Image processing system, image processing method, and image processing program

Abstract

To provide an image processing system, an image processing method, and an image processing program capable of accurately evaluating the shapes of primary particles that form an agglomerated particle.SOLUTION: An image processing system 10 includes: an image acquisition unit 11 which acquires a first original image indicating a plurality of agglomerated particles each formed by aggregating a plurality of primary particles; a binary processing unit 12 which executes first binary processing to generate a first binary image that distinguishes a foreground region composed of a plurality of agglomerated particles from a background region which is other than the foreground region; a generation unit 13 which generates, on the basis of the first original image and the first binary image, a second original image indicating some of the multiple agglomerated particles so that pixel values of pixels of the second original image may be pixel values of pixels in the first original image corresponding to the pixels of the second original image; and a binary processing unit 12 which executes second binary processing to generate a second binary image that distinguishes a foreground region composed of the some agglomerated particles from a background region which is other than the foreground region.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] One aspect of the present disclosure relates to an image processing system, an image processing method, and an image processing program.

Background Art

[0002] Patent Document 1 describes a method for evaluating the degree of connection of primary particles in porous secondary particles from an observation image.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] A method for accurately evaluating the shape of primary particles forming aggregated particles is desired.

Means for Solving the Problems

[0005] An image processing system according to one aspect of the present disclosure includes at least one processor. The at least one processor acquires a first original image showing a plurality of aggregated particles formed by aggregation of a plurality of primary particles, performs a first binarization process on the first original image to generate a first binary image that distinguishes a foreground region composed of the plurality of aggregated particles from a background region outside the foreground region, extracts from the first binary image a partial image that is a part of the first binary image and shows some of the plurality of aggregated particles, converts the pixel value of each pixel of the partial image into the pixel value of each pixel of the first original image corresponding to each pixel of the partial image to generate a second original image, and performs a second binarization process on the second original image to generate a second binary image that distinguishes a foreground region composed of some of the aggregated particles from a background region outside the foreground region.

[0006] The image processing method according to one aspect of the present disclosure is executed by an image processing system including at least one processor. The image processing method includes: obtaining a first original image showing a plurality of aggregated particles respectively formed by aggregation of a plurality of primary particles; performing a first binarization process on the first original image to generate a first binary image distinguishing a foreground region composed of the plurality of aggregated particles from a background region other than the foreground region; generating a second original image showing some of the plurality of aggregated particles based on the first original image and the first binary image, wherein a pixel value of each pixel of the second original image is a pixel value of each pixel of the first original image corresponding to each pixel of the second original image; and performing a second binarization process on the second original image to generate a second binary image distinguishing a foreground region composed of some of the aggregated particles from a background region other than the foreground region.

[0007] The image processing program according to one aspect of the present disclosure causes a computer to execute: obtaining a first original image showing a plurality of aggregated particles respectively formed by aggregation of a plurality of primary particles; performing a first binarization process on the first original image to generate a first binary image distinguishing a foreground region composed of the plurality of aggregated particles from a background region other than the foreground region; generating a second original image showing some of the plurality of aggregated particles based on the first original image and the first binary image, wherein a pixel value of each pixel of the second original image is a pixel value of each pixel of the first original image corresponding to each pixel of the second original image; and performing a second binarization process on the second original image to generate a second binary image distinguishing a foreground region composed of some of the aggregated particles from a background region other than the foreground region.

[0008] In such an aspect, a first binary image showing a plurality of aggregated particles is obtained by a first binarization process. Then, based on the first original image and the first binary image, a second original image showing some of the aggregated particles among the plurality of aggregated particles is generated, where the pixel value of each pixel is the pixel value of each pixel of the first original image. A second binarization process is performed on this second original image to generate a second binary image for the some of the aggregated particles. In the first binarization process, the appearance of individual aggregated particles may affect each other among the plurality of aggregated particles, and such an influence may be reflected in the first binary image. By performing the second binarization process focusing on some of the aggregated particles, such an influence on the appearance is suppressed and the shape of the aggregated particles can be specified more accurately. By using this second binary image, it becomes possible to evaluate the shape of the primary particles forming some of the aggregated particles with high accuracy.

Advantages of the Invention

[0009] According to one aspect of the present disclosure, it becomes possible to evaluate the shape of the primary particles forming the aggregated particles with high accuracy.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Embodiments for Carrying Out the Invention

[0011] Hereinafter, embodiments in the present disclosure will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same or equivalent elements are denoted by the same reference numerals, and duplicate descriptions are omitted.

