Grain size evaluation method

Abstract

To provide a crystal grain size evaluation method capable of easily selecting an appropriate number of intersections for one crystal grain.SOLUTION: The controller executes a preparing step (S1) of preparing an evaluation target image in which respective grain boundaries of a plurality of grains constituting a crystalline structure of a material are displayed. The control unit executes an application step of applying a test line to the evaluation target image (S2). The controller executes a grain-identifying step of identifying, from among the plurality of grains displayed in the evaluation target image, specific grains that are arranged so as to overlap the test line (S3). The control unit 10 executes an intersection point specifying step of specifying an intersection point at which each of the specific grain boundaries of the specific grains intersects with the test line (S5). When there are a plurality of intersection points specified for one of the specific grains, the controller executes an intersection point selection step of selecting, from the plurality of intersection points, intersection points related to a combination in which distances connecting the intersection points are the largest (S6).SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a method for evaluating grain size. [Background technology]

[0002] Various methods for evaluating the grain size of crystalline materials have been proposed. For example, the grain size of metallic materials affects their strength, fatigue resistance, impact resistance, lubricity, and other properties. Therefore, establishing a technique for accurately measuring grain size is crucial for predicting various material properties. Japanese Industrial Standard JIS G 0551 specifies a microscopic test method for measuring the grain size of ferrite and austenite in steel. The intercept method specified in JIS G 0551 includes a method for evaluating grain size by drawing test lines consisting of straight or curved lines on a crystalline structure image showing the grains and counting the number of intersections between the test lines and the grain boundaries. Patent Document 1 discloses a method for calculating the number of intersections between test lines drawn on a binarized crystalline structure image and the grain boundaries, and then using the calculated number of intersections to determine the average grain size of the crystal grains. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 8-101114 Summary of the Invention [Problem to be solved by the invention]

[0004] In some cases, test lines and grain boundaries are aligned in similar positions. In such cases, repeated intersections between test lines and grain boundaries can result in excessive intersections for a single grain. Furthermore, when the grain boundary shape is irregular, excessive intersections can occur for a single grain. In such cases, the evaluation result can indicate that the grain size is smaller than it actually is. The accuracy of the grain size evaluation can be improved by having the examiner manually select appropriate intersections from the excessive intersections. However, the intercept method requires the evaluation of grain size for five test fields, which places a heavy burden on the examiner due to the task of selecting appropriate intersections.

[0005] An object of the present disclosure is to provide a method for evaluating grain size that can easily select an appropriate number of intersections for one grain. [Means for solving the problem]

[0006] The crystal grain size evaluation method according to the present disclosure includes a preparation step of preparing an image to be evaluated in which the grain boundaries of each of a plurality of crystal grains constituting the crystalline structure of a material are displayed; an adding step of adding test lines consisting of straight or curved lines to the image to be evaluated; a crystal grain identification step of identifying, from among the plurality of crystal grains displayed in the image to be evaluated, each of the crystal grains that are arranged overlapping the test lines as a specific crystal grain and storing range information indicating the range that each of the identified specific crystal grains occupies in the crystalline structure image; an intersection identification step of identifying, for each of the specific crystal grains identified in the crystal grain identification step, an intersection where the grain boundary of the specific crystal grain intersects with the test line; and an intersection selection step of selecting, if there are multiple intersections identified in the intersection identification step for one of the specific crystal grains, the intersection associated with the combination that connects the intersections to form the longest distance, and deleting the other intersections.

[0007] According to the present disclosure, it is possible to provide a method for evaluating grain size that can easily select an appropriate number of intersections for one grain. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is an overall structural diagram of a grain size evaluation system 100. FIG. [Figure 2] 10 is a flowchart showing a grain size evaluation process. [Figure 3] 10 is a flowchart showing a crystal grain identification step. [Figure 4] 1A to 1C are explanatory diagrams showing examples of an evaluation target image 30 and an evaluation target image 30 with a test line TL added thereto. [Figure 5] 3A and 3B are explanatory diagrams showing examples of a first extracted image 31 and a second extracted image 41. FIG. [Figure 6] FIG. 2 is an explanatory diagram showing an example of a one-side image 51. [Figure 7] 10A and 10B are explanatory diagrams showing examples of a fourth extracted image 32 and a fifth extracted image 42. [Figure 8] 5A and 5B are explanatory diagrams showing examples of the other-side image 52 and the generated image 55. FIG. [Figure 9] 10 is an explanatory diagram showing examples of specific crystal grains SG, specific crystal grain boundaries SGB and SGB1, intersections P1, P2, and P3 with the test line TL, and the center P0 of the test line TL. FIG. [Figure 10] This is an explanatory diagram showing a case where the angle formed by the straight lines P0Pa and P0Pb connecting the two intersection points Pa, Pb between the specific grain boundary SGBA of the specific crystal grain SGA and the test line TL and the center P0 of the test line TL is angle θ. DETAILED DESCRIPTION OF THE INVENTION

[0009] First, embodiments of the present disclosure will be listed and described. (1) The crystal grain size evaluation method of the present disclosure includes a preparation step of preparing an image to be evaluated in which the crystal boundaries of each of a plurality of crystal grains constituting the crystalline structure of a material are displayed; an application step of applying test lines consisting of straight or curved lines to the image to be evaluated; a crystal grain identification step of identifying, as a specific crystal grain, each of the crystal grains that are arranged overlapping the test lines from among the plurality of crystal grains displayed in the image to be evaluated; an intersection identification step of identifying an intersection where the test line intersects with the crystal grain boundaries of each of the specific crystal grains identified in the crystal grain identification step; and an intersection selection step of selecting, from the plurality of intersections, the intersection associated with the combination that provides the longest distance between the intersections, if there are multiple intersections identified in the intersection identification step for one of the specific crystal grains.

