Material characteristic prediction device, material characteristic prediction system, material characteristic prediction method, and program

Abstract

The present invention provides an information processing device capable of more easily acquiring material characteristics in a plurality of directions from one sample. A material characteristic prediction device (information processing device 20) comprises: an acquisition unit (control unit 21) that acquires a Talbot image of one sample; a setting unit (control unit 21) that sets an analysis region in the Talbot image; a prediction unit (control unit 21) that predicts material characteristics of the analysis region from feature amounts extracted on the basis of the Talbot image; and an output unit (control unit 21) that outputs the material characteristics of the analysis region. The prediction unit predicts material characteristics in at least two directions.

Description

Material property prediction device, material property prediction system, material property prediction method and program

[0001] The present disclosure relates to a material property prediction device, a material property prediction system, a material property prediction method, and a program.

[0002] Conventionally, in the development of resin materials, tensile strength tests, bending tests, etc. are carried out using test samples such as dumbbell test pieces or flat plates cut out from resin molded products.

[0003] Patent Document 1 describes a technique for photographing a resin molded product using the Talbot effect, dividing the Talbot image into meshes, and illustrating the orientation information of fibers and other components contained in the resin molded product for each mesh. The Talbot effect refers to a phenomenon in which, when coherent light passes through a first grating with slits at regular intervals, a grating image is formed at regular intervals in the direction of light propagation. This grating image is called a self-image. A Talbot interferometer places a second grating at a position where the self-image is formed and measures the moiré fringes that are generated by slightly shifting the second grating. Since placing an object in front of the second grating disrupts the moiré, Talbot photography allows a display image (Talbot image) of the object to be obtained by placing the object in front of or behind the first grating, irradiating it with X-rays, and then computing the resulting moiré fringe image.

[0004] Japanese Patent Application Laid-Open No. 2021-089195

[0005] Resin materials have anisotropy in material properties, such as strength. To measure a material with anisotropy in material properties, multiple samples made with the same material and molding conditions must be prepared, and test samples must be cut out from the same measurement location on each sample. For example, to measure strength in the 0-degree and 90-degree directions, two similar samples must be created. In this case, measurement cannot be performed on the same sample, resulting in poor accuracy and requiring a lot of labor. Patent Document 1 (Japanese Patent Laid-Open No. 2004-222211) can obtain orientation information for each region separated in an image, but it cannot predict material properties, so test samples must be cut out from the resin molded product.

[0006] Therefore, an object of the present disclosure is to more easily obtain material properties in multiple directions from a single sample.

[0007] In order to solve the above problems, the material property prediction device of the present disclosure includes an acquisition unit that acquires a Talbot image of one sample; a setting unit that sets an analysis region in the Talbot image; a prediction unit that predicts material properties of the analysis region from feature quantities extracted based on the Talbot image; and an output unit that outputs the material properties of the analysis region, wherein the prediction unit predicts material properties in at least two directions.

[0008] In addition, the material property prediction system of the present disclosure includes an acquisition unit that acquires a Talbot image of one sample; a setting unit that sets an analysis region in the Talbot image; a prediction unit that predicts material properties of the analysis region from feature quantities extracted based on the Talbot image; and an output unit that outputs material properties of the analysis region, wherein the prediction unit predicts material properties in at least two directions.

[0009] Furthermore, the material property prediction method of the present disclosure is a method in which an information processing device predicts material properties from a Talbot image of a sample, and includes: an acquisition step of acquiring a Talbot image; a setting step of setting an analysis region in the Talbot image; a prediction step of predicting material properties of the analysis region from feature quantities extracted based on the Talbot image; and an output step of outputting material properties of the analysis region, wherein the prediction step predicts material properties in at least two directions.

[0010] In addition, the program disclosed herein causes a computer of a material property prediction device that predicts material properties from a Talbot image of a sample to function as an acquisition unit that acquires a Talbot image, a setting unit that sets an analysis region in the Talbot image, a prediction unit that predicts material properties of the analysis region from feature quantities extracted based on the Talbot image, and an output unit that outputs material properties of the analysis region, and the prediction unit predicts material properties in at least two directions.

[0011] According to the present disclosure, material properties in multiple directions can be more easily obtained from a single sample.

