Kikuchi band combination screening, further improving the accuracy and reliability of the orientation estimation. The combination of obtaining the candidate orientation value based on the rough orientation estimation and calculating the target orientation value M based on the precise orientation can gradually optimize the calibration result and ensure the high precision of the orientation estimation. Evaluating the calibration accuracy MAD corresponding to the successfully calibrated Kikuchi band, only when the calibration accuracy MAD reaches the preset accuracy threshold, is it determined that the electron backscatter diffraction pattern calibration is successful, ensuring the reliability of the final calibration result. In addition, by calculating the calibration accuracy MAD, a quantitative evaluation index is provided for the calibration result, which helps to compare and analyze the calibration results under different samples or different experimental conditions. Moreover, this application realizes an automated matching and screening process, reduces manual intervention, improves the calibration efficiency, and ensures the reliability of the calibration of the electron backscatter diffraction pattern through a multi-step screening and matching process.

[0046] In some embodiments, before S100, that is, before creating a series of Kikuchi band normal triangles (KNT) according to the Kikuchi bands of the electron backscatter diffraction pattern of the sample to be detected, it includes:

[0047] S010. Preprocessing the acquired target image including the electron backscatter diffraction pattern;

[0048] S020. Detecting the Kikuchi bands and their centerlines in the electron backscatter diffraction pattern according to the target image.

[0049] Specifically, the Kikuchi band center line is an important feature in the electron backscatter diffraction pattern, which represents the intersection line of the crystal plane where Bragg diffraction occurs with the receiving screen after expanding from the electron scattering point on the sample. The center line of each Kikuchi band corresponds to a reflecting crystal plane, and these crystal planes are related to the crystallographic orientation of the sample. The EBSD is installed in the scanning electron microscope. The sample to be measured is tilted to an appropriate angle (usually 60-70°). The scanning electron microscope emits an electron beam and irradiates the surface of the sample to be measured. The backscattered electrons interact with the crystal planes of the sample to be measured. When the Bragg diffraction condition is satisfied, Kikuchi bands are formed. The CCD camera captures the diffraction pattern formed by the backscattered electrons to obtain the target image, and transmits the target image to the computer system. The computer system preprocesses the target image. The preprocessing includes adaptive histogram equalization and Gaussian filtering. Through adaptive histogram equalization, the contrast of the target image can be enhanced, making the details in the target image clearer, especially the edges and structures of the Kikuchi bands, and avoiding false detection or missed detection of the Kikuchi bands due to brightness differences. Gaussian filtering can effectively remove the noise in the target image, reduce interference, make the features of the Kikuchi pattern more prominent, make the denoised target image more stable, and reduce the false detection rate caused by noise. In short, through preprocessing, the detection rate of the Kikuchi bands can be improved and the false detection rate can be reduced. In this application, the Radon transform or the Hough transform can be used to transform the Kikuchi bands into the Radon space or the Hough space, which can more easily automate the identification and detection of the position and direction of the Kikuchi bands, as well as the identification and detection of the center line of the Kikuchi bands, thereby improving the detection accuracy.

[0050] In some embodiments, S100 includes, that is, creating a series of Kikuchi band normal triangles KNT according to the Kikuchi bands of the electron backscatter diffraction pattern of the sample to be detected, including:

[0051] S110, obtaining the actual normal vectors of the diffraction crystal planes corresponding to different Kikuchi bands, and the actual normal vectors are unit vectors;

[0052] S120, connecting the vertices of any three non-collinear actual normal vectors to form different Kikuchi band normal triangles KNT.

[0053] Specifically, when the incident electron beam satisfies the Bragg diffraction condition with the crystal plane in the sample to be detected, diffraction cones will be formed. These diffraction cones intersect with the fluorescent screen to form Kikuchi bands, and the center line of the Kikuchi band is equivalent to the intersection line of the diffraction crystal plane and the fluorescent screen. There is a direct correspondence between the Kikuchi band and the diffraction crystal plane, and each Kikuchi band corresponds to a specific diffraction crystal plane. The normal vector of the diffraction crystal plane, which is the actual normal vector of this application, is a vector perpendicular to the diffraction crystal plane, which represents the orientation of the crystal plane. As Figure 2As shown, according to the first unit vector AB corresponding to the center line of a certain Kikuchi band in the electron backscatter diffraction pattern, and the second unit vector OA formed by the electron beam source point O and a point A on the center line of the Kikuchi band, the actual normal vector N corresponding to the diffraction crystal plane of this Kikuchi band is calculated. K = AB × OA. Among them, the position of the electron beam source point O can be obtained by iterative search using a genetic algorithm. A point A on the center line of the Kikuchi band can be arbitrarily selected on this Kikuchi band, but usually the point at the Kikuchi pole is selected. In this way, the actual normal vectors corresponding to the diffraction crystal planes of all Kikuchi bands in the electron backscatter diffraction pattern can be obtained. As Figure 3 shown, from all the detected actual normal vectors, any three non - collinear actual normal vectors are selected, and the vertices (the vector endpoints of the actual normal vectors) of the three non - collinear actual normal vectors are connected to form a Kikuchi band normal vector triangle KNT (Kikuchi Band Normal Vector Triangle). By continuously selecting different three non - collinear actual normal vectors, multiple Kikuchi band normal vector triangles KNT can be generated.

[0054] By obtaining the actual normal vector and forming the Kikuchi band normal vector triangle KNT in this application, the accuracy and efficiency of crystal orientation analysis can be significantly improved, and at the same time, important support is provided for the microstructure research of complex materials.

[0055] In some embodiments, S200 includes, that is, the creation of a series of reflection plane normal vector triangles RNT according to a pre - constructed theoretical crystal includes:

[0056] S210, construct the theoretical crystal, and the orientation of the theoretical crystal is the same as the orientation of the detector used to detect the electron backscatter diffraction pattern;

[0057] S220, obtain the theoretical normal vectors corresponding to different diffraction crystal planes of the theoretical crystal that satisfy the diffraction conditions; the theoretical normal vectors are unit vectors;

[0058] S230, connect the vertices of any three non - collinear of the theoretical normal vectors to form different reflection plane normal vector triangles RNT.

