Image analyzer

Abstract

To easily visualize features of an image representing a structure of a material.SOLUTION: An image analyzer 4 includes: a transformer 41 that performs a Fourier transform on each of a plurality of original images representing a structure of a material and acquires a plurality of power spectra; an analyzer 42 that performs principal component analysis on the plurality of power spectra and acquires principal components and principal component scores; and a reconstructor 43 that reconstructs an image on the basis of the principal components and principal component scores and outputs a feature image representing features of the original image. The reconstructor 43 generates a difference image between a changed image reconstructed by changing the principal component scores and a reference image, and outputs the generated difference image as the feature image. The principal component score to be changed may be determined using a trained regression / classification model constructed by machine learning.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an image analysis device. [Background technology]

[0002] Patent Document 1 discloses a method for searching for new substances, in which the relationship between structural information and physical property information of known substances is learned by machine learning, and target physical properties are input to the obtained trained model, thereby determining at least one candidate substance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2017-91526 A Summary of the Invention [Problem to be solved by the invention]

[0004] Feature quantities are extracted from images of materials, and it is determined whether or not they are important features that affect the performance of the material, and the results are used in the development of new materials, etc. In this case, it is desirable to visualize which areas in the image change due to changes in the feature quantities.

[0005] The present invention has been made in view of the above, and has an object to easily visualize features of an image that represents the structure of a material. [Means for solving the problem]

[0006] In order to solve the above problems, the image analysis device of the present invention includes a transformation unit that performs a Fourier transform on each of a plurality of original images representing a material structure to obtain a plurality of power spectra, an analysis unit that performs principal component analysis on the plurality of power spectra to obtain principal components and principal component scores, and a reconstruction unit that reconstructs an image based on the principal components and the principal component scores, and outputs a feature image representing features of the original image, wherein the reconstruction unit generates a difference image between a reference image and a modified image reconstructed by changing the principal component scores, and outputs the generated difference image as the feature image. Effect of the Invention

[0007] According to the present invention, image features that represent the structure of a material can be easily visualized. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram showing the configuration of an information processing system including an image analyzing device according to a first embodiment. [Diagram 2] FIG. 2 is a diagram for explaining the function of the image analysis device shown in FIG. 1. [Diagram 3] 2 is a flowchart showing a process performed by the image analyzing device shown in FIG. 1. [Figure 4] FIG. 11 is a diagram for explaining an image analysis device according to a second embodiment. [Diagram 5] FIG. 11 is a diagram for explaining an image analysis device according to a third embodiment. [Figure 6] FIG. 13 is a diagram for explaining an image analysis device according to a fourth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the embodiments, the same reference numerals denote the same configurations or functions, and unless otherwise specified, the description thereof will be omitted.

[0010] [First embodiment] An image analyzing device 4 according to the first embodiment will be described with reference to Fig. 1 to Fig. 3. Fig. 1 is a diagram showing the configuration of an information processing system 1 including an image analyzing device 4 according to the first embodiment. Fig. 2 is a diagram showing the function of the image analyzing device 4 shown in Fig. 1.

[0011] The information processing system 1 is a system that analyzes an image input by a user using an image analysis device 4 and displays the analysis result to the user. The information processing system 1 includes a user terminal 2 and a server 3 equipped with the image analysis device 4. The user terminal 2 and the server 3 are connected to each other via a network N so as to be able to communicate with each other.

[0012] The user terminal 2 includes a processing device 21 and a display device 22. The processing device 21 includes a processor and a memory, and the processor executes a program to realize the functions of the user terminal 2. The display device 22 is composed of a display, and displays the processing results of the processing device 21.

[0013] The user terminal 2 accepts an image input by a user and transmits it to the image analysis device 4 of the server 3. The user terminal 2 receives the analysis results of the image analysis device 4 from the server 3, displays them, and presents them to the user.