[0012] [Overview of the System] The image processing system according to the present disclosure is a computer system that executes image processing on a raw image showing aggregated particles. In one example, the image processed by the image processing system is used to evaluate the shape of the primary particles forming the aggregated particles. The evaluation may be performed by the image processing system or by a computer system separate from the image processing system. "Evaluating the shape of the primary particles" refers to a process including at least calculating a value related to the shape of the primary particles.

[0013] Aggregated particles refer to particles formed by aggregation of a plurality of primary particles. Primary particles refer to particles that cannot be further divided finely. That is, primary particles refer to the minimum unit of particles. As an example, the aggregated particles are formed from a plurality of carbon blacks and are used as a conductive agent in a lithium-ion battery. In this example, the smaller the particle size of the carbon black, the more parallel circuits are formed by the carbon black, so that many conductive paths are formed even with a small amount of carbon black. That is, in this example, by evaluating the particle size of the carbon black by the image processing system, the user can grasp the performance of the aggregated particles as a conductive agent. As an example, the particle size of this carbon black is about 22 to 26 nm.

[0014] A computer system calculates shape parameters related to primary particles based on shape parameters related to the shape of aggregated particles in order to evaluate the shape of the primary particles. As an example, the shape parameters of the aggregated particles are area, perimeter, perimeter of the convex hull, area of the convex hull, length, width, equivalent circle diameter, circularity, area circularity, envelope degree, area envelope degree, aspect ratio, elongation degree, etc. As an example, the shape parameters of the primary particles are the average particle diameter of the primary particles, etc. The computer system may evaluate the number of primary particles contained in the aggregated particles, the morphology of the aggregated particles, the electron microscope surface area, the fractal dimension of the entire aggregated particles, etc. As described above, an image processing system may perform such calculations and evaluations.

[0015] The original image is obtained by imaging with an imaging device such as a scanning electron microscope (SEM) or a transmission electron microscope (TEM). As an example, when using SEM, the original image showing the aggregated particles is obtained by the following procedure. First, the user disperses a plurality of aggregated particles in a solvent such as chloroform to prepare a sample. Next, the user drops the prepared sample onto a substrate and dries the sample. Then, by imaging the aggregated particles on the substrate with an imaging device, an original image showing the aggregated particles is obtained. When using SEM, the occurrence of unevenness in the background region of the original image is suppressed.

[0016] [Configuration of the System] FIG. 1 is a diagram showing the functional configuration of an image processing system 10 according to an example. The image processing system 10 includes, as functional elements, an image acquisition unit 11, a binarization processing unit 12, a generation unit 13, and an evaluation unit 14.

[0017] The image acquisition unit 11 is a functional module that acquires the first original images 211 and 212 obtained by the imaging device. The first original image 211 shows a plurality of aggregated particles. The first original image 212 shows some of the aggregated particles among the plurality of aggregated particles in a certain first original image 211. Some of the aggregated particles refer to aggregated particles that are 1 or more and less than N, assuming that the number of aggregated particles reflected in the first original image 211 is N. The resolution of the first original image 212 is higher than that of the first original image 211. Resolution refers to a numerical value indicating the density of pixels with respect to the actual length in an image. As the resolution increases, the length per pixel in the image becomes shorter. That is, relatively high resolution means that the length per pixel in the image is relatively short. Due to such a difference in resolution, the first original image 212 can reflect individual aggregated particles more clearly than the first original image 211.

[0018] The binarization processing unit 12 is a functional module that performs binarization processing on the first original images 211 and 212 and the second original images 231 and 232. Binarization processing refers to converting the pixel value of each pixel in an image into a binary value. By performing binarization processing on the original image, a binary image that distinguishes the foreground region composed of aggregated particles from the background region other than the foreground region is generated. The binarization processing unit 12 performs binarization processing on the first original image 211 to generate a first binary image 221, and performs binarization processing on the second original image 231 to generate a second binary image 241. The binarization processing unit 12 performs binarization processing on the first original image 212 to generate a first binary image 222, and performs binarization processing on the second original image 232 to generate a second binary image 242.

[0019] The generation unit 13 is a functional module that generates one or more second original images 231 based on the first original image 211 and the first binary image 221, and generates one or more second original images 232 based on the first original image 212 and the first binary image 222.