[0059] Specifically, the lattice type (such as cubic, hexagonal, tetragonal, etc.) and lattice parameters (such as lattice constants a, b, c, lattice angles α, β, γ, etc.) of the crystal to be measured can be obtained first. Among them, for known materials, their lattice types and lattice parameters can usually be obtained from materials science databases or literature. In some cases, it may be necessary to pre-measure the lattice type and lattice parameters of the crystal to be measured by X-ray diffraction (XRD) or other experimental methods. For some new materials or unknown materials, it may be necessary to estimate their lattice types and lattice parameters through theoretical calculations or simulations. Create a crystal structure model as the theoretical crystal according to the orientation information of the EBSD detector, set the orientation of the theoretical crystal to be the same as that of the EBSD detector, and then determine the crystal planes that can undergo diffraction according to the crystal structure, and further calculate the theoretical normal vectors corresponding to each diffraction crystal plane. As Figure 4 shown, from all the calculated theoretical normal vectors, select any three non-collinear theoretical normal vectors, and connect the vertices (the vector endpoints of the theoretical normal vectors) of the three non-collinear theoretical normal vectors to form a reflector normal vector triangle RNT (Reflector Normal Vector Triangle). By continuously selecting different three non-collinear theoretical normal vectors, multiple reflector normal vector triangles RNT can be generated.

[0060] In some embodiments, S300 includes, that is, the screening out of the target reflector normal vector triangle that matches the Kikuchi band normal vector triangle KNT includes:

[0061] S310. Obtain the lengths of the three sides corresponding to each Kikuchi band normal vector triangle KNT and each reflector normal vector triangle RNT;

[0062] S320. Set the first parameter as the first ratio of the first side length to the third side length of the same triangle, set the second parameter as the second ratio of the second side length to the third side length of the same triangle, and set the third parameter as the sum value of the three side lengths of the same triangle, where the length of the second side is greater than or equal to the length of the first side and less than or equal to the length of the third side;

[0063] S330. Map each Kikuchi band normal vector triangle KNT and each reflector normal vector triangle RNT to a three-dimensional parameter coordinate system, where the first axis of the three-dimensional parameter coordinate system represents the first parameter, the second axis of the three-dimensional parameter coordinate system represents the second parameter, and the third axis of the three-dimensional parameter coordinate system represents the third parameter;

[0064] S340. Screen out candidate reflecting surface normal triangles from all the reflecting surface normal triangles RNT according to the first, second, and third parameters of the Kikuchi band normal triangle KNT, where the differences between the first, second, and third parameters of the candidate reflecting surface normal triangles and the corresponding parameters of the Kikuchi band normal triangle KNT are all within a preset parameter threshold;

[0065] S350. Calculate the distance value between the Kikuchi band normal triangle KNT and each corresponding candidate reflecting surface normal triangle, and determine the candidate reflecting surface normal triangle with the distance value less than the preset distance threshold as the target reflecting surface normal triangle corresponding to the Kikuchi band normal triangle KNT.

[0066] Specifically, perform the matching of the Kikuchi band normal triangle KNT and the reflecting surface normal triangle RNT. Search for potential matching items for each Kikuchi band normal triangle KNT among all the reflecting surface normal triangles RNT. Since each characteristic triangle (including the Kikuchi band normal triangle KNT and the reflecting surface normal triangle RNT) has three sides, the side lengths of the three sides are the first side length L1, the second side length L2, and the third side length L3, and the three side lengths can be expressed in ascending order as L1 ≤ L2 ≤ L3. Calculate the first ratio L1 / L3 of the first side length to the third side length of each characteristic triangle, and use the first ratio as the first parameter. Calculate the second ratio L2 / L3 of the second side length to the third side length of each characteristic triangle, and use the second ratio as the second parameter. In addition, use the perimeter of the characteristic triangle, that is, the sum value L1 + L2 + L3 of the first side length, the second side length, and the third side length, as the third parameter. In this way, three parameters, namely L1 / L3, L2 / L3, and L1 + L2 + L3, are constructed according to the side lengths corresponding to the three sides of the characteristic triangle. Through these three parameters, a three-dimensional parameter space, that is, the three-dimensional parameter coordinate system of the present application, can be constructed. Each Kikuchi band normal triangle KNT and each reflecting surface normal triangle RNT can be represented in the three-dimensional parameter coordinate system, and the first axis to the third axis of the three-dimensional parameter coordinate system respectively represent the first parameter to the third parameter.

[0067] In order to search for potential matching items for each Kikuchi band normal triangle KNT among all the reflecting surface normal triangles RNT, the present application performs two-step screening. As Figure 5 shown, the preliminary screening is to screen out the reflecting surface normal triangles RNT whose differences from the three parameters (i.e., the first parameter, the second parameter, and the third parameter) of the current Kikuchi band normal triangle KNT are all within the preset parameter threshold (for example, 0.01, and other values can be set according to requirements), and obtain a series of candidate reflecting surface normal triangles, that is, Figure 5The dashed area on the left side. In summary, the difference between the first parameter of the current Kikuchi band normal triangle KNT and the first parameter of the candidate reflection plane normal triangle is within the preset parameter threshold range, the difference between the second parameter of the current Kikuchi band normal triangle KNT and the second parameter of the candidate reflection plane normal triangle is within the preset parameter threshold range, and the difference between the third parameter of the current Kikuchi band normal triangle KNT and the third parameter of the candidate reflection plane normal triangle is within the preset parameter threshold range. After obtaining the candidate reflection plane normal triangle corresponding to the current Kikuchi band normal triangle KNT through the above preliminary screening, a secondary screening is carried out. The secondary screening uses the Euclidean distance formula to calculate the distance value between the current Kikuchi band normal triangle KNT and each corresponding candidate reflection plane normal triangle in the three-dimensional parameter coordinate system. The distance value is used to measure the similarity degree between the candidate reflection plane normal triangle and the Kikuchi band normal triangle KNT in the three-dimensional parameter space. The smaller the distance value, the more similar the characteristics of the two are, and the higher the possibility of matching. In this way, the candidate reflection plane normal triangle with a distance value less than the preset distance threshold is selected as the target reflection plane normal triangle corresponding to the current Kikuchi band normal triangle KNT. By continuously selecting different Kikuchi band normal triangles KNT, multiple target reflection plane normal triangles corresponding to each Kikuchi band normal triangle KNT can be screened out, thus completing the matching between the Kikuchi band normal triangle KNT and the reflection plane normal triangle RNT.