[0014] The image analysis device 4 extracts features from an image input by a user, and performs a process of visualizing which area in the image changes due to a change in the features. The image to be analyzed by the image analysis device 4 (hereinafter also referred to as the "original image") is, for example, an image that represents the structure of a material. The image that represents the structure of a material is, for example, an image of the material captured using a microscope, or an image that simulates the captured image. The material is, for example, a metal material, a resin material, or a coating material used in a vehicle. For example, a metal material has a different surface structure depending on the type of metal to be combined and its composition. It is not easy for a user to grasp the difference in surface structure from an image. Therefore, the image analysis device 4 extracts features from multiple input images, and visualizes which area in the image changes due to a change in the features. In this way, the image analysis device 4 can make it easier for a user to grasp important features that affect the performance of a material. The performance of a material is, for example, battery performance, magnet performance, rigidity, thermoplasticity, tensile performance, mechanical durability, etc.

[0015] The image analysis device 4 includes a processing device 40 and a storage device 45. The storage device 45 is configured with a storage that stores images received from the user terminal 2 and analysis results of the image analysis device 4. The processing device 40 includes a processor and a memory, and each function is realized by the processor executing a program. As the respective functions, the processing device 40 includes a conversion unit 41, an analysis unit 42, and a reconstruction unit 43.

[0016] The conversion unit 41 performs a Fourier transform on each of the multiple original images to obtain an amplitude spectrum and a phase spectrum of each original image. The conversion unit 41 obtains a two-dimensional power spectrum by squaring the obtained amplitude spectrum. As shown in FIG. 2, the two-dimensional power spectrum is expressed in a frequency space such that the power spectrum spreads from the center, with the center as the origin. The conversion unit 41 integrates the two-dimensional power spectrum in the azimuth direction to make it one-dimensional, and obtains a one-dimensional power spectrum. The one-dimensional power spectrum is represented by a graph with the horizontal axis representing frequency and the vertical axis representing power value (intensity). In FIG. 2, the one-dimensional power spectrum of the original image is represented as "Xorg". One one-dimensional power spectrum is obtained from one image.

[0017] In this way, the conversion unit 41 performs a Fourier transform on each of the multiple original images to obtain multiple one-dimensional power spectra. The image analysis device 4 can reduce the number of dimensions of the data to be analyzed by obtaining periodic structural information derived from the size, shape, and arrangement of the structures in the original images as a one-dimensional power spectrum.

[0018] The analysis unit 42 performs principal component analysis on the multiple one-dimensional power spectra acquired by the conversion unit 41 to acquire principal components and principal component scores. The principal components are expressed as vectors having, for example, a first principal component PC1, a second principal component PC2, a third principal component PC3, ... as elements. The principal components indicate spectral components with large variance in a data group of multiple one-dimensional power spectra. The principal component scores indicate the proportion of each principal component included.

[0019] The reconstruction unit 43 reconstructs an image based on the principal components and principal component scores acquired by the analysis unit 42, and outputs a feature image representing the features of the original image. At this time, the reconstruction unit 43 generates an image reconstructed by changing the principal component scores (hereinafter also referred to as a "post-modification image"). The reconstruction unit 43 generates a difference image between the generated post-modification image and a reference image. The reconstruction unit 43 outputs the generated difference image as a feature image. The reference image is, for example, the original image. Alternatively, the reference image is an image reconstructed without changing the principal component scores (hereinafter also referred to as a "pre-modification image"). The reconstruction unit 43 of this embodiment employs the pre-modification image as the reference image.

[0020] The changed principal component score values ​​are input by the user to the user terminal 2, and the user terminal 2 transmits them to the server 3, whereby they are set in the reconstruction unit 43. The reconstruction unit 43 changes the principal component scores according to the set values. The reconstruction unit 43 reconstructs a one-dimensional power spectrum based on the principal components acquired by the analysis unit 42 and the changed principal component scores. In FIG. 2, the one-dimensional power spectrum reconstructed by changing the principal component scores is represented as "Xrec_changed".