[0010] When evaluating grain size using the intercept method, test lines are placed on the image to be evaluated. When test lines and grain boundaries are arranged side by side in similar positions, the repeated intersections between the test lines and the grain boundaries may result in excessive intersections for a single grain. Furthermore, when the shape of the grain boundaries is irregular, excessive intersections for a single grain may result. The grain size evaluation method according to the present disclosure identifies each of the crystal grains arranged overlapping the test lines as a specific crystal grain. Then, the intersections where each grain boundary of the specific crystal grain intersects with the test line are identified. If multiple intersections are identified for a single specific crystal grain, the intersections corresponding to the combination with the longest distance between the intersections are selected from the multiple intersections. In this way, the grain size evaluation method according to the present disclosure can easily select an appropriate number of intersections for a single crystal grain.

[0011] (2) In the crystal grain size evaluation method described in (1), it is preferable that in the intersection selection step, the intersections corresponding to the combination in which the angle formed by two straight lines connecting each of the two intersections extracted from the plurality of intersections identified in the intersection identification step for one specific crystal grain and a predetermined point located at a position that does not overlap with the test line in the image to be evaluated is maximized are selected.

[0012] In this case, the grain size evaluation method can easily select an appropriate combination of intersection points from among a plurality of intersection points for one grain.

[0013] (3) In the method for evaluating grain size described in (2), it is preferable that the test line forms a circle, and the predetermined point is set at the center of the circle that forms the test line.

[0014] In this case, since the length between the predetermined point and the intersection point is uniformly determined, it is easier to derive the angle between two lines connecting the predetermined point and each of two intersection points extracted from the multiple intersection points, and the derived angles can be compared. Therefore, the grain size evaluation method can easily select an appropriate combination of intersection points from multiple intersection points for one crystal grain.

[0015] (4) In the crystal grain size evaluation method according to any one of (1) to (3), the crystal grain specifying step includes a first extraction step of extracting from the evaluation target image an image showing the crystal grain boundary present on one side of the test line assigned to the evaluation target image in the assigning step, the second extraction step of extracting a first largest-area image which is an image of the closed region having the largest area among a plurality of closed regions formed by the images showing the crystal grain boundary extracted in the first extraction step, the third extraction step of extracting from the evaluation target image a one-side image present on the one side of the test line in the first largest-area image extracted in the second extraction step, and the third extraction step of extracting from the evaluation target image a other side of the test line assigned to the evaluation target image in the assigning step, the other side of the test line assigned to the evaluation target image in the assigning step, the other side of the test line assigned to the evaluation target image in the assigning step. It is preferable that the method includes a fourth extraction step of extracting an image showing the crystal grain boundary present on one side; a fifth extraction step of extracting a second largest area image which is an image of the closed area having the largest area among a plurality of closed areas formed by the image showing the crystal grain boundary extracted in the fourth extraction step; a sixth extraction step of extracting an other-side image present on the other side of the test line in the second largest area image extracted in the fifth extraction step; a generation step of generating images of each of the specific crystal grains by combining the one-side image extracted in the third extraction step and the other-side image extracted in the sixth extraction step; and an identification step of identifying the specific crystal grains by performing a labeling process on the images of each of the specific crystal grains generated by the generation step.

[0016] In this case, each of the crystal grains that overlap the test line is appropriately identified as a specific crystal grain. Therefore, an appropriate number of intersections are selected for each specific crystal grain. Therefore, the crystal grain size evaluation method can improve the accuracy of the crystal grain size evaluation.

[0017] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. The drawings are used to explain technical features that may be employed by the present invention, and the configurations of the described devices are not intended to be limiting but are merely illustrative examples. First, with reference to FIG. 1 , the system configuration of a grain size evaluation system 100 that executes the grain size evaluation method according to the present disclosure will be described. The grain size evaluation method is a method for evaluating grain size based on a crystal structure image, which is an image that represents the crystal structure of a predetermined region in a material. The grain size evaluation system 100 includes a grain size evaluation device 1 and a microscope 2. The microscope 2 is an example of a means for preparing a crystal structure image to be evaluated by the grain size evaluation system 100.

[0018] In the grain size evaluation method according to the present disclosure, the material to be evaluated for grain size is not particularly limited. The material to be evaluated may be either a metallic material or a non-metallic material, or may include both metals and non-metals. Furthermore, the material to be evaluated may include substances composed of crystals, such as inorganic and organic substances. For example, the material to be evaluated may be an alloy composed of one or more metallic and non-metallic elements. In this embodiment, the material to be evaluated is primarily steel. When the material to be evaluated is steel, the surface of the material or the cut surface obtained by cutting the material in any direction is polished, and then an appropriate etching solution is used to reveal the grain boundaries of the crystal grains. A crystalline structure image is obtained by photographing the crystalline structure with the revealed grain boundaries using a microscope 2.

[0019] In this embodiment, the microscope 2 is a scanning electron microscope (SEM). The crystalline structure image is an SEM image obtained by capturing an image using the microscope 2, which is a scanning electron microscope. The microscope 2 can be connected to an external device connection unit 18 (described later) of the crystalline grain size evaluation device 1 and transmit image data showing the crystalline structure image to the crystalline grain size evaluation device 1. The microscope 2 may be, for example, a transmission electron microscope, an optical microscope, or the like, as long as it can capture an image of the crystalline structure of the material. The magnification of the crystalline structure image may be any magnification suitable for the cutting method defined in JIS G 0551 (described later).

[0020] The grain size evaluation device 1 may be a device dedicated to grain size evaluation, or may be a general-purpose device such as a personal computer (PC) with a dedicated application for performing grain size evaluation installed. In this embodiment, a grain size evaluation device 1 using a general-purpose PC is exemplified.

[0021] The grain size evaluation apparatus 1 includes a control unit 10, a storage unit 11, and an external device connection unit 18. The control unit 10 includes a CPU that controls the entire grain size evaluation apparatus 1. The control unit 10 is electrically connected to the storage unit 11 via a data bus (not shown). The storage unit 11 includes a nonvolatile storage device such as a hard disk drive (HDD). In this embodiment, the HDD is an example of a non-transitory storage medium. The non-transitory storage medium may be any storage medium capable of storing information regardless of the storage period. The non-transitory storage medium may not include a temporary storage medium (e.g., a transmitted signal). While the storage device in this embodiment is a hard disk drive (HDD), the storage device may be configured with other non-transitory storage media, such as flash memory or RAM, that can retain data regardless of the storage period. The storage unit 11 stores a program that causes the control unit 10 to execute a grain size evaluation process, which will be described later with reference to FIG. 2. The control unit 10 functions as an example of a processor that executes the grain size evaluation process by loading the program stored in the storage unit 11. The storage unit 11 may be provided outside the grain size evaluation device 1.