[0068] In this application, three parameters (L1 / L3, L2 / L3, L1+L2+L3) that are all within the threshold range are selected through preliminary screening for the candidate reflecting surface normal triangles. The reflecting surface normal triangles RNT are preliminarily screened. The preliminary screening ensures that the candidate reflecting surface normal triangles have a high similarity with the current Kikuchi band normal triangle KNT in three key parameters, which can effectively reduce the possibility of mis-matching, so that only those candidate reflecting surface normal triangles that are geometrically similar to the current Kikuchi band normal triangle KNT will enter the next step of precise matching. Moreover, through preliminary screening, those reflecting surface normal triangles RNT that are obviously mismatched can be excluded, thereby narrowing the scope of subsequent search and matching, significantly reducing the number of reflecting surface normal triangles RNT that need to be further processed, significantly reducing the computational amount, and improving the efficiency of the algorithm. Since there may be some candidate reflecting surface normal triangles that are geometrically similar to the current Kikuchi band normal triangle KNT but are actually irrelevant during the matching process. Therefore, in this application, secondary screening is performed to match and screen out the candidate reflecting surface normal triangles with a distance value less than the preset distance threshold, which can more accurately determine the matching item, that is, the target reflecting surface normal triangle, thereby improving the flexibility and accuracy of the matching, and further improving the success rate and accuracy of the calibration. In short, through the two-step filtering of preliminary screening and secondary screening in this application, these irrelevant reflecting surface normal triangles RNT can be effectively excluded, ensuring that the finally matched reflecting surface normal triangle RNT is highly consistent with the Kikuchi band normal triangle KNT to be matched in geometric features, reducing both the computational amount and improving the matching accuracy and algorithm efficiency, which helps to improve the robustness and reliability of the entire calibration process.

[0069] In some embodiments, S400, that is, performing matching counting according to the target reflecting surface normal triangle and the Kikuchi band normal triangle, and assigning the information of the target crystal plane family determined according to the matching counting result to the corresponding Kikuchi band, so as to determine the crystal plane family to which the Kikuchi band belongs to obtain the target Kikuchi band includes:

[0070] S410, assigning the information of the crystal plane family corresponding to the target reflecting surface normal triangle to the candidate Kikuchi band, where the candidate Kikuchi band is the Kikuchi band corresponding to the Kikuchi band normal triangle KNT that matches the target reflecting surface normal triangle;

[0071] S420, counting the number of times the information of the crystal plane family is assigned to each candidate Kikuchi band, and setting the target crystal plane family of the candidate Kikuchi band as the crystal plane family corresponding to the maximum number of assignments;

[0072] S430, assigning the crystal plane indices of the target crystal plane family to the corresponding candidate Kikuchi band to obtain the target Kikuchi band.

[0073] Specifically, during the feature triangle matching process of the above embodiments, after finding the target reflection plane normal triangle that matches the normal triangle of the Kikuchi band (KNT), the information of the crystal plane family represented by the target reflection plane normal triangle is assigned to the corresponding candidate Kikuchi band. Among them, one normal triangle of the Kikuchi band (KNT) can match at least one of the target reflection plane normal triangles, and one target reflection plane normal triangle may match multiple different normal triangles of the Kikuchi band (KNT). When assigning the information of the crystal plane family, the information of the crystal plane family of one target reflection plane normal triangle may be assigned to all the normal triangles of the Kikuchi band (KNT) that match it, rather than only to one normal triangle of the Kikuchi band (KNT). Therefore, each candidate Kikuchi band may be assigned multiple possible crystal plane families. Then, cumulative voting is performed, that is, for each candidate Kikuchi band, the assignment times of each crystal plane family assigned to it are recorded. Repeating this process, after all feature triangle matching is completed, the assignment times of the crystal plane families assigned to each candidate Kikuchi band are counted, and the crystal plane family with the largest assignment times is taken as the target crystal plane family that the candidate Kikuchi band is most likely to belong to. Then, the crystal plane indices of the target crystal plane family are assigned to the corresponding candidate Kikuchi band to obtain the target Kikuchi band.

[0074] For example, assume that there is a first normal triangle of the Kikuchi band (KNT1) actually detected, and its first to third parameters are L1 / L3 = 0.2, L2 / L3 = 0.5, L1 + L2 + L3 = 1.2. Assume that the first to third parameters of the first candidate reflection plane normal triangle 1, the second candidate reflection plane normal triangle 2, and the third candidate reflection plane normal triangle 3 of the theoretical crystal are as follows:

[0075] RNT1: L1 / L3 = 0.21, L2 / L3 = 0.51, L1 + L2 + L3 = 1.21. The three crystal plane families corresponding to the three vertices (normal vectors) of this triangle are {100}, {110}, {111};

[0076] RNT2: L1 / L3 = 0.19, L2 / L3 = 0.49, L1 + L2 + L3 = 1.19. The three crystal plane families corresponding to the three vertices (normal vectors) of this triangle are {110}, {100}, {100};

[0077] RNT3: L1 / L3 = 0.22, L2 / L3 = 0.52, L1 + L2 + L3 = 1.22. The three crystal plane families corresponding to the three vertices (normal vectors) of this triangle are {111}, {100}, {111}.

[0078] Assume that the preset distance threshold is set to 0.02. Therefore, the distance values between RNT1 and RNT2 and KNT1 are less than the preset distance threshold and are selected as the matching items of KNT1, that is, RNT1 and RNT2 are the target reflection plane normal triangles of KNT1. The crystal plane families corresponding to RNT1 are {100}, {110}, {111}, and the crystal plane families corresponding to RNT2 are {110}, {100}, {100}. The information of the crystal plane families corresponding to RNT1 and RNT2 is assigned to the three corresponding candidate Kikuchi bands in KNT1. For each candidate Kikuchi band, record the number of times each crystal plane family it is assigned.

[0079] Assume that after all characteristic triangle matching, a certain Kikuchi band is assigned the crystal plane family {100} 3 times, the crystal plane family {110} 5 times, and the crystal plane family {111} 2 times. Then, select the crystal plane family {110} with the largest assignment count as the target crystal plane family that this Kikuchi band most likely belongs to, which means that this candidate Kikuchi band most likely corresponds to the {110} crystal plane family in the theoretical crystal.

[0080] This application screens out the target crystal plane family through cumulative voting and assigns the information of the target crystal plane family to the corresponding candidate Kikuchi bands. It can not only associate the information of the crystal plane family of the theoretical crystal corresponding to the target reflection plane normal triangle with the candidate Kikuchi bands corresponding to the normal triangle of the Kikuchi band, so that the candidate Kikuchi bands can be initially determined to belong to a specific crystal plane family, in order to transfer the crystal plane family information of the theoretical crystal to the actually detected Kikuchi bands, providing a basis for more accurate crystal plane index determination and orientation calculation in the follow-up. Moreover, in the actual EBSD pattern, there may be multiple similar crystal plane families or Kikuchi band distortions caused by factors such as crystal defects and stress. Through this cumulative voting method, it can effectively reduce the mis-matching caused by reasons such as image noise, Kikuchi band detection errors, and accidental factors (such as fluctuations of the electron beam, changes in the sensitivity of the detector, etc.). Even in some cases where there is a certain deviation between the normal vector of the Kikuchi band and the crystal plane normal vector, through multiple matches and counting, the most practical target crystal plane family is screened out from many possible crystal plane families, avoiding calibration failures or errors caused by the contingency of single match, thereby reducing the probability of mis-matching and improving the accuracy of the calibration result.