[0021] The reconstruction unit 43 calculates a first ratio (Xrec_changed / Xorg) indicating a ratio of change in the one-dimensional power spectrum reconstructed by changing the principal component scores to the one-dimensional power spectrum of the original image. The reconstruction unit 43 modulates the amplitude spectrum of the original image based on the calculated first ratio. For example, the reconstruction unit 43 creates a mask for modulating the two-dimensional power spectrum of the original image according to the calculated first ratio. The reconstruction unit 43 modulates the amplitude spectrum of the original image according to the first ratio by multiplying the two-dimensional power spectrum by the created mask. The reconstruction unit 43 generates a modified image by performing an inverse Fourier transform on the amplitude spectrum and phase spectrum modulated based on the first ratio.

[0022] Moreover, the reconstruction unit 43 reconstructs a one-dimensional power spectrum based on the principal components acquired by the analysis unit 42 and the principal component scores before the change. In Fig. 2, the one-dimensional power spectrum reconstructed without changing the principal component scores is represented as "Xrec".

[0023] The reconstruction unit 43 calculates a second ratio (Xrec / Xorg) indicating the ratio of change of the one-dimensional power spectrum reconstructed without changing the principal component scores to the one-dimensional power spectrum of the original image. The reconstruction unit 43 modulates the amplitude spectrum of the original image based on the calculated second ratio. The reconstruction unit 43 generates a pre-modulation image by performing an inverse Fourier transform on the amplitude spectrum and phase spectrum modulated based on the second ratio.

[0024] Then, the reconstruction unit 43 generates a difference image between the generated post-modification image and the pre-modification image. The reconstruction unit 43 outputs the generated difference image as a feature image representing the feature of the original image. The feature image output from the reconstruction unit 43 is transmitted to the user terminal 2 and displayed on the display device 22.

[0025] FIG. 3 is a flowchart showing the process performed by the image analysis device 4 shown in FIG.

[0026] In step S1, the processing device 40 performs a Fourier transform on each of a plurality of original images to obtain an amplitude spectrum and a phase spectrum of each original image.

[0027] In step S2, the processing device 40 calculates a one-dimensional power spectrum of each of the original images based on each of the acquired amplitude spectra.

[0028] In step S3, the processing device 40 performs principal component analysis on the calculated one-dimensional power spectra to obtain principal component vectors and principal component scores.

[0029] In step S4, the processing unit 40 reconstructs a one-dimensional power spectrum based on the changed principal component score values. The processing unit 40 reconstructs a one-dimensional power spectrum based on the unchanged principal component score values.

[0030] In step S5, the processing device 40 calculates a first ratio indicating the ratio of change in the one-dimensional power spectrum reconstructed by changing the principal component scores to the one-dimensional power spectrum of the original image. The processing device 40 calculates a second ratio indicating the ratio of change in the one-dimensional power spectrum reconstructed without changing the principal component scores to the one-dimensional power spectrum of the original image.

[0031] In step S6, the processing device 40 modulates the amplitude spectrum of the original image based on the calculated first ratio, and the processing device 40 modulates the amplitude spectrum of the original image based on the calculated second ratio.

[0032] In step S7, the processing device 40 performs an inverse Fourier transform on the amplitude spectrum and the phase spectrum modulated based on the first ratio to generate a post-transformation image, and performs an inverse Fourier transform on the amplitude spectrum and the phase spectrum modulated based on the second ratio to generate a pre-transformation image.

[0033] In step S8, the processing device 40 calculates the difference between the post-modification image and the pre-modification image to generate a difference image. The processing device 40 outputs the generated difference image to the storage device 45 as a feature image representing the feature of the original image, and stores it there. The processing device 40 outputs the feature image stored in the storage device 45 to the communication device of the server 3. The communication device of the server 3 transmits the difference image output from the processing device 40 to the user terminal 2. The user terminal 2 displays the difference image transmitted from the server 3 on the display device 22.