[0012] 1 is a schematic diagram showing an overall image of an analysis system; FIG. 2 is a block diagram showing a schematic configuration of an information processing device; FIG. 3 is a flowchart showing material property prediction processing; FIG. 4 is an example of an analysis condition input screen; FIG. 5 is an example of an analysis condition input screen; FIG. 6 is an example of an analysis result display screen; FIG. 7 is an example of an analysis result display screen;

[0013] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. However, although the embodiments described below have various limitations that are technically preferable for implementing the present disclosure, the technical scope of the present disclosure is not limited to the following embodiments and illustrated examples.

[0014] In this embodiment, a material property prediction method is described using an information processing device 20 (material property prediction device) that predicts and outputs the material properties of a subject 100 using a Talbot image obtained by photographing the subject 100 using an X-ray Talbot imaging device 10.

[0015] [Subject] The subject 100 in this embodiment is specifically fiber reinforced plastic in general, resin in general, resin with added filler in general, foam material such as sponge, cast material such as aluminum die cast, and the like. More specifically, known materials include CFRP (Carbon-Fiber-Reinforced Plastics), CFRTP (Carbon Fiber Reinforced Thermo Plastics), and FRP (Fiber-Reinforced Plastics) such as GFRP (Glass-Fiber-Reinforced Plastics), which use carbon fiber or glass fiber as reinforcing fibers, and CMC (Ceramic Matrix Composites), which use ceramic fiber as a reinforcing material. In a broad sense, the term may also include composite materials made of multiple types of wood, such as plywood. In addition, composite materials that do not contain fibers, such as MMC (Metal Matrix Composites) concrete and reinforced concrete, may also be included. Resins used in the materials include, but are not limited to, general-purpose plastics, engineering plastics, and super engineering plastics. Resins are often used as resin composite materials to which fillers with micro- or nano-sized structures are added to impart specific properties (such as strength), and are used as plastic molded products. Fillers include organic materials, inorganic materials, magnetic materials, and metallic materials. For example, when strength and rigidity are required for a plastic molded product, a composite material such as PPS (polyphenylene sulfide), POM (polyacetal, polyoxymethylene), PA (polyamide), PC (polycarbonate), or PP (polypropylene) is used as the resin, and aramid fiber, talc, or cellulose fiber is used as the filler.Furthermore, when the plastic molded product is a plastic mag, a composite material such as nylon as the resin and strontium ferrite or samarium cobalt as the filler may be used.

[0016] 1 includes an X-ray Talbot imaging device 10, a controller 19, and an information processing device 20. The X-ray Talbot imaging device 10 is connected to the information processing device 20 via the controller 19 and a communication network N. The communication network N is a LAN (Local Area Network), a WAN (Wide Area Network), the Internet, or the like.

[0017] [X-ray Talbot Imaging Device] In this embodiment, the X-ray Talbot imaging device 10 uses a Talbot-Lau interferometer equipped with a source grating 12. Note that an X-ray Talbot imaging device using a Talbot interferometer equipped only with a first grating 14 and a second grating 15 without the source grating 12 can also be used.

[0018] 1 is a schematic diagram showing an overall view of an X-ray Talbot imaging device 10. The X-ray Talbot imaging device 10 according to this embodiment includes an X-ray generator 11, a source grating 12, a subject table 13, a first grating 14, a second grating 15, an X-ray detector 16, a support 17, and a base 18. The grating directions of the source grating 12, the first grating 14, and the second grating 15 are the same.

[0019] With this X-ray Talbot imaging device 10, a moiré image Mo of a subject 100 positioned at a predetermined position relative to the subject table 13 can be captured using a method based on the principles of fringe scanning, and the moiré image Mo can be analyzed using a Fourier transform to reconstruct at least three types of images (two-dimensional images) (referred to as reconstructed images). These three types of images include an absorption image (same as a normal X-ray absorption image) that visualizes the average component of the moiré fringes in the moiré image Mo, a differential phase image that visualizes the phase information of the moiré fringes, and a small-angle scattering image that visualizes the visibility of the moiré fringes. The fringe scanning method is a method of obtaining a high-resolution reconstructed image by capturing moiré images Mo M times by moving one of a plurality of gratings in the slit period direction by 1 / M of the grating slit period (M is a positive integer, M > 2 for absorption images, and M > 3 for differential phase images and small-angle scattering images). The Fourier transform method is a method in which, in the presence of a subject, one moiré image Mo is captured using an X-ray Talbot imaging device, and then, in image processing, the moiré image Mo is subjected to a Fourier transform or the like to reconstruct and generate an image such as a differential phase image.