[0022] The program executed by the control unit 10 is provided by being recorded on, for example, a magnetic disk, an optical disk, a semiconductor memory, or a similar recording medium, and is stored in the memory unit 11 or the like. The recording medium on which the program is recorded may be in any storage format as long as it is a recording medium readable by the control unit 10. The program may also be configured to be pre-installed in the grain size evaluation device 1.

[0023] The control unit 10 is electrically connected to an operation unit 12, such as a keyboard, touch panel, or mouse, and a display unit 13, such as a display that displays images, via an input / output interface (not shown). The external device connection unit 18 is a connector for electrically connecting to an external information device, such as the microscope 2. The control unit 10 is electrically connected to the microscope 2 via the external device connection unit 18 and can receive image data representing a crystalline structure image acquired by the microscope 2. The control unit 10 may also be electrically connected to an information device that stores image data representing a crystalline structure image acquired by the microscope 2 via the external device connection unit 18 and receive image data from the information device. The control unit 10 may also be electrically connected to a storage medium, such as a USB memory, that stores image data representing a crystalline structure image acquired by the microscope 2 via the external device connection unit 18 and receive image data representing a crystalline structure image from the storage medium. The control unit 10 may also be electrically connected to an external network (not shown) via the external device connection unit 18 and receive image data representing a crystalline structure image from an external device connected to the network. The external device connection unit 18 is not limited to a connector that allows wired communication, but may be configured to allow wireless connection of an external device.

[0024] The grain size evaluation method according to this embodiment will be described with reference to FIG. 2. The control unit 10 of the grain size evaluation device 1 executes the grain size evaluation process shown in FIG. 2 to execute the grain size evaluation method. The grain size evaluation process is executed when the control unit 10 detects an instruction to start the grain size evaluation process. In this embodiment, the instruction to start the grain size evaluation process is executed by the user via the operation unit 12. In the following description, each processing step will be abbreviated as "S".

[0025] When the grain size evaluation process is started, the control unit 10 executes a preparation step (S1). In the preparation step, the control unit 10 prepares an evaluation target image 30 in which the grain boundaries of each of the multiple crystal grains that make up the crystalline structure of the material are displayed. FIG. 4(A) shows an example of the evaluation target image 30. Specifically, the control unit 10 transmits an instruction to the microscope 2 via the external device connection unit 18 to transmit image data of the crystalline structure image captured by the microscope 2. Upon receiving this instruction, the microscope 2 generates image data of the captured crystalline structure image and transmits it to the grain size evaluation device 1.

[0026] It is preferable that the grain boundaries of the crystal grains are clearly visible in the evaluation target image 30 that is the subject of the grain size evaluation process. The control unit 10 may use the image corresponding to the image data received from the microscope 2 as the evaluation target image 30 to be used as the subject of the grain size evaluation process. The control unit 10 may also obtain the evaluation target image 30 by processing the crystalline structure image corresponding to the image data transmitted from the microscope 2.

[0027] As an example of processing the crystalline structure image corresponding to the image data, a binarization process may be performed in which a predetermined pixel value is set as a threshold for the pixel values ​​of each pixel constituting the crystalline structure image, and the crystalline structure image is converted into two colors, black and white. This aims to clarify the grain boundaries appearing in the crystalline structure image. The image obtained as a result of this binarization process may then be used as the evaluation target image 30. For example, the binarization process may blacken many of the grain boundaries contained in the crystalline structure image and whiten many of the grains, thereby obtaining an evaluation target image 30 in which the grain boundaries of each of the multiple grains constituting the crystalline structure of the material are visible. Alternatively, the binarization process may whiten many of the grain boundaries contained in the crystalline structure image and blacken many of the grains, thereby obtaining the evaluation target image 30. In this embodiment, the control unit 10 obtains, as the evaluation target image 30, an image in which many of the grain boundaries contained in the crystalline structure image are blackened and many of the grains are whitened by the binarization process. As shown in FIG. 4(A), in the evaluation object image 30, a large number of crystal grains G and the respective grain boundaries GB of the large number of crystal grains G appear.

[0028] As another example of processing a crystalline structure image corresponding to image data, a learning model obtained by learning a large number of crystalline structure images through machine learning may be used to clearly extract grain boundaries within the crystalline structure image. The image obtained as a result of such grain boundary extraction may be the evaluation target image 30. As another example of processing a crystalline structure image corresponding to image data, an evaluator may manually add or delete grain boundaries in areas of the crystalline structure image where the grain boundaries are not clearly visible. The image obtained as a result of such manual addition or deletion of grain boundaries may be the evaluation target image 30. The evaluation target image 30 may also be obtained by performing other processing such as removing noise from the crystalline structure image or correcting the overall brightness. The control unit 10 may acquire, via an external network (not shown), the crystalline structure image itself captured by the microscope 2 or an image after various processing operations have been performed on the crystalline structure image as the evaluation target image 30.

[0029] The control unit 10 executes an assignment step (S2) in which a test line TL is assigned to the evaluation target image 30 obtained in the preparation step. The cutting method defined by JIS G 0551 counts the number of intersections between grain boundaries and test lines, and derives a grain size number indicating the grain size based on the average number of intersections per mm of the test line. The test line TL is the test line used in this cutting method. In this embodiment, the test line TL is a curve that forms a circle. However, as defined by JIS G 0551, the test line TL may be another curve or a straight line. FIG. 4(B) shows an example in which the test line TL is assigned to the evaluation target image 30.

[0030] The control unit 10 executes a crystal grain identification step (S3) to identify each of the crystal grains G that are arranged overlapping the test line TL as a specific crystal grain SG from among the multiple crystal grains G displayed in the evaluation target image 30.