[0081] In some embodiments, the S500, that is, the calculated orientation estimation value E corresponding to each Kikuchi band combination includes:

[0082] S510. Combine all the target Kikuchi bands to obtain multiple Kikuchi band combinations;

[0083] S520. Calculate the target crystal plane angle between the two target Kikuchi bands in the Kikuchi band combination according to the actual normal vectors of the diffraction crystal planes corresponding to the two target Kikuchi bands respectively in the Kikuchi band combination;

[0084] S530. Obtain the first target crystal plane index of the Kikuchi band combination, where the first target crystal plane index is the first crystal plane index in the target crystal plane family to which one of the target Kikuchi bands in the Kikuchi band combination belongs;

[0085] S540. Calculate the respective corresponding theoretical crystal plane angles between the first target crystal plane index and all crystal plane indices of the target crystal plane family to which the other target Kikuchi band in the Kikuchi band combination belongs;

[0086] S550. Determine that the crystal plane angle difference between the theoretical crystal plane angle and the target crystal plane angle is less than a preset crystal plane angle difference, and use the crystal plane index corresponding to the theoretical crystal plane angle that is closest to the target crystal plane angle as the second target crystal plane index of the other target Kikuchi band in the Kikuchi band combination;

[0087] S560. Compose a first matrix according to the actual normal vectors corresponding to the Kikuchi band combination, where the first matrix includes the actual normal vectors corresponding to the two target Kikuchi bands respectively in the Kikuchi band combination and their cross products;

[0088] S570. Compose a second matrix according to the theoretical normal vectors of the Kikuchi band combination after being assigned values according to the first target crystal plane index and the second target crystal plane index, where the second matrix includes the theoretical normal vectors corresponding to the two target Kikuchi bands respectively in the Kikuchi band combination and their cross products;

[0089] S580. Use the first matrix corresponding to the Kikuchi band combination as the divisor and the second matrix corresponding to the Kikuchi band combination as the dividend to calculate the orientation estimation value E corresponding to the Kikuchi band combination.

[0090] Specifically, the theoretical crystal plane angle is the angle between two crystal planes calculated according to the crystal plane indices of the crystal. In crystallography, the angle between crystal planes can be calculated through the crystal plane indices, and this angle is an important parameter of the crystal structure and is used to describe the geometric relationship between different crystal planes inside the crystal. Select any two different target Kikuchi bands from all the target Kikuchi bands for combination, and then, according to the actual normal vectors of the diffraction crystal planes corresponding to the two target Kikuchi bands in the current Kikuchi band combination, use the vector angle formula, such as the vector dot product formula θ = ∣arc cos[(N R1 ·N R2 ) / (∣N R1 ∣·∣N R2∣)]∣ is used to calculate the target crystal plane angle between two target Kikuchi bands in the current Kikuchi band combination. In this way, the target crystal plane angle between two target Kikuchi bands in each Kikuchi band combination can be calculated.

[0091] Then, select the first crystal plane index (such as 110) from the target crystal plane family to which a target Kikuchi band in the current Kikuchi band combination belongs, and calibrate the selected crystal plane index as the first target crystal plane index of the current Kikuchi band combination. Since the corresponding crystal plane normal vector can be calculated according to the crystal plane index, and then the target crystal plane angle can be calculated according to the crystal plane normal vector. In this way, the respective corresponding theoretical crystal plane angles between the first target crystal plane index and all crystal plane indexes of the target crystal plane family to which the other target Kikuchi band in the current Kikuchi band combination belongs can be calculated. For example, for the crystal plane index (hkl) of a cubic crystal, it can directly correspond to the components of the normal vector. The normal vector of the crystal plane index (110) is (1, 1, 0). Then, compare whether the crystal plane angle difference between each theoretical crystal plane angle and the target crystal plane angle is less than a preset angle difference (for example, 3°. Other values can be set according to requirements). If the crystal plane angle difference between the current theoretical crystal plane angle and the target crystal plane angle is greater than or equal to the preset angle difference, it indicates that the matching fails, discard it, and switch to judge whether the crystal plane angle difference between the next theoretical crystal plane angle and the target crystal plane angle is less than the preset angle difference. Of course, if the crystal plane angle difference between the current theoretical crystal plane angle and the target crystal plane angle is less than the preset angle difference, it indicates that the current theoretical crystal plane angle matches the target crystal plane angle, and the matching is successful, and it is retained. By selecting different theoretical crystal plane angles for comparison, screen and retain all theoretical crystal plane angles with crystal plane angle differences less than the preset angle difference. Then, take the theoretical crystal plane angle corresponding to the smallest crystal plane angle difference as the theoretical crystal plane angle closest to the target crystal plane angle, and determine the crystal plane index corresponding to it as the second target crystal plane index of the other target Kikuchi band in the current Kikuchi band combination.

[0092] After determining the crystal plane indexes of the two target Kikuchi bands included in the Kikuchi band combination, that is, after determining the first target crystal plane index and the second target crystal plane index, next, use the target Kikuchi band with the determined crystal plane index for rough orientation estimation. That is, according to the respective actual normal vectors of the two target Kikuchi bands included in the current Kikuchi band combination, and the vector obtained by the cross product of the actual normal vectors of the two target Kikuchi bands, a first matrix M is formed K 。 The theoretical normal vectors corresponding to the two Kikuchi bands included in the current Kikuchi band combination after being assigned values according to the first target crystal plane index and the second target crystal plane index, and the vector obtained by the cross product of the theoretical normal vectors corresponding to the two Kikuchi bands respectively form a second matrix M RUsing the first matrix corresponding to the current Kikuchi band combination as the divisor and the second matrix corresponding to the current Kikuchi band combination as the dividend, calculate the orientation estimation value E corresponding to the current Kikuchi band combination, i.e., E = M R / M K By selecting different Kikuchi band combinations, the orientation estimation values E corresponding to all Kikuchi band combinations can be calculated respectively.