[0034] As described above, the image analysis device 4 of the first embodiment includes a conversion unit 41 that performs Fourier transform on each of a plurality of original images representing the structure of a material to obtain a plurality of power spectra (one-dimensional power spectra), an analysis unit 42 that performs principal component analysis on the plurality of power spectra to obtain principal components and principal component scores, and a reconstruction unit 43 that reconstructs an image based on the principal components and principal component scores and outputs a feature image representing the features of the original image. The reconstruction unit 43 generates a difference image between the reference image and the modified image reconstructed by changing the principal component scores, and outputs the generated difference image as a feature image.

[0035] As a result, the image analysis device 4 acquires periodic structure information of the original image as a power spectrum, so that the number of dimensions of the data to be analyzed can be reduced. Since the image analysis device 4 performs principal component analysis on a plurality of power spectra, the feature amount of the original image can be extracted relatively easily as a principal component score. Since the image analysis device 4 generates and outputs a difference image between the reference image and the changed image reconstructed by changing the principal component score, the change caused in the image due to the change in the principal component score can be directly reflected and expressed on the image. Therefore, the image analysis device 4 can easily visualize which area in the image changes due to the change in the feature amount, so that the user can easily intuitively grasp it. Therefore, the image analysis device 4 can easily visualize the features of the image representing the structure of the material.

[0036] Furthermore, in the image analysis device 4 of the first embodiment, the reconstruction unit 43 generates a pre-transformation image reconstructed without changing the principal component scores as a reference image, and generates and outputs a difference image.

[0037] As a result, the image analysis device 4 generates a difference image by comparing the reconstructed images (the post-change image and the pre-change image), and therefore it is possible to prevent errors occurring during reconstruction from being reflected in the difference image. The image analysis device 4 can easily and accurately visualize which areas in the image have changed due to changes in the feature amount. Therefore, the image analysis device 4 can easily and accurately visualize the features of the image that represent the structure of the material.

[0038] Furthermore, in the image analysis device 4 of the first embodiment, the conversion unit 41 performs a Fourier transform on the original image to obtain a spectrum and a phase spectrum, and calculates a power spectrum based on the obtained amplitude spectrum. The reconstruction unit 43 calculates a first ratio indicating a ratio of change in the power spectrum reconstructed by changing the principal component scores to the power spectrum of the original image, and modulates the amplitude spectrum of the original image based on the calculated first ratio. The reconstruction unit 43 generates a transformed image based on the amplitude spectrum and phase spectrum modulated based on the first ratio.

[0039] This allows the image analysis device 4 to reconstruct the post-modification image and the pre-modification image without losing the phase information of the original image. The image analysis device 4 can easily associate an area in the post-modification image where a change has occurred due to a change in the principal component score with an area in the pre-modification image. The image analysis device 4 can more easily and accurately visualize which area in the image has changed due to a change in the feature amount. Therefore, the image analysis device 4 can more easily and accurately visualize the features of the image that represent the structure of the material.

[0040] Furthermore, in the image analyzing device 4 of the first embodiment, the reconstruction unit 43 calculates a second ratio indicating a ratio of change in the power spectrum reconstructed without changing the principal component scores to the power spectrum of the original image, and modulates the amplitude spectrum of the original image based on the calculated second ratio. The reconstruction unit 43 generates a pre-modulation image based on the amplitude spectrum and phase spectrum modulated based on the second ratio.

[0041] This allows the image analysis device 4 to easily and accurately associate a region in the post-change image where a change has occurred due to a change in the principal component score with a region in the pre-change image. The image analysis device 4 can more easily and accurately visualize which region in the image has changed due to a change in the feature amount. Thus, the image analysis device 4 can more easily and accurately visualize the features of the image that represent the structure of the material.

[0042] [Second embodiment] An image analyzing device 4 of the second embodiment will be described with reference to Fig. 4. In the image analyzing device 4 of the second embodiment, the description of the same configuration and operation as in the first embodiment will be omitted. Fig. 4 is a diagram for explaining the image analyzing device 4 of the second embodiment.