[0031] The details of the grain identification step indicated by S3 in Fig. 2 will be described with reference to Fig. 3. When the grain identification step is started, the control unit 10 executes a first extraction step (S11) in which an image indicating a grain boundary GB existing on one side of the test line TL assigned to the evaluation target image 30 in the assignment step is extracted from the evaluation target image 30. Specifically, as shown in Fig. 5(C), the control unit 10 generates a first extracted image 31 by extracting a line drawing indicating a grain boundary GB existing on the outside of the test line TL, which is one side of the direction intersecting the test line TL, from the evaluation target image 30. As a result, the line drawing indicating a grain boundary GB existing on the inside of the test line TL, which is the other side opposite to the one side of the direction intersecting the test line TL, is removed, and the first extracted image 31 is obtained as an image in which only the line drawing indicating a grain boundary GB existing on the outside corresponding to one side of the test line TL remains.

[0032] The control unit 10 executes a second extraction step (S12) to extract a first maximum-area image 411, which is an image of the largest area among the multiple closed regions formed by the image showing the grain boundary GB extracted in the first extraction step. Specifically, the control unit 10 executes a labeling process to define each closed region formed by the line drawing showing the grain boundary GB in the first extracted image 31 as one group and label the group. The control unit 10 can use a known image processing method for this labeling process. The control unit 10 analyzes the image features of each labeled closed region and acquires various information about each closed region, such as information indicating its area and position. The information indicating the position may be acquired based on coordinate values ​​obtained by setting an origin at an appropriate position in the first extracted image 31 and establishing an xy coordinate system based on the origin. The origin of the xy coordinate system may be set at a predetermined position, such as a corner of the first extracted image 31, or at the center of the circle drawn by the test line TL.

[0033] Next, the control unit 10 compares the areas of the labeled regions and extracts the image of the closed region with the largest area in the second extracted image 41 as the first largest-area image 411. FIG. 5(D) shows an example of the second extracted image 41 obtained as a result of performing a labeling process on each of the multiple closed regions formed in the first extracted image 31. The second extracted image 41 conceptually illustrates that each of the multiple closed regions is labeled by painting adjacent closed regions with different densities. As shown in FIG. 5(D), in the evaluation target image 30, the grain boundaries GB of the multiple crystal grains G that existed outside the test line TL form individual closed regions. Therefore, each of the regions corresponding to each crystal grain G is labeled by the labeling process. Meanwhile, the grain boundaries GB of the crystal grains G that existed inside the test line TL in the evaluation target image 30 were removed in the first extraction step. Therefore, the regions corresponding to the numerous crystal grains G present inside the test line TL in the evaluation target image 30 are grouped into one region. This grouped region is combined with a region formed by connecting the lines indicating the grain boundaries GB of the numerous crystal grains G that extended toward the outside of the test line TL to form one region. This region formed by the regions corresponding to the numerous crystal grains G is extracted as the first maximum-area image 411.

[0034] The control unit 10 executes a third extraction step (S13) of extracting a one-side image 51 located on one side of the test line TL from the first maximum area image 411 extracted in the second extraction step. Specifically, the control unit 10 removes the image located inside the test line TL from the first maximum area image 411 extracted in the second extraction step, and extracts only the image located outside the test line TL, thereby generating the one-side image 51. As shown in FIG. 6(E), the one-side image 51 is formed by lines indicating grain boundaries GB that are tangent to the test line TL and located outside the test line TL.

[0035] The control unit 10 also executes a fourth extraction step (S15) of extracting from the evaluation target image 30 an image showing grain boundaries GB that exist on the other side of the test line TL that was assigned to the evaluation target image 30 in the assignment step, opposite to one side of the direction intersecting the test line TL. Specifically, as shown in FIG. 7(F), the control unit 10 generates a fourth extracted image 32 by extracting line drawings showing grain boundaries GB that exist on the inside of the test line TL, which is the other side of the direction intersecting the test line TL, from the evaluation target image 30. As a result, the line drawings showing grain boundaries GB that exist on the outside of the test line TL, which is the one side of the direction intersecting the test line TL, are removed, and the fourth extracted image 32 is obtained as an image in which only line drawings showing grain boundaries GB that exist on the inside corresponding to the other side of the test line TL remain.

[0036] The control unit 10 executes a fifth extraction step (S16) to extract a second maximum-area image 412, which is an image of the largest area among the multiple closed regions formed by the image showing the grain boundary GB extracted in the fourth extraction step. Specifically, the control unit 10 executes a labeling process to label each of the closed regions formed by the line drawings showing the grain boundary GB in the fourth extracted image 32. The control unit 10 analyzes the image features of each labeled closed region to acquire various information about each closed region, such as information indicating the area and information indicating the position. The information indicating the position may be acquired based on coordinate values ​​obtained by setting an origin at an appropriate position in the fourth extracted image 32 and establishing an x-y coordinate system based on the origin. The origin of the x-y coordinate system may be set at a predetermined position, such as a corner of the fourth extracted image 32, or may be set at the center of the circle drawn by the test line TL. Furthermore, it is preferable that the origin of the x-y coordinate system of the fourth extracted image 32 be set at the same position as the origin of the first extracted image 31.

[0037] Next, the control unit 10 compares the areas of the labeled regions and extracts the image of the closed region with the largest area in the fifth extracted image 42 as the second largest-area image 412. FIG. 7(G) shows an example of the fifth extracted image 42 obtained as a result of performing a labeling process on each of the multiple closed regions formed in the first extracted image 31. Similarly to the second extracted image 41, the fifth extracted image 42 conceptually illustrates that each of the multiple closed regions is labeled by filling adjacent closed regions with different densities. As shown in FIG. 7(G), in the evaluation target image 30, the grain boundaries GB of the multiple crystal grains G that existed inside the test line TL form individual closed regions. Therefore, each region corresponding to each crystal grain G is labeled by the labeling process. Meanwhile, the grain boundaries GB of the crystal grains G that existed outside the test line TL in the evaluation target image 30 were removed in the fourth extraction step. Therefore, the regions corresponding to the numerous crystal grains G that existed outside the test line TL in the evaluation target image 30 are grouped into one region. This grouped region is combined with a region formed by connecting the lines indicating the grain boundaries GB of the numerous crystal grains G that extended toward the inside of the test line TL to form one region. This region formed by the correspondences to the numerous crystal grains G is extracted as the second maximum-area image 412.