[0020] Each second original image 231 shows some of the aggregated particles in the first original image 211. The pixel value of each pixel in the second original image 231 is the pixel value of each pixel in the first original image 211 corresponding to the respective pixel in the second original image 231.

[0021] When a certain first original image 212 shows two or more aggregated particles, each second original image 232 generated based on the first original image 212 shows some of the two or more aggregated particles. When a certain first original image 212 shows one aggregated particle, the second original image 232 is consequently the same as the first original image 212. The pixel value of each pixel in the second original image 232 is the pixel value of each pixel in the first original image 212 corresponding to the respective pixel in the second original image 232.

[0022] The evaluation unit 14 is a functional module that evaluates the shape of primary particles. The evaluation unit 14 acquires the second binary image 241 and the second binary image 242, and calculates the shape parameters of any aggregated particle in these images. Then, the evaluation unit 14 evaluates the shape of the primary particles forming the any aggregated particle based on the calculated shape parameters.

[0023] FIG. 2 is a diagram showing a general hardware configuration of a computer 100 that can function as the image processing system 10. For example, the computer 100 includes a processor 101, a main memory unit 102, an auxiliary storage unit 103, a communication control unit 104, an input device 105, and an output device 106. The processor 101 executes an operating system and an application program. The main memory unit 102 is composed of, for example, ROM and RAM. The auxiliary storage unit 103 is composed of, for example, a hard disk or a flash memory, and generally stores a larger amount of data than the main memory unit 102. The auxiliary storage unit 103 non-temporarily stores a program 110 for causing at least one computer to function as the image processing system 10. The communication control unit 104 is composed of, for example, a network card or a wireless communication module. The input device 105 is composed of, for example, a keyboard, a mouse, a touch panel, etc. The output device 106 is composed of, for example, a monitor and a speaker.

[0024] Each functional module of the image processing system 10 is realized by causing the processor 101 or the main memory unit 102 to load the program 110 thereon and causing the processor 101 to execute the program 110. The program 110 includes code for realizing each functional element of the image processing system 10. The processor 101 operates the communication control unit 104, the input device 105, or the output device 106 according to the program 110, and reads and writes data in the main memory unit 102 or the auxiliary storage unit 103. Each functional element of the image processing system 10 is realized by this processing. Data or databases necessary for the processing may be stored in the main memory unit 102 or the auxiliary storage unit 103.

[0025] The program 110 corresponds to an image processing program. The program 110 may be provided after being non-temporarily recorded on a tangible recording medium such as a CD-ROM, a DVD-ROM, or a semiconductor memory. Alternatively, the program 110 may be provided via a communication network as a data signal superimposed on a carrier wave.

[0026] The image processing system 10 can be configured by one or more computers. When a plurality of computers are used, the image processing system 10 is configured by connecting these computers to each other via a communication network.

[0027] [Image] Various images processed by the image processing system 10 will be described. As described above, the resolution of the first original image 212 is higher than that of the first original image 211. The resolutions of the first binary image 221, the second original image 231, and the second binary image 241 are the same as that of the first original image 211, and the resolutions of the first binary image 222, the second original image 232, and the second binary image 242 are the same as that of the first original image 212. Therefore, it can be said that each of the first original image 211, the first binary image 221, the second original image 231, and the second binary image 241 is a relatively low-resolution image, and it can be said that each of the first original image 212, the first binary image 222, the second original image 232, and the second binary image 242 is a relatively high-resolution image.

[0028] As described above, assuming that the first original image 211 shows N aggregated particles, the first original image 212 shows 1 or more and less than N aggregated particles, which are part of the N aggregated particles. The image acquisition unit 11 can acquire 2 or more first original images 212 with different imaging ranges for one first original image 211. Among the first original image 211 and the first binary image 221, the plurality of aggregated particles shown are the same. Among the corresponding first original image 212 and the first binary image 222, the 1 or more aggregated particles shown are also the same.

[0029] The second original image 231 shows some of the aggregated particles among the plurality of aggregated particles in the first original image 211. In other words, it can be said that the second original image 231 is an image that captures a part of the imaging range of the first original image 211. Each individual second original image 231 may show one aggregated particle or 2 or more aggregated particles. Assuming that the first original image 211 shows N aggregated particles, the generation unit 13 may generate a plurality of second original images 231 such that all of the N aggregated particles are shown by any of the second original images 231. When each second original image 231 shows one aggregated particle, N second original images 231 are generated. The generation unit 13 may also generate a plurality of second original images 231 such that some of the N aggregated particles are shown by any of the second original images 231. In this case, the remaining aggregated particles are not shown by any of the second original images 231.