[0043] When the one-dimensional power spectrum of the original image has peaks in multiple frequency ranges, it is preferable to be able to visualize which region in the image the peak in each frequency range represents. In the image analysis device 4 of the second embodiment, a difference image is generated and output by limiting the frequency range to that specified by the user.

[0044] Specifically, when a principal component score changed by a user is set as shown by reference numeral 101 in Fig. 4, the reconstruction unit 43 of the second embodiment reconstructs a one-dimensional power spectrum with a changed principal component score and a one-dimensional power spectrum with an unchanged principal component score as shown by reference numeral 102 in Fig. 4. Then, as shown by reference numeral 103 in Fig. 4, the reconstruction unit 43 calculates a differential power spectrum indicating a difference between the one-dimensional power spectrum reconstructed by changing the principal component score and the one-dimensional power spectrum reconstructed without changing the principal component score. For example, the reconstruction unit 43 calculates the differential power spectrum by subtracting each power value of the one-dimensional power spectrum reconstructed without changing the principal component score from each power value of the one-dimensional power spectrum reconstructed by changing the principal component score.

[0045] Then, when a frequency range is designated by the user as shown by reference numerals 104 and 105 in Fig. 4, the reconstruction unit 43 extracts a differential power spectrum of the designated frequency range from the calculated differential power spectrum as shown by reference numeral 106 in Fig. 4. At this time, the reconstruction unit 43 sets the portions outside the designated frequency range among the power values ​​of the one-dimensional power spectrum in which the principal component scores have been changed to the same values ​​as the power values ​​of the one-dimensional power spectrum in which the principal component scores have not been changed. This enables the reconstruction unit 43 to extract a differential power spectrum of the designated frequency range.

[0046] Then, the reconstruction unit 43 generates and outputs a difference image for a specified frequency range based on the extracted difference power spectrum, as indicated by reference numeral 107 in Fig. 4. Note that reference numeral 108 in Fig. 4 indicates a difference image for the entire frequency range.

[0047] As a result, the image analyzing device 4 of the second embodiment can generate a difference image limited to a specified frequency range even when the one-dimensional power spectrum of the original image has peaks in multiple frequency ranges. The image analyzing device 4 of the second embodiment can visualize which region in the image is represented by each peak in the multiple frequency ranges. Therefore, the image analyzing device 4 of the second embodiment can visualize the features of the image representing the structure of the material in more detail.

[0048] [Third embodiment] The image analyzing device 4 of the third embodiment will be described with reference to Fig. 5. In the image analyzing device 4 of the third embodiment, the description of the same configuration and operation as those of the first and second embodiments will be omitted. Fig. 5 is a diagram for explaining the image analyzing device 4 of the third embodiment.

[0049] The image analyzing device 4 of the third embodiment superimposes the original image and a difference image in which each power value of the difference power spectrum has been converted to an absolute value, and outputs the superimposed image.

[0050] Specifically, as shown by reference numeral 201 in Fig. 5, the reconstruction unit 43 of the third embodiment calculates a differential power spectrum by a method similar to that of the reconstruction unit 43 of the second embodiment, and generates a differential image. Then, as shown by reference numeral 202 in Fig. 5, the reconstruction unit 43 converts each power value of the differential power spectrum into an absolute value, and generates an absolute value differential image. Then, the reconstruction unit 43 superimposes the absolute value differential image shown by reference numeral 202 in Fig. 5 on the original image shown by reference numeral 203 in Fig. 5. Then, the reconstruction unit 43 outputs a composite image in which the absolute value differential image and the original image are superimposed, as shown by reference numeral 204 in Fig. 5.

[0051] As a result, the image analyzing device 4 of the third embodiment overlays the difference image on the original image, making it easier for the user to understand which areas in the image have changed due to changes in the feature amounts. The image analyzing device 4 of the third embodiment generates an absolute valued difference image, making it easier for the user to understand which areas in the image have changed significantly due to changes in the feature amounts. Thus, the image analyzing device 4 of the third embodiment can visualize the features of an image that represents the structure of a material in an easier-to-understand manner for the user.