[0038] The control unit 10 executes a sixth extraction step (S17) of extracting an other-side image 52 located on the other side of the test line TL from the second maximum-area image 412 extracted in the fifth extraction step. Specifically, the control unit 10 removes the image outside the test line TL from the second maximum-area image 412 extracted in the fifth extraction step, and extracts only the image inside the test line TL to generate the other-side image 52. As shown in FIG. 8(H), the other-side image 52 is formed by lines indicating grain boundaries GB that are tangent to the test line TL and located inside the test line TL.

[0039] The control unit 10 executes a generating step (S18) in which the one-side image 51 extracted in the third extraction step and the other-side image 52 extracted in the sixth extraction step are combined to generate images of the specific crystal grains SG. The images generated by the generating step are referred to as generated images 55. The one-side image 51 is configured with lines representing the grain boundaries GB that are tangent to the test line TL and located outside the test line TL. The other-side image 52 is configured with lines representing the grain boundaries GB that are tangent to the test line TL and located inside the test line TL. Therefore, as shown in FIG. 8(I), by combining the one-side image 51 and the other-side image 52, the lines representing the grain boundaries GB that are located at the same positions on the test line TL are connected. As a result, in the generated image 55, closed regions representing the crystal grains G that are located on the test line TL are formed. Note that the generated image 55 generated based on the evaluation target image 30 is also included in the evaluation target image 30 because it is an image that is subject to grain size evaluation in the grain size evaluation process.

[0040] As described above, among the many crystal grains G displayed in the evaluation target image 30, the crystal grains G that are arranged overlapping the test line TL are referred to as specific crystal grains SG. That is, the control unit 10 can generate each image of the specific crystal grains SG by synthesizing the one-side image 51 and the other-side image 52 to generate the generated image 55. Hereinafter, the crystal grain boundary GB of the specific crystal grain SG will be referred to as the specific crystal grain boundary SGB.

[0041] The control unit 10 executes an identification step (S19) in which the control unit 10 identifies each of the specific crystal grains SG by performing a labeling process on each image showing the specific crystal grains SG in the generated image 55 generated by the generation step. Specifically, the control unit 10 executes a labeling process in which each of the closed regions overlapping the test line TL is defined as one group. As a result, the control unit 10 acquires various information about the labeled specific crystal grains SG, such as information indicating the area and information indicating the position. The control unit 10 ends the crystal grain identification step and returns the process to the crystal grain size evaluation process.

[0042] Returning to the explanation of Fig. 2, the control unit 10 executes an intersection identifying step (S5) of identifying an intersection where the specific grain boundary SGB of each specific grain SG identified in the grain identifying step intersects with the test line TL. Specifically, the control unit 10 extracts, for each specific grain SG, an intersection between the specific grain boundary SGB corresponding to the specific grain SG and the test line TL, and obtains information indicating the position of the extracted intersection.

[0043] The control unit 10 may execute the intersection identifying step by extracting the intersection between the specific grain boundary SGB and the test line TL for one specific crystal grain SG, and then extracting the intersection between the specific grain boundary SGB and the test line TL for another specific crystal grain SG. Alternatively, the control unit 10 may extract all of the intersections between the test line TL and each of the specific crystal grain boundaries SGB corresponding to all of the specific crystal grains SG, and then determine which region of the specific crystal grain SG each of the extracted intersections belongs to, thereby identifying the intersection between the specific crystal grain boundary SGB corresponding to each specific crystal grain SG and the test line TL.

[0044] When multiple intersections are identified for one specific crystal grain SG in the intersection identification step, the control unit 10 executes an intersection selection step (S6) to select, from the multiple intersections, the intersections associated with the combination that provides the longest distance between the intersections. When the test line TL is a curve that forms a circle, as in this embodiment, there are often two intersections between the specific crystal grain boundary SGB corresponding to one specific crystal grain SG and the test line TL. However, depending on the shape of the specific crystal grain SG, three or more intersections may be identified for one specific crystal grain SG. Furthermore, when the specific crystal grain boundary SGB is arranged in a similar position to the test line TL, the test line TL and the specific crystal grain boundary SGB may intersect repeatedly. In such cases, multiple intersections may be identified for one specific crystal grain SG.

[0045] 9(J) and 9(K), a specific crystal grain SG1, which is one of the many specific crystal grains SG and is located in the upper left corner of the test line TL, will be specifically described. The specific crystal grain boundary SGB of the specific crystal grain SG1 is designated as the specific crystal grain boundary SGB1. As shown in FIG. 9(J), the upper left corner of the specific crystal grain boundary SGB1 has a portion extending outside the test line TL and a portion extending inside the test line TL. Therefore, as shown in FIG. 9(K), three intersections between the specific crystal grain boundary SGB1 and the test line TL are identified for the specific crystal grain SG1. The three identified intersections are designated as intersections P1, P2, and P3. In the grain size evaluation using the intercept method, the more intersections with the test line TL are counted, the larger the derived grain size number. The grain size number represents the grain size of the crystal grain G. The larger the grain size number, the smaller the grain size of the crystal grain G. Therefore, if the number of intersections between the specific grain boundaries SGB corresponding to each of the many specific grains SG and the test line TL is simply counted, a grain size number larger than the grain size number expected from the grain size of the specific grains SG that are actually arranged on the test line TL may be derived. Because the grain size of a material affects the characteristics of the material, a high degree of accuracy is required in evaluating the grain size.

[0046] The control unit 10 counts the number of intersections identified in the intersection identifying step that exist within a region occupied by one specific crystal grain SG. The control unit 10 can perform this counting by comparing information indicating the position of the specific crystal grain SG to be counted with information indicating the positions of the intersections identified in the intersection identifying step. If the number of counted intersections is 1 or 2, the control unit 10 directly uses the counted number of intersections as the number of intersections to be used in the cutting method. If the number of counted intersections is 3 or more, the control unit 10 selects the intersections associated with the combination with the longest distance between the intersections as the number of intersections to be used in the cutting method. This allows the control unit 10 to select, from among the multiple intersections, an intersection that allows for more accurate evaluation of the grain size of the specific crystal grain SG.