[0030] The relationship between the first original image 212 and the second original image 232 can vary depending on the number of aggregated particles indicated by the first original image 212.

[0031] When a certain first original image 212 shows two or more aggregated particles, each second original image 232 generated based on the first original image 212 shows some of the aggregated particles among the two or more aggregated particles. In this case, each second original image 232 may show one aggregated particle or two or more aggregated particles. Assuming the first original image 212 shows M aggregated particles, the generation unit 13 may generate a plurality of second original images 232 such that all of the M aggregated particles are shown by any of the second original images 232. When each second original image 232 shows one aggregated particle, M second original images 232 are generated. The generation unit 13 may generate a plurality of second original images 232 such that some of the M aggregated particles are shown by any of the second original images 232. In this case, the remaining aggregated particles are not shown by any of the second original images 232.

[0032] On the other hand, when a certain first original image 212 shows one aggregated particle, the second original image 232 is consequently the same as the first original image 212.

[0033] Among the corresponding second original image 231 and the second binary image 241, the one or more aggregated particles shown are the same. Among the corresponding second original image 232 and the second binary image 242 as well, the one or more aggregated particles shown are the same.

[0034] [Operation of the System] While explaining the operation of the image processing system 10 with reference to FIG. 3, an image processing method according to an example will also be described. FIG. 3 is a flowchart showing an example of the processing in the image processing system 10 as a processing flow S1. The processing from the first original image 212 until the generation of the second binary image 242 through the first binary image 222 and the second original image 232 is the same as the processing from the first original image 211 until the generation of the second binary image 241 through the first binary image 221 and the second original image 231.

[0035] In step S11, the image acquisition unit 11 acquires a first original image 211 and a first original image 212 with different resolutions from each other. The image acquisition unit 11 may receive the first original images 211 and 212 from an imaging device or another computer. Alternatively, the image acquisition unit 11 may access a given database or file system to read out the first original images 211 and 212. Alternatively, the image acquisition unit 11 may acquire the first original images 211 and 212 input by the user of the image processing system 10. The acquired first original images 211 and 212 are grayscale images. In a grayscale image, the pixel value of each pixel is any integer value between 0 corresponding to black and 255 corresponding to white. The image acquisition unit 11 may acquire the first original images 211 and 212 by converting a color image into a grayscale image. In one example, the image acquisition unit 11 acquires an SEM image obtained by imaging with an SEM. Since the SEM image can be acquired with high contrast, it is suitable for binarization processing.

[0036] The resolution of the first original image 212 is higher than the resolution of the first original image 211. That is, assuming that the length per pixel in the first original image 211 is the pixel length L1 and the length per pixel in the first original image 212 is the pixel length L2, the pixel length L2 is shorter than the pixel length L1. As an example, the pixel length L1 is 2.48 [nm / pixel], and the pixel length L2 is 0.55 [nm / pixel]. The pixel length L1 may be 1.00 to 5.00 [nm / pixel], or may be 1.50 to 3.00 [nm / pixel]. The pixel length L2 may be 0.10 to 1.00 [nm / pixel], or may be 0.50 to 0.71 [nm / pixel]. The pixel length L2 may be 0.5 times or less, 0.3 times or less, or 0.1 times or less of the pixel length L1. The difference in resolution between the first original image 211 and the first original image 212 may be represented by the difference in magnification of the imaging device when the first original image 211 and the first original image 212 are imaged. As an example, the first original image 211 may be an image imaged at a magnification of 20,000 times, and the first original image 212 may be an image imaged at a magnification of 90,000 times.

[0037] FIG. 4 is a diagram showing an example of the first original images 211 and 212. The first original image 211 captures a plurality of aggregated particles 300 existing in a relatively wide subject range. On the other hand, the first original image 212 captures a part of the subject range with a higher resolution than the first original image 211. Therefore, the first original image 212 shows some of the aggregated particles 300 among the plurality of aggregated particles 300 shown by the first original image 211 more clearly than the first original image 211. FIG. 4 shows two first original images 212 corresponding to the partial regions 211a and 211b of the first original image 211.