[0052] [Fourth embodiment] An image analyzing device 4 of the fourth embodiment will be described with reference to Fig. 6. In the image analyzing device 4 of the fourth embodiment, the description of the same configuration and operation as those of the first to third embodiments will be omitted. Fig. 6 is a diagram for explaining the image analyzing device 4 of the fourth embodiment.

[0053] The image analyzing device 4 of the fourth embodiment separates the frequency range in which each power value of the differential power spectrum represents a positive value from the frequency range in which each power value represents a negative value, and generates and outputs a differential image.

[0054] Specifically, the reconstruction unit 43 of the fourth embodiment calculates the differential power spectrum by a method similar to that of the reconstruction unit 43 of the second embodiment, as indicated by reference numeral 301 in Fig. 6. Then, the reconstruction unit 43 specifies a frequency range (fpos) in which each power value of the differential power spectrum represents a positive value and a frequency range (fneg) in which each power value represents a negative value.

[0055] Then, as shown by reference numeral 302 in Fig. 6, the reconstruction unit 43 sets each power value in the frequency range (fneg) representing negative values ​​to the same value as each power value in the one-dimensional power spectrum in which the principal component score is not changed. In this way, the reconstruction unit 43 extracts a differential power spectrum in the frequency range representing positive values. Similarly, as shown by reference numeral 303 in Fig. 6, the reconstruction unit 43 sets each power value in the frequency range (fpos) representing positive values ​​to the same value as each power value in the one-dimensional power spectrum in which the principal component score is not changed. In this way, the reconstruction unit 43 extracts a differential power spectrum in the frequency range representing negative values.

[0056] Then, the reconstruction unit 43 generates and outputs a difference image based on the difference power spectrum in the frequency range representing positive values, as indicated by reference numeral 304 in Fig. 6. Similarly, the reconstruction unit 43 generates and outputs a difference image based on the difference power spectrum in the frequency range representing negative values, as indicated by reference numeral 305 in Fig. 6. In this way, the reconstruction unit 43 generates and outputs a difference image in the frequency range representing positive values ​​and a difference image in the frequency range representing negative values ​​separately.

[0057] As a result, the image analysis device 4 of the fourth embodiment can generate a difference image by separating the frequency range in which the power value of the one-dimensional power spectrum increases and the frequency range in which it decreases due to the change in the principal component score. The image analysis device 4 of the fourth embodiment can separate and visualize which region in the image is characterized by the frequency range having a positive correlation with the change in the feature amount and which region in the image is characterized by the frequency range having a negative correlation with the change in the feature amount. Thus, the image analysis device 4 of the fourth embodiment can visualize the image features representing the structure of the material in more detail.

[0058] The user terminal 2 may accept a plurality of material performance values ​​(material characteristics) input by a user and associated with a plurality of original images, respectively, and transmit them to the image analyzing device 4 of the server 3. The storage device 45 of the image analyzing device 4 stores the plurality of original images and the plurality of material performance values ​​received from the user terminal 2, in association with each other.

[0059] The processing device 40 of the image analysis device 4 may further include a determination unit that determines a principal component score corresponding to an important feature amount that affects the performance of the material based on the result of the principal component analysis by the analysis unit 42. The determination result by the determination unit is transmitted to the user terminal 2 and displayed on the display device 22. The user can grasp the principal component score corresponding to the important feature amount that affects the performance of the material. The user can change the value of the principal component score corresponding to the important feature amount and input it to the user terminal 2. The principal component score value input by the user is transmitted to the server 3 and set in the reconstruction unit 43.

[0060] The reconstruction unit 43 changes the principal component scores that are determined by the determination unit to correspond to important features, and then reconstructs a one-dimensional power spectrum based on the changed principal component scores to generate a post-change image.