[0047] The control unit 10 may select the intersection points according to the combination that maximizes the distance between the intersection points by measuring the distances between the intersection points. For example, for the specific grain boundary SGB1 in which three intersection points P1, P2, and P3 are identified, the control unit 10 may select the intersection points according to the combination that maximizes the distance between the intersection points by measuring the lengths of the line segments P1P2, P1P3, and P2P3. The coordinate values ​​of the intersection points P1, P2, and P3 may be used to measure the lengths of the line segments. For the specific grain SG1 shown in FIG. 9(K), the length of the line segment P1P3 is longer than the lengths of the line segments P1P2 and P2P3, so the control unit 10 selects the intersection points P1 and P3 according to the combination of the line segment P1P3.

[0048] Furthermore, the control unit 10 may select an intersection point relating to a combination that maximizes the angle formed by two lines connecting each of two extracted intersection points from the plurality of intersection points and a predetermined point provided at a position that does not overlap with the test line TL in the generated image 55, which is the evaluation target image 30. Even with this method, the control unit 10 can select an intersection point from the plurality of intersection points that allows for more accurate evaluation of the grain size of the specific grain boundary SGB.

[0049] 10, an explanation will be given of the evaluation of the angle formed by two lines connecting each of the two intersections between the specific crystal grain boundary SGB and the test line TL and a predetermined point provided at a position that does not overlap the test line TL in the generated image 55 when there are two intersections between the specific crystal grain boundary SGB and the test line TL. In the evaluation target image 30 shown in FIG. 10, the explanation will be given using the specific crystal grain SGA that is arranged overlapping the upper right part of the circle formed by the test line TL, among the multiple specific crystal grains SG arranged overlapping the test line TL.

[0050] The specific grain boundary SGB of the specific grain SGA is designated as the specific grain boundary SGBA. As shown in FIG. 10, the specific grain boundary SGBA and the test line TL form two intersections. These two intersections are designated as intersections Pa and Pb. To evaluate the angles formed by two lines connecting each of the intersections Pa and Pb with a predetermined point on the evaluation target image 30 that does not overlap with the test line TL, the predetermined point is set at the center P0 of the circle that constitutes the test line TL. The control unit 10 can actually measure the angle θ formed by the line P0Pa connecting the center P0 and the intersection Pa, and the line P0Pb connecting the center P0 and the intersection Pb. The control unit 10 can apply this method to cases where there are three or more intersections between the specific grain boundary SGB and the test line TL. Specifically, the control unit 10 measures the angles formed by the two lines connecting each of two intersections extracted from the three or more intersections and the center P0. The larger the angle formed by the two lines connecting each of the two intersections and the center P0, the greater the distance connecting the intersections. Therefore, the control unit 10 can select the intersections related to the combination that results in the largest measured angle, thereby selecting the intersections related to the combination that results in the largest distance connecting the intersections.

[0051] 9(K), the angle between the lines P0P1 and P0P3 is larger than the angle between the lines P0P1 and P0P2 and the angle between the lines P0P2 and P0P3. Therefore, the control unit 10 selects the intersection points P1 and P3 associated with the combination that maximizes the measured angle.

[0052] In addition to actually measuring each of the angles formed by the two lines, the control unit 10 can also evaluate each of the angles formed by the two lines by performing a predetermined calculation. As shown in Fig. 10, vector P0Pa is represented as vector a, and vector P0Pb is represented as vector b. Furthermore, since the angle formed by lines P0Pa and P0Pb is angle θ, the angle formed by vector a and vector b is also angle θ. Therefore, using the definition of the dot product of vector a and vector b, the following equation (1) holds:

[0053]

number

[0054] In Equation (1), the magnitudes of vector a and vector b correspond to the lengths of line segments P0Pa and P0Pb, respectively, and the lengths of these line segments correspond to the radius of the circle formed by the test line TL. The control unit 10 can acquire the value of the radius of the circle formed by the test line TL in advance when adding the test line TL to the evaluation target image 30 in the adding step. Alternatively, the control unit 10 can acquire the radius of the circle formed by the test line TL by actually measuring it when executing the intersection point selecting step. Furthermore, the control unit 10 can calculate the dot product of vector a and vector b by acquiring the x- and y-components of the center P0 of the circle formed by the test line TL and the intersection points Pa and Pb in the xy coordinate system. Thus, the control unit 10 can calculate the value of cos θ. Because the size of the circle formed by the test line TL is sufficiently large compared to the size of the crystal grain G, the angle θ takes a value in the range of 0° to 90°. In this case, 0≦cos θ≦1 holds, and therefore the control unit 10 can evaluate that the smaller the calculated value of cos θ, the larger the angle θ.

[0055] The control unit 10 can apply this method when there are three or more intersections between the specific grain boundary SGB and the test line TL. Specifically, the control unit 10 calculates the value of cos θ using equation (1) based on two vectors connecting the center P0 and each of two intersections extracted from the three or more intersections. The control unit 10 can select the intersection related to the combination that provides the smallest calculated value of cos θ, thereby selecting the intersection related to the combination that provides the greatest distance between the intersections. In this way, the control unit 10 can accurately select the intersection related to the combination that provides the greatest distance between the intersections.

[0056] 9(K), the value of cos θ calculated based on the angle θ between the lines P0P1 and P0P3 is smaller than the value of cos θ calculated based on the angle θ between the lines P0P1 and P0P2 and the angle θ between the lines P0P2 and P0P3. Therefore, the control unit 10 selects the intersection points P1 and P3 associated with the combination that provides the smallest value of cos θ.

[0057] By using this method, the control unit 10 determines whether there are multiple identified intersections for each of all the specific crystal grains SG, and if there are multiple identified intersections, selects the intersections related to the combination that connects the intersections to each other at the longest distance. This allows the control unit 10 to appropriately select the intersections between the test line TL and the specific crystal grain boundary SGB.