[0020] Even more types of images can be generated by recombining the three types of reconstructed images. For example, small-angle scattering images taken at multiple (three or more) grating opposing angles are used, and after aligning each image, fitting is performed for each pixel using a sine wave to extract fitting parameters. The sine wave graph has the relative angle α between the sample and the grating on the horizontal axis and the small-angle scattering signal value of a given pixel on the vertical axis. The fitting parameters are the amplitude, average, and phase of the sine wave. An image showing the amplitude value for each pixel is called the orientation image, an image showing the average value for each pixel is called the scattering intensity image, and an image showing the phase for each pixel is called the orientation angle image. Note that the fitting method is not limited to sine waves. Hereinafter, the images generated by recombining the reconstructed images (orientation image, scattering intensity image, and orientation angle image) are collectively referred to as the orientation analysis image.

[0021] It is also possible to perform image processing such as filtering, clarification, and contour extraction on the reconstructed image and the orientation analysis image, as well as image processing for combining two or more types of images. Hereinafter, such an image that has undergone image processing will be referred to as a secondary image.

[0022] Hereinafter, the term "Talbot photography" will refer to not only the photography of the moiré image Mo, but also the generation of the reconstructed image, orientation analysis image, and secondary image described above. Also, hereinafter, the reconstructed image, orientation analysis image, and secondary image will be collectively referred to as "Talbot images."

[0023] The configuration of other parts of the X-ray Talbot imaging device 10 according to this embodiment will be described below. The controller 19 is a device that performs overall control of the X-ray Talbot imaging device 10. That is, for example, the controller 19 is connected to the X-ray generator 11, and is capable of setting the tube voltage, tube current, irradiation time, etc., for the X-ray source 11a.

[0024] [Regarding the Information Processing Device] In this embodiment, a general-purpose computer device (control PC) is used as the information processing device 20 that executes various processes. However, this is not limited to this, and some of the functions of the information processing device 20 may be provided on a network so that each process can be executed by exchanging data via communication. As shown in Figure 2, the information processing device 20 includes a control unit 21, an operation unit 22, a communication unit 23, a storage unit 24, and a display unit 25.

[0025] The control unit 21 is configured with a CPU (Central Processing Unit), a RAM (Random Access Memory), etc. The CPU of the control unit 21 reads out various programs stored in the storage unit 24, loads them into the RAM, and executes various processes (e.g., material property prediction process described below) in accordance with the loaded programs, thereby controlling the operation of each unit of the information processing device 20.

[0026] The control unit 21 functions as an acquisition unit that acquires Talbot images. Specifically, the control unit 21 acquires Talbot images from the X-ray Talbot imaging device 10 via the communication unit 23, which will be described later. The control unit 21 may also acquire Talbot images stored in the storage unit 24.

[0027] The control unit 21 functions as a setting unit that sets an analysis region in the Talbot image. Specifically, the control unit 21 sets the analysis region in the Talbot image based on a signal input by a user using an operation unit 22, which will be described later.

[0028] The control unit 21 functions as a prediction unit that predicts material properties of the analysis region from feature quantities extracted based on the Talbot image. The feature quantities extracted based on the Talbot image include, for example, statistical values ​​of pixel values ​​of the Talbot image, such as the average value, deviation value, maximum value, minimum value, coefficient of variation, and percentile value. The material properties are material properties of the subject 100, including, for example, tensile strength, modulus of elasticity, bending strength, bending modulus of elasticity, Charpy strength, warpage, and coefficient of linear expansion. The control unit 21 predicts the material properties using an analysis model such as a machine learning model that uses the feature quantities as input information and material properties as output information.

[0029] The control unit 21 functions as an output unit that outputs material properties of the analysis region. As described below, the control unit 21 outputs material properties in at least two directions as an output unit. Specifically, the control unit 21 causes the display unit 25, described below, to display the material properties of the analysis region.

[0030] The operation unit 22 is a keyboard equipped with cursor keys, numeric input keys, various function keys, etc., a pointing device such as a mouse, a touch panel laminated on the surface of the display unit 25, etc. The operation unit 22 is configured to be operable by an operator, and outputs various signals to the control unit 21 based on operations performed by the operator.

[0031] The communication unit 23 is capable of transmitting and receiving various signals and various data to and from other devices connected via the communication network N.