[0038] Returning to FIG. 3, in step S12, the binarization processing unit 12 performs first binarization processing on the first original image to generate a first binary image that distinguishes a foreground region composed of a plurality of aggregated particles from a background region other than the foreground region. The binarization processing unit 12 performs first binarization processing on the first original image 211 to generate a first binary image 221, and performs first binarization processing on the first original image 212 to generate a first binary image 222. As an example, the first binarization processing is a process based on a mixture Gaussian model. The details of the mixture Gaussian model will be described with reference to FIG. 5. FIG. 5 is a diagram showing an example of a histogram of pixel values in the first original image 211.

[0039] As shown in FIG. 5, in the first original image which is a grayscale image, histograms in the foreground region and the background region each form a mountain-shaped distribution. In FIG. 5, the histogram in the background region is shown as the skirt portion of the mountain-shaped distribution. The mixture Gaussian model assumes that each mountain-shaped distribution follows a Gaussian distribution, and then calculates the occurrence frequency F(x i (i = 0 to 255)) as a weighted linear sum of a plurality of Gaussian distributions. Here, x0 to x i correspond to 0 to 255 which are grayscale pixel values respectively. The binarization processing unit 12 calculates the occurrence frequency F(x 255 ) according to the following formula (1) based on the mixture Gaussian model. Here, let the Gaussian distributions be f1(x i ), f2(x i ). f1(x i ) andi ) corresponds to the Gaussian distribution in the foreground region, f2(x i ) corresponds to the Gaussian distribution in the background region. The mean of f1(x i ) is μ1, and the variance of f1(x i ) is σ1 2 . The mean of f2(x i ) is μ2, and the variance of f2(x i ) is σ2 2 . Also, let the weight of f1(x i ) be w1, and the weight of f2(x i ) be w2. Both w1 and w2 are real numbers greater than or equal to 0, and the sum of w1 and w2 is 1.

[0040]

Number

[0041] Then, the binarization processing unit 12 converts each pixel value to binary based on a certain threshold T. As an example, the binarization processing unit 12 converts each pixel value to either 0 or 255 based on the threshold T. As an example, the binarization processing unit 12 calculates the threshold T as follows. First, the binarization processing unit 12 calculates the log-likelihood lnL of the above formula (1) by the following formula (2).

[0042]

Number

[0043] Then, the binarization processing unit 12 estimates the values of the parameters w1, w2, μ1, σ1, μ2, σ2 that maximize the log-likelihood lnL shown in the above formula (2) based on the EM algorithm (expectation-maximization algorithm), etc. The binarization processing unit 12 calculates the threshold T based on the estimated parameters. As an example, the binarization processing unit 12 uses two Gaussian distributions f1(x i ), f2(x i) Calculate the pixel value at the intersection as the threshold value T. Alternatively, the binarization processing unit 12 calculates the average value of μ1 and μ2 as the threshold value T. Then, the binarization processing unit 12 executes the first binarization process by converting pixel values less than the calculated threshold value T to 0 and pixel values greater than or equal to the threshold value T to 255. In a grayscale image, each distribution forming the histogram of pixel values follows a Gaussian distribution or has a shape close to a Gaussian distribution. Therefore, when the first original image is a grayscale image, especially when it is an SEM image, the binarization processing unit 12 can generate the first binary image with higher accuracy based on the mixture Gaussian model.

[0044] Alternatively, the first binarization process may be a process based on Otsu's binarization method. Otsu's binarization process refers to a process of dividing a histogram into two classes and calculating the separation degree between the two classes, and calculating the pixel value at which the separation degree is maximized as the threshold value T. Alternatively, the first binarization process may be a process of performing binarization based on a threshold value T input by the user.

[0045] FIG. 6 is a diagram showing an example of the first binarization process by the binarization processing unit 12. The first original image 211 shown in this example is a grayscale SEM image. The first binary image 221 is an image after the first binarization process is executed, and the pixel value of each pixel of the first binary image 221 is 0 or 255. In the first binary image 221, the shape of the aggregated particles 300 is more clearly distinguishable than in the first original image 211. Similar to this example, in the first binary image 222, the shape of the aggregated particles is more clearly distinguishable than in the first original image 212.

[0046] Returning to FIG. 3, in step S13, the generation unit 13 generates one or more second original images based on the first original image and the first binary image. The generation unit 13 generates one or more second original images 231 based on the first original image 211 and the first binary image 221, and generates one or more second original images 232 based on the first original image 212 and the first binary image 222. The details of step S13 will be described with reference to FIG. 7. FIG. 7 is a flowchart showing an example of the method for generating the second original image.