[0058] As described above, the control unit 10 of the grain size evaluation apparatus 1 executes a grain size evaluation process to perform the grain size evaluation method according to the present disclosure. The control unit 10 executes a preparation step of preparing an evaluation target image 30 in which the grain boundaries GB of each of the multiple crystal grains G that constitute the crystalline structure of the material are displayed (S1). The control unit 10 also executes an assignment step of assigning test lines TL to the evaluation target image 30 (S2). The control unit 10 executes a grain identification step of identifying, as specific crystal grains SG, each of the multiple crystal grains G displayed in the evaluation target image 30 that overlap the test lines TL (S3). The control unit 10 executes an intersection identification step of identifying an intersection where the test line TL intersects with the specific crystal grain boundaries SGB, which are the grain boundaries GB of each of the specific crystal grains SG identified in the grain identification step (S5). If there are multiple intersections identified for one specific crystal grain SG in the intersection identification step, the control unit 10 executes an intersection selection step (S6) to select from the multiple intersections the intersection associated with the combination that provides the longest distance between the intersections.

[0059] Conventionally, when evaluating grain size using the intercept method, an evaluator marks a test line TL on the evaluation target image 30 and then visually identifies the intersections between the test line TL and the specific grain boundary SGB. Visually identifying the intersections is cumbersome. Furthermore, the degree of intersection identification may vary depending on the evaluator. Conventionally, rather than visually identifying the intersections, the intersections between the test line TL and the specific grain boundary SGB may be identified by performing image analysis on the evaluation target image 30 marked with the test line TL. However, depending on the shape of the specific grain SG, three or more intersections may be identified for one specific grain SG. Furthermore, if the specific grain boundary SGB of a specific grain SG is arranged in a similar position to the test line TL, the test line TL and the specific grain boundary SGB may intersect repeatedly. In such cases, an excessive number of intersections may occur for one specific grain SG. In the grain size evaluation method according to this embodiment, each of the crystal grains G arranged on the test line TL is identified as a specific crystal grain SG. Then, the intersections where the specific grain boundaries SGB of each of the specific crystal grains SG intersect with the test line TL are identified. If multiple intersections are identified for one specific crystal grain SG, the intersections associated with the combination that provides the longest distance between the intersections are selected from the multiple intersections. In this way, the control unit 10 executes the grain size evaluation process, thereby automatically selecting an appropriate number of intersections for one specific crystal grain SG without relying on visual inspection by an evaluator.

[0060] In the intersection selection step, the control unit 10 selects the intersections associated with the combination that maximizes the angle between two straight lines connecting each of the two intersections extracted from the multiple intersections identified in the intersection identification step for one specific crystal grain SG and a predetermined point located in a position that does not overlap with the test line TL in the image 30 to be evaluated.

[0061] In this case, the grain size evaluation method can easily select an appropriate combination of intersection points from among a plurality of intersection points for one specific grain SG.

[0062] The test line TL forms a circle, and the predetermined point is set at the center of the circle that forms the test line TL.

[0063] In this case, since the lengths between the predetermined point and the intersections are uniformly determined, the control unit 10 can more easily derive the angles between the two lines connecting the predetermined point and each of the two intersections extracted from the plurality of intersections, and compare the derived angles. Therefore, the control unit 10 can easily select an appropriate combination of intersections from the plurality of intersections for one specific crystal grain SG.

[0064] As a grain identification step, the control unit 10 executes a first extraction step (S11) to extract from the evaluation target image 30 an image showing a grain boundary GB that exists on one side of the test line TL that was assigned to the evaluation target image 30 in the assigning step in a direction intersecting the test line TL. The control unit 10 executes a second extraction step (S12) to extract a first maximum-area image 411, which is an image of the closed region with the largest area among multiple closed regions formed by the images showing the grain boundary GB extracted in the first extraction step. The control unit 10 executes a third extraction step (S13) to extract a one-side image 51 that exists on one side of the test line TL from the first maximum-area image 411 extracted in the second extraction step. The control unit 10 executes a fourth extraction step (S15) to extract from the evaluation target image 30 an image showing a grain boundary GB that exists on the other side opposite to the one side of the test line TL that was assigned to the evaluation target image 30 in the assigning step in a direction intersecting the test line TL. The control unit 10 executes a fifth extraction step of extracting a second maximum-area image 412, which is an image of the closed region having the largest area among the multiple closed regions formed by the image showing the grain boundary GB extracted in the fourth extraction step. The control unit 10 executes a sixth extraction step of extracting an other-side image 52 existing on the other side of the test line TL from the second maximum-area image 412 extracted in the fifth extraction step. The control unit 10 executes a generation step of generating each image of the specific crystal grain SG by combining the one-side image 51 extracted in the third extraction step and the other-side image 52 extracted in the sixth extraction step (S18). The control unit 10 executes an identification step of identifying the specific crystal grain SG by performing a labeling process on each image of the specific crystal grain SG generated in the generation step (S19).

[0065] In this case, each of the crystal grains G arranged on the test line TL is appropriately identified as a specific crystal grain SG. Therefore, an appropriate number of intersections are selected for each specific crystal grain SG. Therefore, the crystal grain size evaluation method can improve the accuracy of the evaluation of the crystal grain size.

[0066] The present disclosure is not limited to the embodiments described above and in the drawings, and for example, the following embodiments are also included in the technical scope of the present disclosure. Furthermore, various modifications other than those described below can be made without departing from the spirit of the present disclosure.

[0067] (1) The first extraction step, second extraction step, third extraction step, fourth extraction step, fifth extraction step, and sixth extraction step do not necessarily have to be performed in this order. For example, the fourth extraction step, fifth extraction step, and sixth extraction step may be performed in this order, and then the first extraction step, second extraction step, and third extraction step may be performed in this order.

[0068] (2) The test line TL may be a straight line. For example, if the test line TL is a horizontally extending straight line, the control unit 10 may extract, in the first extraction step, an image showing the grain boundary GB present on the upper side of the test line TL, which is one side of the direction intersecting the test line TL. Furthermore, in the fourth extraction step, the control unit 10 may extract, in the fourth extraction step, an image showing the grain boundary GB present on the lower side of the test line TL, which is the other side of the direction intersecting the test line TL. Furthermore, in the intersection identification step, the control unit 10 may select an intersection corresponding to a combination in which the angle formed by two straight lines connecting an arbitrary point that does not overlap the test line TL and each of the multiple intersections identified for one specific crystal grain SG is the largest.