[0093] For example, if 10 target Kikuchi bands are detected, 45 Kikuchi band combinations (selecting 2 from 10) can be formed. Determine the crystal plane angle between the two target Kikuchi bands (such as band A and band B) in this combination. Then, select the first crystal plane index from the crystal plane family to which band A most likely belongs, and calculate the theoretical crystal plane angle between the diffraction crystal plane corresponding to the first crystal plane index and all crystal plane indices in the target crystal plane family to which band B most likely belongs. To avoid the situation of incorrect crystal plane family voting, it is considered that the match is valid only when the difference between the calculated theoretical crystal plane angle and the target crystal plane angle measured by the Kikuchi band is less than the preset angle difference. Then, from the screening results with valid matches, select the crystal plane index that best matches the target crystal plane angle, that is, the crystal plane index corresponding to the theoretical crystal plane angle closest to the target crystal plane angle as the target crystal plane index.

[0094] This application establishes Kikuchi band combinations composed of two target Kikuchi bands from all detected target Kikuchi bands. When determining the crystal plane indices corresponding to the two target Kikuchi bands in the Kikuchi band combination, by calculating the crystal plane angle difference between the theoretical crystal plane angle and the measured target crystal plane angle, and setting a preset angle difference, it is considered that the match is valid only when the crystal plane angle difference is less than the preset angle difference, effectively avoiding the situation of incorrect crystal plane family voting, improving the accuracy of crystal plane index determination, and filtering out the scheme that strips not belonging to the same grain are selected into the same Kikuchi band combination, thus achieving a good calibration effect at the grain boundary. In addition, this application calibrates each Kikuchi band combination, increasing the redundancy of calibration and improving the reliability of the calibration result. This application uses matrix division (E = M R / M K ) to calculate the rough orientation estimation value E, providing a high-quality initial value for calculating the target orientation value M, accelerating the convergence of the exact solution, achieving high-precision and high-robustness calibration in complex scenarios (such as grain boundaries), and laying a reliable foundation for subsequent accurate orientation calculation.

[0095] The orientation estimate E obtained by rough calculation may have errors (such as limited information based on two bands), and it needs to be further optimized to improve the calibration accuracy MAD. For example, in the grain boundary or pattern overlap region, there are multiple possible calibration combinations (such as the mixing of Kikuchi bands from different grains), and it is difficult for traditional methods to distinguish the valid solutions. To solve this problem, in some embodiments, the S600, that is, screening out the optimal Kikuchi band combination from all the Kikuchi band combinations according to all the orientation estimates E, and calculating the target orientation value M according to the optimal Kikuchi band combination includes:

[0096] S610. Obtain the rotated normal vector corresponding to the target Kikuchi band according to the orientation estimate E; the rotated normal vector is equal to the product of the orientation estimate E and the actual normal vector corresponding to the target Kikuchi band;

[0097] S620. Calculate the vector angle between the theoretical normal vectors of all the diffraction crystal planes of the target crystal plane family to which the target Kikuchi band belongs and the rotated normal vector, and determine that the crystal plane index corresponding to the smallest vector angle is the final crystal plane index of the target Kikuchi band;

[0098] S630. Count the number of successfully calibrated Kikuchi bands corresponding to each Kikuchi band combination, and determine the Kikuchi band combination with the largest number of successfully calibrated Kikuchi bands as the optimal Kikuchi band combination. The target Kikuchi band for which the vector angle is less than the preset angle threshold by the orientation estimate E is determined to be successfully calibrated;

[0099] S640. Obtain the first set and the second set according to the optimal Kikuchi band combination, and calculate the optimal rotation matrix between the first set and the second set as the target orientation value M. The first set includes the actual normal vectors corresponding to all the detected target Kikuchi bands, and the second set includes the theoretical normal vectors determined according to the final crystal plane indices corresponding to the optimal Kikuchi band combination.

[0100] Specifically, calibrate the crystal plane indices of all the target Kikuchi bands according to the roughly estimated orientation estimate E, that is, rotate the actual normal vector N of the target Kikuchi band K by the orientation estimate E to obtain N K ’ = E·N K . Starting from the first target Kikuchi band among all the Kikuchi bands detected in the above embodiments, determine the crystal plane indices of all the target Kikuchi bands in turn, calculate the actual normal vector N of the first target Kikuchi band K1 and the rotated normal vector N K1 ’ after rotation by the orientation estimate E. Calculate the rotated normal vector N K1 ’ of the first target Kikuchi band and the normal vectors N of each diffraction crystal plane in the target crystal plane family to which the first target Kikuchi band belongsRi The corresponding vector angles β are respectively, β = ∣arc cos[(N K1 ’·N Ri ) / (∣N K1 ’∣·∣N Ri ∣)]∣. Find the minimum value of the vector angle, that is, the crystal plane index that coincides most with the rotated normal vector N K1 ’. Convert the normal vector of the diffraction crystal plane corresponding to the minimum value of the vector angle to obtain the corresponding crystal plane index, and specify this crystal plane index as the final crystal plane index of the first target Kikuchi band. And so on, determine the final crystal plane indices of other target Kikuchi bands. Compare the minimum value of the vector angle of each target Kikuchi band with a preset angle threshold (for example, 5°, and other values can be set according to requirements). If the final angle index of the current target Kikuchi band is less than the preset angle threshold, it indicates that the target Kikuchi band is successfully calibrated. On the contrary, if the final angle index of the current target Kikuchi band is greater than or equal to the preset angle threshold, it indicates that the target Kikuchi band calibration fails. In short, the orientation estimate value E corresponding to the current Kikuchi band combination can be used to calibrate all target Kikuchi bands, and then count how many target Kikuchi bands corresponding to the orientation estimate value E of this Kikuchi band combination are successfully calibrated. In this way, the orientation estimate value E corresponding to each Kikuchi band combination may calibrate more target Kikuchi bands, not just two. Thus, count the number of successfully calibrated target Kikuchi bands after calibrating all target Kikuchi bands with the orientation estimate value E corresponding to each Kikuchi band combination, and determine the Kikuchi band combination corresponding to the orientation estimate value E with the largest number of successfully calibrated target Kikuchi bands as the optimal Kikuchi band combination. The optimal Kikuchi band combination is considered to be the combination that is most likely to correctly reflect the crystal orientation and crystal plane index.