[0032] The storage unit 24 is configured with a non-volatile semiconductor memory, a hard disk, etc., and stores various programs executed by the control unit 21, parameters required for executing the programs, various data, etc. The storage unit 24 stores the above-mentioned analytical model and Talbot image.

[0033] The display unit 25 is configured with a monitor such as an LCD (Liquid Crystal Display), and displays various screens and the like in accordance with instructions of a display signal input from the control unit 21 .

[0034] [Material Property Prediction Processing] The material property prediction processing in the information processing device 20 will be described with reference to Fig. 3. The material property prediction processing is processing for predicting material properties of an analysis region set in a Talbot image obtained by photographing the subject 100. The material property prediction processing starts when the control unit 21 receives a signal to start the material property prediction processing input by the user using the operation unit 22. It is assumed that the Talbot image has been sent from the X-ray Talbot imaging device 10 to the information processing device 20 and has already been stored in the storage unit 24 before the material property prediction processing starts. It is also assumed that pre-learning of the analysis model has been completed.

[0035] First, the control unit 21 acquires a Talbot image from the storage unit 24 (step S1; acquisition step).

[0036] Next, the control unit 21 displays an analysis condition input screen (FIGS. 4, 5, and 6) described later on the display unit 25 (step S2). The analysis conditions are conditions for the control unit 21 to analyze the Talbot image.

[0037] Here, the analysis condition input screen 2510 will be described with reference to FIG. 4 . In FIG. 4 , the analysis conditions are “shape,” “area / size,” “direction,” and “index.” Area 2511 is an area for selecting the shape of the analysis area. Specific examples of the shape of the analysis area include a rectangle, a trapezoid, a circle, and a triangle. When the shape of the analysis area is selected in area 2511, a frame indicating the analysis area is displayed in area 2512 in the selected shape. Area 2512 is an area for selecting the position and size of the analysis area. For example, the position and size of the analysis area can be selected by moving the frame indicating the analysis area on the displayed Talbot image. Note that, as described above, a Talbot image includes multiple images, and the control unit 21 displays one or more Talbot images. Furthermore, multiple analysis areas may be selected by adding frames indicating the analysis areas. Area 2513 is an area for selecting the direction of the material properties to be predicted. For example, the upward direction of the Talbot image displayed in area 2512 is set as the reference (0°), and the direction is selected by a right rotation degree (45°, 90°, ...) relative to the reference. In this case, it is sufficient to perform the analysis only in the selected direction. Area 2514 is an area for selecting an index representing the material properties to be predicted. For example, the index is selected from indexes representing various material properties such as tensile strength, linear expansion coefficient, ...

[0038] Here, the analysis condition input screen 2520 will be described with reference to FIG. 5. In FIG. 5, the analysis conditions are "shape," "area / size," "conditions (statistical values)," and "index." Area 2521 is the same as area 2511. Area 2522 is the same as area 2512. Area 2523 is an area for selecting statistical values ​​(maximum value, minimum value, etc.) of predicted material properties. In this case, the analysis itself is performed in all directions. "All directions" refers to, for example, directions of 360° at predetermined angles relative to a reference direction. Area 2524 is the same as area 2513.

[0039] Here, the analysis condition input screen 2530 will be described using FIG. 6 . In FIG. 6 , the analysis conditions are “index” and “size.” Area 2531 is similar to area 2513. Area 2532 is an area for inputting the size of the analysis area. In the case of FIG. 6 , the shape and position of the analysis area are automatically set by the control unit 21. The control unit 21 automatically extracts the contour of the subject 100 in the Talbot image and determines the shape of the analysis area based on the contour. For example, the shape of the analysis area is a rectangle, trapezoid, circle, triangle, etc. The control unit 21 also automatically sets the analysis area on the Talbot image by arranging analysis areas having the shape of the analysis area with the input size. In the case of FIG. 6 , the direction of the predicted material properties is also automatically set by the control unit 21. For example, the upward direction of the Talbot image displayed in area 2512 is set as the reference (0°), and the direction is selected by a right rotation degree (45°, 90°, ...) relative to the reference. In this case, analysis is performed only in the selected direction. It should be noted that statistical values ​​of material properties (maximum values, minimum values, etc.) may be automatically set instead of automatically setting the direction by the control unit 21. In this case, the analysis itself is performed for all directions.

[0040] Next, the control unit 21 acquires the analysis conditions input by the user using the operation unit 22 (step S3; setting step).