[0069] (3) The grain size evaluation system 100 may acquire the evaluation object image 30 from various devices capable of capturing an image of the crystalline structure of a material, instead of the microscope 2.

[0070] (4) It is not necessary for the crystal grain size evaluation device 1 to acquire the evaluation target image 30. For example, the crystal grain size evaluation method according to the present disclosure may be executed by the control unit 10 performing the processes in S2 and subsequent steps on the evaluation target image 30 present on the network.

[0071] (5) The crystal grain identifying step is not limited to being performed as shown in Fig. 3. For example, the control unit 10 may label all of the closed areas formed by line drawings indicating the crystal grain boundaries GB that appear in the evaluation target image 30, and treat each of the labeled closed areas as a crystal grain G. Thereafter, the control unit 10 may execute the crystal grain identifying step by attaching a test line TL to the evaluation target image 30 and identifying the crystal grain G that is arranged overlapping the test line TL as a specific crystal grain SG.

[0072] (6) In the above embodiment, in the intersection point selection step, a predetermined point provided at a position that does not overlap with the test line TL in the generated image 55 is set as the center P0 of the circle that constitutes the test line TL. In addition, the predetermined point provided at a position that does not overlap with the test line TL in the generated image 55 is not limited to the center P0 of the circle that constitutes the test line TL, and may be a point provided at any position in the generated image 55.

[0073] (7) In the above embodiment, when three or more intersections are identified for one specific crystal grain SG in the intersection selection step, the control unit 10 selects, from among the plurality of intersections, an intersection associated with a combination that provides the longest distance between the intersections. For example, when a predetermined number of intersections (three or more) are identified for one specific crystal grain SG, the control unit 10 may select, from among the plurality of intersections, an intersection associated with a combination that provides the longest distance between the intersections.

[0074] (8) The CPU included in the control unit 10 functions as a processor that executes the grain size evaluation process by expanding a program stored in the HDD included in the storage unit 11 onto the RAM included in the storage unit 11. A general-purpose processor may be used as the CPU. A microcomputer, ASIC, FPGA, etc. may be used as a processor instead of the CPU. When the microscope 2 functions as the grain size evaluation device 1, the CPU of the control unit included in the microscope 2 functions as a processor that executes the grain size evaluation process by expanding a program stored in a storage unit such as a ROM included in the microscope 2 onto the RAM included in the microscope 2. As above, a microcomputer, ASIC, FPGA, etc. may be used as a processor instead of the CPU.

[0075] (9) The program for executing the grain size evaluation process may be downloaded, for example, from a server connected to a network (not shown) via the external device connection unit 18 and stored in the storage unit 11. In this case, the program for executing the grain size evaluation process may be stored in a non-transitory storage medium such as an HDD provided in the server.

[0076] (10) The grain size evaluation process may be distributed among multiple electronic devices (i.e., multiple CPUs). For example, part of the grain size evaluation process may be executed by a control unit included in the microscope 2 or another server connected to the grain size evaluation device 1 via a network (not shown). [Explanation of symbols]

[0077] 1. Grain size evaluation device 2. Microscope 10 Control Unit 100 Grain Size Evaluation System G grain GB grain boundary SG Specific crystal grain SGB ​​specific grain boundary TL test line

Claims

1. a preparation step of preparing an evaluation target image in which each grain boundary of a plurality of crystal grains constituting a crystalline structure of a material is displayed; an applying step of applying a test line formed of a straight line or a curved line to the image to be evaluated; a crystal grain identifying step of identifying, as a specific crystal grain, each of the crystal grains that overlaps the test line from among the plurality of crystal grains displayed in the evaluation target image; an intersection point identifying step of identifying an intersection point where the grain boundary of each of the specific crystal grains identified in the crystal grain identifying step intersects with the test line; an intersection selecting step of selecting, from among the plurality of intersections, the intersections associated with a combination that connects the intersections to each other by the greatest distance when a plurality of intersections are identified for one of the specific crystal grains in the intersection identifying step; A method for evaluating grain size, comprising:

2. 2. The crystal grain size evaluation method according to claim 1, wherein in the intersection selection step, the intersections corresponding to the combination in which the angle formed by two straight lines connecting each of the two intersections extracted from the plurality of intersections identified in the intersection identification step for one specific crystal grain and a predetermined point located in a position that does not overlap with the test line in the image to be evaluated is maximized are selected.

3. 3. The method for evaluating crystal grain size according to claim 2, wherein the test line forms a circle, and the predetermined point is set at the center of the circle forming the test line.

4. The crystal grain identification step includes: a first extraction step of extracting from the evaluation target image an image showing the grain boundary present on one side of the test line in a direction intersecting the test line assigned to the evaluation target image in the assignment step; a second extraction step of extracting a first largest-area image, which is an image of the closed region having the largest area among a plurality of closed regions formed by the image showing the grain boundary extracted in the first extraction step; a third extraction step of extracting a one-side image existing on one side of the test line from the first maximum area image extracted in the second extraction step; a fourth extraction step of extracting from the evaluation target image an image showing the grain boundary present on the other side of the test line assigned to the evaluation target image in the assigning step, the other side being opposite to the one side in a direction intersecting the test line; a fifth extraction step of extracting a second largest-area image, which is an image of the closed region having the largest area among a plurality of closed regions formed by the images showing the grain boundaries extracted in the fourth extraction step; a sixth extraction step of extracting an other-side image that exists on the other side of the test line from the second maximum area image extracted in the fifth extraction step; a generating step of generating an image of each of the specific crystal grains by combining the one-side image extracted in the third extraction step and the other-side image extracted in the sixth extraction step; a step of identifying the specific crystal grains by performing a labeling process on each image of the specific crystal grains generated by the generating step; The method for evaluating crystal grain size according to claim 1 , comprising:

Images