[0101] After determining the optimal Kikuchi band combination through the above embodiments, the calibration results of this optimal Kikuchi band combination will mark all the target Kikuchi bands that meet the conditions and are successfully calibrated. Establish a first set W1 for these successfully calibrated target Kikuchi bands with the actual normal vector N K (calculated from the geometric relationship detected by the Radon transform or Hough transform). W1 is represented as {N K1 , N K2 , ……, N Ki}, where i is the number of successfully calibrated target Kikuchi bands. In addition, for each of the above successfully calibrated target Kikuchi bands, its final crystal plane index has been determined through the above embodiment steps S620 and S630. In this way, according to the theoretical crystallography data, the theoretical normal vector N R corresponding to these final crystal plane indices can be calculated, and finally a second set W2 is obtained. W2 is represented as {N Rz1 , N Rz2 , ……, N Rzi}. Then, the optimal rotation matrix between the first set and the second set can be calculated using the Kabsch algorithm least squares method to minimize the difference between the two sets. In this way, the calculated optimal rotation matrix can be used as the target orientation value M.

[0102] In this application, through the roughly estimated orientation estimate value E, the actual normal vector N of the target Kikuchi band K is rotated into the theoretical crystal space and matched with the theoretical normal vector N in the crystal plane family R to calibrate the crystal plane indices of all target Kikuchi bands, which can ensure that the distribution of crystal plane indices is consistent with the actual crystal structure. The rotation matching of crystal plane normal vectors is effectively utilized, avoiding errors caused by direct matching and improving the accuracy of calibration. By performing the same operation on all combinations and selecting the optimal solution in this application, the reliability of the calibration result can be ensured. The calibration success criterion with the vector angle less than the preset angle threshold helps to filter out unreasonable combinations or effectively filter out Kikuchi bands that do not belong to the same grain, avoiding calibration errors caused by pattern overlap, thereby improving the calibration efficiency and success rate, and significantly improving the accuracy, efficiency, and adaptability of EBSD calibration. In this application, the Kabsch algorithm least squares method is used to calculate the optimal rotation matrix M, which can accurately calculate the target orientation value M, effectively filter out Kikuchi bands that do not belong to the same grain, and avoid calibration errors caused by pattern overlap. Through the least squares method, the influence of errors can be reduced, and the robustness of the calibration result can be improved.

[0103] For example, assume that 10 target Kikuchi bands are detected, and all pairwise combinations of Kikuchi band combinations (a total of 45 combinations) will be traversed. For each Kikuchi band combination (such as band A and band B), after calculating the roughly orientation estimate value E through the above embodiments, the calibration is extended to all target Kikuchi bands. Using the orientation estimate value E of each Kikuchi band combination, rotation and crystal plane matching are performed on all detected target Kikuchi bands (not limited to the two in the combination): the actual normal vector N of each target Kikuchi band K After being rotated by the orientation estimate value E, it is matched with the theoretical normal vector N of the diffraction crystal plane of the target crystal plane family to which the target Kikuchi band belongs R If the match is successful (the angle <5°), it is considered that the target Kikuchi band is successfully calibrated. Count the number of successfully calibrated bands. Each Kikuchi band combination corresponds to a calibration solution. Count the number of all successfully calibrated Kikuchi bands under this calibration solution, and finally select the Kikuchi band combination with the most successfully calibrated bands as the optimal solution.

[0104] In some embodiments, the S700, that is, the calculating the calibration accuracy MAD according to the target orientation value M includes:

[0105] S710. Calculate the deviation angle between the actual normal vector of the target Kikuchi band with successful calibration and the theoretical normal vector determined according to the target orientation value M, the actual normal vector of the successfully calibrated target Kikuchi band, and the final crystal plane indices corresponding to the optimal Kikuchi band combination.

[0106] S720. Calculate the average deviation angle based on the deviation angles corresponding to all successfully calibrated target Kikuchi bands, thereby calculating the calibration accuracy MAD.

[0107] Specifically, perform MAD (mean angular deviation) calculation on all successfully calibrated target Kikuchi bands using the target orientation value M, which can be used as an evaluation of the calibration accuracy MAD. Substitute the target orientation value M, the actual normal vector of the successfully calibrated target Kikuchi band, and the theoretical normal vector determined according to the final crystal plane indices corresponding to the optimal Kikuchi band combination into the following formula to calculate the calibration accuracy MAD:

[0108]

[0109] In the above formulas (1) and (2), N Kj ” is the normal vector after rotation of the actual normal vector result of the j-th successfully calibrated target Kikuchi band by the target orientation value M, and N R is the theoretical normal vector determined according to the final crystal plane indices corresponding to the optimal Kikuchi band combination, and α j is the angle between the actual normal vector of the j-th successfully calibrated target Kikuchi band and the theoretical normal vector determined according to the final crystal plane indices corresponding to the optimal Kikuchi band combination. Among them, α j directly reflects the calibration error of a single Kikuchi band, while the calibration accuracy MAD globally measures the average error of all successfully calibrated Kikuchi bands. The smaller the MAD value, the closer the calibration result is to the true value, and it can more accurately reflect the orientation of the crystal. For electron backscatter patterns with a calibration accuracy MAD less than a preset accuracy threshold (for example, 0.035, and other values can be set according to requirements), it is regarded as successful calibration.

[0110] In this application, by calculating α for each successfully calibrated target Kikuchi band j, which is finally summarized as the calibration accuracy MAD, used to evaluate the overall calibration quality. By setting a preset accuracy threshold, those calibration results or outliers with large deviations can be filtered out, ensuring the consistency of the final calibration results within a certain accuracy range, quantitatively evaluating the calibration quality, ensuring that the results are highly consistent with the theory, filtering noise and false matches, especially prominent at grain boundaries. Moreover, calibration results that do not meet the accuracy requirements can be quickly excluded, thereby reducing unnecessary calculations and time waste, improving the overall calibration speed, effectively improving the accuracy and reliability of calibration, and at the same time optimizing the performance of the calibration algorithm.