[0041] Next, the control unit 21 predicts the material properties (step S4; prediction step). Specifically, the control unit 21 inputs the feature amounts of the analysis region in the Talbot image into the analysis model, thereby predicting the material properties of the analysis region based on the analysis conditions.

[0042] Next, the control unit 21 outputs the material properties of the analysis region (step S5; output step). Specifically, the control unit 21 displays an analysis result display screen (FIGS. 7, 8, and 9) described later on the display unit 25, and ends the material property prediction process.

[0043] Here, the analysis result display screen 2540 will be described with reference to Fig. 7. The analysis result display screen 2540 shown in Fig. 7 is a screen that displays the analysis results corresponding to the analysis condition input screen 2510 shown in Fig. 4. In the analysis region selected in region 2512, the indicators (80 MPa, 75 MPa) selected in region 2514 for the direction selected in region 2513 are displayed by the direction and size of the arrow.

[0044] Here, the analysis result display screen 2550 will be described with reference to Fig. 8. The analysis result display screen 2550 shown in Fig. 8 is a screen that displays the analysis results corresponding to the analysis condition input screen 2520 shown in Fig. 5. In the analysis region selected in region 2522, the indicators (80 MPa, 65 MPa) selected in region 2514 of the statistical values ​​(maximum value, minimum value) selected in region 2523 are displayed by the direction and size of the arrow.

[0045] Here, the analysis result display screen 2560 will be described with reference to Fig. 9. The analysis result display screen 2560 shown in Fig. 9 is a screen that displays the analysis results corresponding to the analysis condition input screen 2530 shown in Fig. 6. In each automatically set analysis region (rectangular with the input size), the indicator selected in the region 2531 (compared to 50 MPa and 45 MPa) in the automatically set direction (0° and 45°) is displayed using an arrow and an ellipse. Note that the shape is not limited to an ellipse, and a polygon or the like may also be used.

[0046] <Others> The X-ray Talbot imaging device 10 may be configured to perform Talbot CT imaging by providing an imaging jig (not shown) on the subject table 13 to fix the subject 100 in a predetermined orientation and rotating the subject 100 three-dimensionally. While the above description has been given of Talbot imaging, which generates various two-dimensional images using a Talbot interferometer and a Talbot-Lau interferometer, the following description will be given of Talbot CT imaging, which generates various three-dimensional images using a Talbot-Lau interferometer. Talbot CT imaging is a three-dimensional extension of Talbot imaging, which generates various two-dimensional images. In Talbot CT imaging, the subject is rotated around the CT rotation axis by a predetermined angle (e.g., 1°) to acquire Moiré fringe images covering 180° or 360°, and processing is performed based on a fringe scanning method or a Fourier transform method to generate two-dimensional projection images (absorption image, small-angle scattering image, differential phase image). Next, the signal value of each voxel is calculated using the two-dimensional projection image to generate a three-dimensional CT image (absorption tomographic image, small-angle scattering tomographic image, phase tomographic image) corresponding to the two-dimensional projection image. Furthermore, in Talbot CT imaging, three-dimensional orientation analysis images (orientation degree tomographic image, orientation angle tomographic image, scattering intensity image) are generated by performing CT reconstruction and image processing using small-angle scattering images captured in multiple directions relative to the CT rotation axis. In this case, the Talbot image further includes a three-dimensional CT image and a three-dimensional orientation analysis image in addition to the Talbot image described above. It is also possible to perform image processing such as filtering, clarification, and contour extraction on the three-dimensional CT image and the three-dimensional orientation analysis image, as well as image processing for combining two or more types of images.

[0047] <Effects> As described above, the material property prediction device (information processing device 20) includes an acquisition unit (control unit 21) that acquires a Talbot image of one sample, a setting unit (control unit 21) that sets an analysis region in the Talbot image, a prediction unit (control unit 21) that predicts material properties of the analysis region from feature quantities extracted based on the Talbot image, and an output unit (control unit 21) that outputs material properties of the analysis region, and the prediction unit predicts material properties in at least two directions. This makes it easier to obtain material properties in multiple directions from one sample.

[0048] The material property prediction system (analysis system 1) includes an acquisition unit (control unit 21) that acquires a Talbot image of one sample, a setting unit (control unit 21) that sets an analysis region in the Talbot image, a prediction unit (control unit 21) that predicts material properties in the analysis region from feature quantities extracted based on the Talbot image, and an output unit (control unit 21) that outputs material properties in the analysis region, where the prediction unit predicts material properties in at least two directions, thereby making it possible to predict and output material properties at any location in a resin molded product.