[0047] In step S131, the generation unit 13 generates one or more partial images from each of the first binary images 221 and 222. Each partial image is a part of the first binary image and shows some of the aggregated particles among the plurality of aggregated particles. The generation unit 13 may generate a partial image showing one aggregated particle for each of the first binary images 221 and 222, or may generate a partial image showing two or more aggregated particles. As an example, the generation unit 13 may generate a plurality of partial images each showing one aggregated particle for at least one of the first binary images 221 and 222. In this example, the number of partial images corresponding to the number of aggregated particles shown in the first binary image 221 (or the first binary image 222) is obtained from the first binary image 221 (or the first binary image 222). Note that the generation unit 13 may not generate a partial image showing the aggregated particle after determining that an aggregated particle with a small size is noise.

[0048] In step S132, the generation unit 13 converts the pixel values of each pixel for one or more partial images to generate one or more second original images. For each of the one or more partial images obtained from the first binary image 221, the generation unit 13 converts the pixel value of each pixel of the partial image to the pixel value of the pixel of the first original image 211 corresponding to the pixel, and generates the second original image 231. For each of the one or more partial images obtained from the first binary image 222, the generation unit 13 converts the pixel value of each pixel of the partial image to the pixel value of the pixel of the first original image 212 corresponding to the pixel, and generates the second original image 232. The pixel value of each pixel of the second original image is the pixel value of each pixel of the first original image corresponding to each pixel of the second original image. The second original image is generated from the partial image. Therefore, the second original image shows some of the aggregated particles among the plurality of aggregated particles. For example, each second original image shows one aggregated particle.

[0049] FIG. 8 is a diagram showing an example of a method for generating a second original image, and shows an enlarged part of the first binary image 221. In this example, the generation unit 13 shows two partial images 251 and two second original images 231 corresponding to the partial regions 211c and 211d of the first binary image 221.

[0050] Alternatively, in step S13, the generation unit 13 may directly generate one or more second original images from at least one of the first original images 211 and 212. In this case, first, the generation unit 13 refers to the first original image and the first binary image, and identifies the region in the first original image corresponding to the region indicating some aggregated particles in the first binary image. Then, the generation unit 13 cuts out the identified region from the first original image to generate the second original image 231. In this way, the generation unit 13 can generate one or more second original images from the first original image without going through the generation of partial images.

[0051] Regarding step S13, when a certain first binary image 222 indicates one aggregated particle, the generation unit 13 may generate the second original image 232 from the first binary image 222 based on the pixel values of the corresponding first original image 212 without generating a partial image from the first binary image 222 or performing cutting out from the first original image 212.

[0052] Returning to FIG. 3, in step S14, the binarization processing unit 12 performs a second binarization process on each of the one or more second original images to generate one or more second binary images that distinguish the foreground region composed of some aggregated particles from the background region outside the foreground region. The binarization processing unit 12 generates one or more second binary images 241 from the one or more second original images 231 and one or more second binary images 242 from the one or more second original images 232 by the second binarization process. The second binarization process may be a process based on a mixture Gaussian model or a process based on Otsu's binarization method, similar to the first binarization process. Alternatively, the second binarization process may be a process of performing binarization based on the threshold value T input by the user. The binarization method used for the second binarization process may be the same as or different from the first binarization process.

[0053] Even if the same binarization is adopted in the first binarization process and the second binarization process, the results of the first binarization process and the second binarization process for a single aggregated particle can be different from each other. This is because the histograms of the pixel values of the first original image and the second original image can be different from each other. The first original image 211 depicts relatively many aggregated particles, and the range or distribution of pixel values can be different among individual aggregated particles. On the other hand, the second original image depicts some of the aggregated particles. Therefore, the histogram of pixel values in the entire image is different between the first original image and the second original image. Due to the difference in the histograms, the threshold value T in the binarization process is also different. Therefore, the results of the first binarization process and the second binarization process for a single aggregated particle can be different from each other.

[0054] In step S15, the evaluation unit 14 calculates a correction coefficient based on the shape parameters of some of the aggregated particles in each of one or more second binary images 241 and one or more second binary images 242. Hereinafter, the second binary image 241 is also referred to as the first image, and the second binary image 242 is also referred to as the second image. In the following description, an example is illustrated in which the set of one or more second binary images 241 (first images) shows a plurality of aggregated particles, and the set of one or more second binary images 242 (second images) shows some of the plurality of aggregated particles. In this example, some of the aggregated particles are shown in both the second binary image 241 (first image) and the second binary image 242 (second image).