[0061] The determination unit has a regression / classification model in which the principal component scores acquired by the analysis unit 42 are explanatory variables and each material performance value is an objective variable. The regression / classification model is constructed by machine learning. The regression / classification model is constructed using a known model. For example, the regression / classification model is constructed using a linear model such as Lasso or Ridge, or a decision tree model such as RandomForest. The determination unit calculates the contribution of the explanatory variable to the prediction of the objective variable from the learned regression / classification model. The contribution is calculated using a known method. For example, when the regression / classification model is a linear model, the contribution is calculated based on the regression coefficient. For example, when the regression / classification model is a decision tree model, the contribution is calculated based on the Gini impurity or cross entropy.

[0062] By including the above-mentioned determination unit, the image analysis device 4 can limit the principal component scores to be changed to those that correspond to important feature quantities that affect the performance of the material. Therefore, the image analysis device 4 can visualize which area in the image will change due to a change in an important feature quantity that affects the performance of the material. Therefore, the image analysis device 4 can visualize the features of the image that represent the structure of the material, and can visualize the correlation between the material performance and the feature quantities of the image.

[0063] Although the embodiments of the present invention have been described in detail above, the present invention is not limited to the above-mentioned embodiments, and various modifications can be made without departing from the spirit of the present invention. In the present invention, the configuration of one embodiment can be added to the configuration of another embodiment, the configuration of one embodiment can be replaced with the configuration of another embodiment, or a part of the configuration of one embodiment can be deleted. [Explanation of symbols]

[0064] 1...information processing system, 2...user terminal, 21...processing device, 22...display device, 3...server, 4...image analysis device, 40...processing device, 41...conversion unit, 42...analysis unit, 43...reconstruction unit, 45...storage device, N...network

Claims

1. a conversion unit that performs a Fourier transform on each of a plurality of original images representing a structure of a material to obtain a plurality of power spectra; an analysis unit that performs a principal component analysis on the plurality of power spectra to obtain principal components and principal component scores; a reconstruction unit that reconstructs an image based on the principal components and the principal component scores, and outputs a feature image that represents a feature of the original image, The reconstruction unit generates a difference image between a reference image and a post-change image reconstructed by changing the principal component score, and outputs the generated difference image as the feature image.

1. An image analysis device comprising:

2. The reconstruction unit generates a pre-transformation image reconstructed without changing the principal component scores as the reference image, generates the difference image, and outputs the difference image.

2. The image analysis device according to claim 1 .

3. the transform unit performs the Fourier transform on the original image to obtain an amplitude spectrum and a phase spectrum, and calculates the power spectrum based on the obtained amplitude spectrum; The reconstruction unit is Calculating a first ratio indicating a ratio of change in the power spectrum reconstructed by changing the principal component scores to the power spectrum of the original image; modulating the amplitude spectrum of the original image based on the calculated first ratio; generating the transformed image based on the amplitude spectrum and the phase spectrum modulated based on the first ratio; 3. The image analysis device according to claim 2.

4. The reconstruction unit is Calculating a second ratio indicating a ratio of change in the power spectrum reconstructed without changing the principal component scores to the power spectrum of the original image; modulating the amplitude spectrum of the original image based on the calculated second ratio; generating the pre-conversion image based on the amplitude spectrum and the phase spectrum modulated based on the second ratio; 4. The image analysis device according to claim 3.

5. The reconstruction unit is calculating a differential power spectrum indicating a difference between the power spectrum reconstructed by changing the principal component scores and the power spectrum reconstructed without changing the principal component scores; Extracting the differential power spectrum in a specified frequency range from the calculated differential power spectrum; The difference image of the specified frequency range is generated based on the extracted difference power spectrum, and is output.

3. The image analysis device according to claim 2.

6. The reconstruction unit is calculating a differential power spectrum indicating a difference between the power spectrum reconstructed by changing the principal component scores and the power spectrum reconstructed without changing the principal component scores; Identifying a frequency range in which each power value of the calculated difference power spectrum represents a positive value and a frequency range in which each power value represents a negative value; The difference image in the frequency range representing the positive values ​​and the difference image in the frequency range representing the negative values ​​are generated separately, and the generated difference images are output.

3. The image analysis device according to claim 2.

Images