[0111] The calibration method of this application first constructs the Kikuchi band normal triangle KNT and the reflection plane normal triangle RNT, then constructs a three-dimensional parameter space for triangle matching to determine the crystal plane family to which the Kikuchi band most likely belongs. Then construct Kikuchi band combinations for rough orientation estimation, assign a self-consistent calibration solution to the Kikuchi bands, and then select the optimal solution from all Kikuchi band combinations, that is, the optimal Kikuchi band combination. Finally, calculate a more accurate target orientation value M according to the optimal solution, and thus obtain the calibration accuracy MAD according to the target orientation value M. One aspect of this application is that the description of the characteristic triangle is extended to three-dimensional space, so that the geometric characteristics of each triangle can be more accurately described by the three parameters L1 / L3, L2 / L3, and L1+L2+L3. This parameter combination can effectively distinguish different triangles, such as similar triangles, etc., thereby reducing false matches, making the matching easier, and improving the accuracy and efficiency of the matching. Another aspect of this application is that after determining the target crystal plane family, it is necessary to construct Kikuchi band combinations, and find the theoretical crystal plane angle that coincides with the target crystal plane angle from the respective target crystal plane families for the two target Kikuchi bands in the Kikuchi band combination, that is, the target crystal plane angle matching. And only when the difference between the theoretical crystal plane angle and the target crystal plane angle is less than the preset angle difference, the matching is considered valid, so that the scheme of Kikuchi bands belonging to different grains being selected into the same Kikuchi band combination is filtered out. In short, the method of this application reduces false matches through the construction of a three-dimensional parameter space and the filtering of crystal plane angle thresholds, automatically filters interference data through the preset angle difference, reduces the need for manual intervention, speeds up the calibration speed, and is especially suitable for complex grain boundary scenarios.

[0112] Exemplarily, 38,232 experimental diffraction patterns were collected from a cubic YSZ bulk sample in a scanning electron microscope (FEI, Magellan 400) equipped with an Oxford Instruments NordlysNano EBSD detector. The acceleration voltage for collecting the diffraction patterns was 20 keV, and the size of the diffraction patterns was 672×512 pixels. The above process was sampled for calibration. The calibration accuracy MAD of the successfully calibrated bands was calculated using the target orientation value M. Diffraction patterns with an MAD value less than the threshold (0.035) were regarded as successfully calibrated. This application can calibrate EBSD diffraction patterns, detect Kikuchi bands in the diffraction patterns, determine the crystal plane indices corresponding to the Kikuchi bands, and the orientation of the diffraction patterns. This application has a good calibration effect at grain boundaries. Assuming that 4 Kikuchi bands (bands 1-4) are detected, 6 pairwise combinations are generated:

[0113]

[0114]

[0115] Optimal solution selection: Combinations 1, 2, and 4 all successfully calibrated 3 bands, and it is necessary to further compare their MAD values. If the MAD value of combination 1 is the lowest, then it is selected as the optimal solution. The original diffraction pattern and the corresponding calibration results are as Figure 6 shown. 10 bands were successfully calibrated for this diffraction pattern, and the MAD value was 0.005977. The orientation calibration distribution map of all diffraction patterns in the dataset is as Figure 7 shown. There is good consistency with the calibration results of Oxford's commercial EBSD system, and the calibration method of this application has achieved better results at grain boundaries. Calibrating 38,232 experimental diffraction patterns was completed in approximately 18 minutes. Figure 7 Comparison of the Oxford orientation map (a) and the orientation map (b) calibrated by the solution of this application. The calibration success rates of the two are 89.07% (a) and 94.42% (b), respectively. Figure 7 The white uncalibrated area in [reference] is due to the absence of diffraction patterns at the holes on the sample surface.

[0116] Furthermore, this application also provides a computer storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the calibration method of the electron backscatter diffraction pattern in the above embodiment.

[0117] The above specific embodiments have further elaborated on the purpose, technical solutions, and beneficial effects of the present application. It should be understood that the above is only one specific embodiment of the present application and is not limited to the protection scope of the present application. Without departing from the gist of the basic features of the present application, the present application can be embodied in various forms. Therefore, the embodiments in the present application are for illustration rather than limitation. Since the scope of the present application is defined by the claims rather than the specification, and all changes falling within the scope defined by the claims or the equivalent scope of the defined scope should be understood to be included in the claims. Any modifications, equivalent replacements, improvements, etc., made within the spirit and principles of the present application shall be included within the protection scope of the present application.

Claims

1. A calibration method for electron backscatter diffraction patterns, characterized in that Including: Creating a series of Kikuchi band normal triangles based on the Kikuchi bands of the electron backscatter diffraction pattern of the sample to be detected, where the Kikuchi band normal triangle is composed of three non-collinear actual normal vectors corresponding to the electron backscatter diffraction pattern; Creating a series of reflection plane normal triangles based on a pre-constructed theoretical crystal, where the reflection plane normal triangle is composed of three non-collinear theoretical normal vectors corresponding to the theoretical crystal; Selecting a target reflection plane normal triangle that matches the Kikuchi band normal triangle; Performing matching counting based on the target reflection plane normal triangle and the Kikuchi band normal triangle, and assigning the information of the target crystal plane family determined according to the matching counting result to the corresponding Kikuchi band, so as to determine the crystal plane family to which the Kikuchi band belongs to obtain a target Kikuchi band; Calculating the orientation estimation value corresponding to each Kikuchi band combination, where the Kikuchi band combination includes two different target Kikuchi bands; Selecting an optimal Kikuchi band combination from all the Kikuchi band combinations according to all the orientation estimation values, and calculating a target orientation value according to the optimal Kikuchi band combination; the optimal Kikuchi band combination is the Kikuchi band combination that makes the number of successfully calibrated target Kikuchi bands the largest; Calculating the calibration accuracy according to the target orientation value, and if the calibration accuracy is lower than a preset accuracy threshold, determining that the electron backscatter diffraction pattern is successfully calibrated.

2. The calibration method of the electron backscatter diffraction pattern according to claim 1, wherein Before creating a series of Kikuchi band normal triangles based on the Kikuchi bands of the electron backscatter diffraction pattern of the sample to be detected, it includes: Preprocessing the acquired target image including the electron backscatter diffraction pattern; Detecting the Kikuchi bands and their centerlines in the electron backscatter diffraction pattern according to the target image.

3. The calibration method of electron backscatter diffraction patterns according to claim 1, characterized in that, Creating a series of Kikuchi band normal triangles based on the Kikuchi bands of the electron backscatter diffraction pattern of the sample to be detected includes: Obtaining the actual normal vectors of the diffraction crystal planes corresponding to different Kikuchi bands, where the actual normal vectors are unit vectors; Connecting the vertices of any three non-collinear actual normal vectors to form different Kikuchi band normal triangles.

4. The calibration method of electron backscatter diffraction pattern according to claim 1, characterized in that Creating a series of reflection plane normal triangles based on a pre-constructed theoretical crystal includes: Constructing the theoretical crystal, where the orientation of the theoretical crystal is the same as the orientation of the detector used to detect the electron backscatter diffraction pattern; Obtaining the theoretical normal vectors corresponding to different diffraction crystal planes of the theoretical crystal that satisfy the diffraction conditions; the theoretical normal vectors are unit vectors; Connecting the vertices of any three non-collinear theoretical normal vectors to form different reflection plane normal triangles.