[0055] As an example, the evaluation unit 14 analyzes each of one or more first images and one or more second images to obtain shape parameters. Then, the evaluation unit 14 calculates a correction coefficient based on these shape parameters.

[0056] FIG. 9 is a diagram for explaining an example of a method for calculating a correction coefficient. First, for each of the one or more aggregated particles shown in the second image of 1 or more, the evaluation unit 14 acquires the shape parameter obtained by analyzing the first image and the shape parameter obtained by analyzing the second image. The association between the first image and the second image showing a common single aggregated particle may be performed in advance by the user or may be performed by the evaluation unit 14. Then, the evaluation unit 14 approximates the relationship between the shape parameters of some of the aggregated particles obtained from the first image and the shape parameters of the some of the aggregated particles obtained from the second image with a linear function G, and calculates the slope of the linear function G as the correction coefficient. The linear function G can be expressed as y = ax, where one of the shape parameters obtained from the first image and the shape parameter obtained from the second image is x and the other is y.

[0057] Explain the technical significance of the calculated correction coefficient. By using the first image with low resolution, information on many aggregated particles can be obtained, but the accuracy of the shape parameters obtained from the first image tends to be lower than the accuracy of the shape parameters obtained from the second image with high resolution. As an example, the perimeter length of the aggregated particle obtained from the first image tends to be shorter than the perimeter length of the aggregated particle obtained from the second image with high resolution. It can be said that the correction coefficient calculated as described above indicates the degree of difference in the shape parameters between such a first image and a second image.

[0058] Returning to FIG. 3, in step S16, the evaluation unit 14 calculates a shape parameter for at least one of the aggregated particles shown by the one or more first images based on the correction coefficient. For each of the at least one aggregated particle, the evaluation unit 14 multiplies the shape parameter obtained by analyzing the first image by the correction coefficient corresponding to the shape parameter to calculate a corrected shape parameter for the aggregated particle.

[0059] In step S17, the evaluation unit 14 evaluates the shape of the primary particles of at least one of the aggregated particles shown by one or more first images based on the calculated shape parameters. For example, the evaluation unit 14 evaluates the shape of the plurality of primary particles forming each aggregated particle for each aggregated particle.

[0060] The evaluation unit 14 outputs the evaluation result. The evaluation unit 14 may display the processing result on a display device, may store the processing result in a predetermined storage device such as a memory or a database, or may transmit the processing result to another computer system.

[0061] The evaluation unit 14 may calculate the perimeter of each of some of the aggregated particles as a shape parameter based on a correction coefficient. The evaluation unit 14 may calculate the average particle size of the plurality of primary particles forming each of the some aggregated particles based on the perimeter. As an example, the evaluation unit 14 calculates the average particle size of the plurality of primary particles forming the aggregated particle based on the area of the aggregated particle and the calculated perimeter of the aggregated particle. The evaluation unit 14 may calculate the area of the aggregated particle using the correction coefficient, or may obtain the area of the aggregated particle without using the correction coefficient. As an example, the evaluation unit 14 calculates the average particle size of the primary particles in accordance with the standard of ASTM D3849. In this example, the evaluation unit 14 calculates the average particle size of the primary particles by the following formula (3). The evaluation unit 14 may calculate the number of primary particles forming one aggregated particle by the following formula (4). In formulas (3) and (4), let the perimeter of the aggregated particle be P and the area of the aggregated particle be A. α is the aggregation coefficient, dp is the average particle size of the primary particles, and n is the number of primary particles. When the calculated value of α is less than 0.4, the evaluation unit 14 sets the value of α to 0.4.

[0062]

Number

[0063]

Number

[0064] The evaluation unit 14 may execute the above processing for each of two or more types of shape parameters. In this regard, the evaluation unit 14 may calculate a correction coefficient for each type of shape parameter.