[0049] The material property prediction method (material property prediction process) is a method in which an information processing device 20 predicts material properties from a Talbot image of a sample, and includes an acquisition step (step S1) of acquiring a Talbot image, a setting step (step S3) of setting an analysis region in the Talbot image, a prediction step (step S4) of predicting material properties in the analysis region from feature quantities extracted based on the Talbot image, and an output step (step S5) of outputting material properties in the analysis region, where the prediction step predicts material properties in at least two directions. This makes it easier to obtain material properties in multiple directions from a single sample.

[0050] The program also causes a computer of a material property prediction device (information processing device 20) that predicts material properties from a Talbot image of a sample to function as an acquisition unit (control unit 21) that acquires the Talbot image, a setting unit (control unit 21) that sets an analysis region in the Talbot image, a prediction unit (control unit 21) that predicts material properties in the analysis region from feature quantities extracted based on the Talbot image, and an output unit (control unit 21) that outputs material properties in the analysis region, and the prediction unit predicts material properties in at least two directions. This makes it easier to obtain material properties in multiple directions from a single sample.

[0051] The embodiment of the present disclosure has been described above, but the description of the above embodiment is a preferred example according to the present disclosure, and the present disclosure is not limited to this.

[0052] In the above description, examples have been given in which a hard disk or a semiconductor nonvolatile memory is used as a computer-readable medium for the program according to the present disclosure, but the present disclosure is not limited to these examples. Other computer-readable media may also be portable recording media such as a CD-ROM.

[0053] In addition, the detailed configuration and operation of each device can be modified as appropriate without departing from the spirit of the invention.

[0054] The present disclosure can be used in a material property prediction device, a material property prediction system, a material property prediction method, and a program.

[0055] REFERENCE SIGNS LIST 1 Analysis system (material property prediction system) 10 X-ray Talbot imaging device 11 X-ray generator 19 Controller 20 Information processing device (material property prediction device) 21 Control unit (acquisition unit, setting unit, prediction unit, output unit) 22 Operation unit 23 Communication unit 24 Storage unit 25 Display unit 100 Subject

Claims

1. A material property prediction device comprising: an acquisition unit that acquires a Talbot image of a sample; a setting unit that sets an analysis area in the Talbot image; a prediction unit that predicts material properties of the analysis area from feature quantities extracted based on the Talbot image; and an output unit that outputs the material properties of the analysis area, wherein the prediction unit predicts material properties in at least two directions.

2. The material property prediction device according to claim 1, wherein the prediction unit predicts the material property using an analytical model that uses feature quantities as input information and material properties as output information.

3. The material property prediction device according to claim 1, wherein the material properties include tensile strength, modulus of elasticity, flexural strength, flexural modulus of elasticity, Charpy strength, warpage, and coefficient of linear expansion.

4. The material property prediction device according to claim 1, wherein the output unit displays the material property of the analysis region superimposed on the Talbot image.

5. The material property prediction device according to claim 4, wherein the setting unit sets a plurality of the analysis regions, the prediction unit predicts the material properties for each of the analysis regions, and the output unit visually displays the direction and magnitude of the material properties in each of the analysis regions of the Talbot image.

6. A material property prediction system comprising: an acquisition unit that acquires a Talbot image of a sample; a setting unit that sets an analysis area in the Talbot image; a prediction unit that predicts material properties of the analysis area from feature quantities extracted based on the Talbot image; and an output unit that outputs material properties of the analysis area, wherein the prediction unit predicts material properties in at least two directions.

7. A method for an information processing device to predict material properties from a Talbot image of a sample, comprising: an acquisition step of acquiring a Talbot image; a setting step of setting an analysis region in the Talbot image; a prediction step of predicting material properties of the analysis region from feature quantities extracted based on the Talbot image; and an output step of outputting material properties of the analysis region, wherein the prediction step predicts material properties in at least two directions.

8. A program for making a computer of a material property prediction device that predicts material properties from a Talbot image of a sample function as: an acquisition unit that acquires a Talbot image; a setting unit that sets an analysis region in the Talbot image; a prediction unit that predicts material properties of the analysis region from feature quantities extracted based on the Talbot image; and an output unit that outputs material properties of the analysis region, wherein the prediction unit predicts material properties in at least two directions.

Images