5. The calibration method of the electron backscatter diffraction pattern according to claim 1, characterized in that, Selecting a target reflection plane normal triangle that matches the Kikuchi band normal triangle includes: Obtaining the lengths of the three sides corresponding to each Kikuchi band normal triangle and each reflection plane normal triangle; Set the first parameter as the first ratio of the first side length to the third side length of the same triangle, set the second parameter as the second ratio of the second side length to the third side length of the same triangle, and set the third parameter as the sum value of the three side lengths of the same triangle, where the length of the second side is greater than or equal to the length of the first side and less than or equal to the length of the third side; Map each of the Kikuchi band normal triangles and each of the reflection plane normal triangles onto a three-dimensional parameter coordinate system, where the first axis of the three-dimensional parameter coordinate system represents the first parameter, the second axis represents the second parameter, and the third axis represents the third parameter; According to the first, second, and third parameters of the Kikuchi band normal triangle, screen out candidate reflection plane normal triangles from all the reflection plane normal triangles, where the differences between the first, second, and third parameters of the candidate reflection plane normal triangles and the corresponding parameters of the Kikuchi band normal triangle are all within a preset parameter threshold; Calculate the distance values between the Kikuchi band normal triangle and each of its corresponding candidate reflection plane normal triangles, and determine the candidate reflection plane normal triangles with the distance values less than a preset distance threshold as the target reflection plane normal triangles corresponding to the Kikuchi band normal triangle.

6. The calibration method of the electron backscatter diffraction pattern according to claim 1, characterized in that, The matching counting based on the target reflection plane normal triangle and the Kikuchi band normal triangle, and assigning the information of the target crystal plane family determined according to the matching counting result to the corresponding Kikuchi band, so as to determine the crystal plane family to which the Kikuchi band belongs to obtain the target Kikuchi band includes: Assign the information of the crystal plane family corresponding to the target reflection plane normal triangle to the candidate Kikuchi band, where the candidate Kikuchi band is the Kikuchi band corresponding to the Kikuchi band normal triangle that matches the target reflection plane normal triangle; Count the number of times the information of the crystal plane family is assigned to each candidate Kikuchi band, and set the target crystal plane family of the candidate Kikuchi band as the crystal plane family corresponding to the maximum number of assignment times; Assign the crystal plane indices of the target crystal plane family to the corresponding candidate Kikuchi band to obtain the target Kikuchi band.

7. The calibration method of the electron backscatter diffraction pattern according to claim 1, characterized in that The calculation of obtaining the orientation estimation value corresponding to each Kikuchi band combination includes: Combine all the target Kikuchi bands to obtain a plurality of Kikuchi band combinations; According to the actual normal vectors of the diffraction crystal planes corresponding to two target Kikuchi bands in the Kikuchi band combination, calculate the target crystal plane angle between the two target Kikuchi bands in the Kikuchi band combination; Obtain the first target crystal plane index of the Kikuchi band combination, where the first target crystal plane index is the first crystal plane index in the target crystal plane family to which one of the target Kikuchi bands in the Kikuchi band combination belongs; Calculate the theoretical crystal plane angles corresponding to each between the first target crystal plane index and all the crystal plane indices of the target crystal plane family to which the other target Kikuchi band in the Kikuchi band combination belongs; Determine that the difference in crystal plane angles between the theoretical crystal plane angle and the target crystal plane angle is less than a preset angle difference, and use the crystal plane index corresponding to the theoretical crystal plane angle that is closest to the target crystal plane angle as the second target crystal plane index of the other target Kikuchi band in the Kikuchi band combination; Form a first matrix according to the actual normal vectors corresponding to the Kikuchi band combination. The first matrix includes the actual normal vectors corresponding to the two target Kikuchi bands in the Kikuchi band combination and their cross products; Form a second matrix according to the theoretical normal vectors of the Kikuchi band combination after being assigned values according to the first target crystal plane index and the second target crystal plane index. The second matrix includes the theoretical normal vectors corresponding to the two target Kikuchi bands in the Kikuchi band combination and their cross products; Use the first matrix corresponding to the Kikuchi band combination as the divisor and the second matrix corresponding to the Kikuchi band combination as the dividend to calculate the orientation estimation value corresponding to the Kikuchi band combination.

8. The calibration method of the electron backscatter diffraction pattern according to claim 7, characterized in that, The screening of the optimal Kikuchi band combination from all the Kikuchi band combinations according to all the orientation estimation values and the calculation of the target orientation value according to the optimal Kikuchi band combination include: Obtain the rotated normal vector corresponding to the target Kikuchi band according to the orientation estimation value; the rotated normal vector is equal to the product of the orientation estimation value and the actual normal vector corresponding to the target Kikuchi band; Calculate the vector angle between the theoretical normal vectors of all diffraction crystal planes of the target crystal plane family to which the target Kikuchi band belongs and the rotated normal vector, and determine that the crystal plane index corresponding to the smallest vector angle is the final crystal plane index of the target Kikuchi band; Count the number of successfully calibrated Kikuchi bands corresponding to each Kikuchi band combination, and determine the Kikuchi band combination with the largest number of successfully calibrated Kikuchi bands as the optimal Kikuchi band combination. The target Kikuchi band with the orientation estimation value making the vector angle less than the preset angle threshold is determined to be successfully calibrated; Obtain a first set and a second set according to the optimal Kikuchi band combination, and calculate the optimal rotation matrix between the first set and the second set as the target orientation value. The first set includes the actual normal vectors corresponding to all the detected target Kikuchi bands, and the second set includes the theoretical normal vectors determined according to the final crystal plane indices corresponding to the optimal Kikuchi band combination.

9. The calibration method of the electron backscatter diffraction pattern according to claim 1, wherein, Calculating the calibration accuracy according to the target orientation value includes: Calculate the deviation angle between the actual normal vector corresponding to the successfully calibrated target Kikuchi band and the theoretical normal vector according to the target orientation value, the actual normal vector of the successfully calibrated target Kikuchi band, and the theoretical normal vector determined according to the final crystal plane index corresponding to the optimal Kikuchi band combination; Obtain the average deviation angle according to the deviation angles corresponding to all the successfully calibrated target Kikuchi bands, so as to calculate the calibration accuracy.

10. A computer storage medium, on which a computer program is stored, characterized in that, When the program is executed by a processor, it implements the calibration method of the electron backscatter diffraction pattern as described in any one of claims 1-9.

Images