[0065] [Evaluation Example] Table 1 below shows an example of the calculation results of the average particle size of primary particles by the image processing system 10. The image processing system 10 acquired a first original image 211 with a pixel length L1 of 2.48 [nm / pixel] and a first original image 212 with a pixel length L2 of 0.55 [nm / pixel]. Both the first original images 211 and 212 were SEM images. The image processing system 10 executed a first binarization process and a second binarization process based on a mixture Gaussian model. The image processing system 10 acquired the area and perimeter for each of 15 aggregated particles, and corrected the perimeter obtained from the first original image 211 with a correction coefficient. Then, the image processing system 10 substituted the above area and the corrected perimeter into the above formulas (3) and (4) to calculate the average particle size and the number of primary particles in each aggregated particle. Then, the image processing system 10 calculated the average value of the average particle size of primary particles and the average value of the number of primary particles for the entire 15 aggregated particles. When the particle sizes of a plurality of primary particles forming one of the 15 aggregated particles were manually measured and the average particle size of the primary particles was obtained, the average particle size was 25 nm. That is, the average value of the average particle size of primary particles obtained in this evaluation example was close to the average particle size obtained by manual measurement. Therefore, in this evaluation example, it can be said that the image processing system 10 could evaluate the average particle size of primary particles with high accuracy.

[0066] [Table 1]

[0067] [Modification Example] The above has described in detail the technology according to the present disclosure based on various examples. However, the present disclosure is not limited to the above examples. For the technology according to the present disclosure, various modifications are possible without departing from the gist thereof.

[0068] In the above example, the image processing system 10 calculates the shape parameters of the aggregated particles in the second binary image 241 which is the first image after calculating the correction coefficient. However, the image processing system may evaluate the shape of the primary particles based on the shape parameters after obtaining the shape parameters of the aggregated particles in at least one of the second binary images 241 and 242 without calculating the correction coefficient. In this example, the evaluation unit of the image processing system may obtain at least one of the second binary images 241 and 242. That is, the image acquisition unit of the image processing system may obtain at least one of the first original images 211 and 212.

[0069] The processing procedures of the method executed by at least one processor are not limited to the above examples. For example, some of the above-described steps may be omitted, or each step may be executed in a different order. Also, any two or more of the above-described steps may be combined, or a part of the steps may be modified or deleted. Alternatively, other steps may be executed in addition to each of the above steps.

[0070] In the comparison of the magnitude relationship between two numerical values in the present disclosure, either of the two criteria of "greater than or equal to" and "greater than" may be used, and either of the two criteria of "less than or equal to" and "less than" may be used.

[0071] In the present disclosure, the expression "at least one processor executes the first process, executes the second process,... executes the L-th process." or the corresponding expression indicates a concept including the case where the execution subject of the L processes from the first process to the L-th process, that is, the processor, changes in the middle. That is, this expression indicates a concept including both the case where all of the L processes are executed by the same processor and the case where the processor changes in an arbitrary manner in the L processes.

[0072] [Supplementary Note] As can be understood from the various examples above, the present disclosure includes the aspects shown below. (Supplementary Note 1) Comprising at least one processor, wherein the at least one processor acquires a first original image showing a plurality of aggregated particles each formed by aggregation of a plurality of primary particles, performs a first binarization process on the first original image to generate a first binary image that distinguishes a foreground region composed of the plurality of aggregated particles from a background region outside the foreground region, based on the first original image and the first binary image, generates a second original image showing some of the plurality of aggregated particles, wherein the pixel value of each pixel of the second original image is the pixel value of each pixel of the first original image corresponding to each pixel of the second original image, performs a second binarization process on the second original image to generate a second binary image that distinguishes a foreground region composed of some of the aggregated particles from a background region outside the foreground region, An image processing system. (Supplementary Note 2) wherein the at least one processor performs the first binarization process based on a mixture Gaussian model, performs the second binarization process based on the mixture Gaussian model, The image processing system according to Supplementary Note 1. (Supplementary Note 3) wherein the at least one processor generates a partial image that is a part of the first binary image and shows some of the plurality of aggregated particles from the first binary image, converts the pixel value of each pixel of the partial image into the pixel value of each pixel of the first original image corresponding to each pixel of the partial image to generate the second original image, The image processing system according to Supplementary Note 1 or 2. (Supplementary Note 4) The at least one processor generates the second original image indicating one of the plurality of aggregated particles as the partial aggregated particles among the plurality of aggregated particles. The image processing system according to any one of Appendices 1 to 3. (Appendix 5) For each of the partial aggregated particles in the second binary image, the at least one processor calculates an average particle size of the plurality of primary particles forming the aggregated particle based on a perimeter length of the aggregated particle. The image processing system according to any one of Appendices 1 to 4. (Appendix 6) The first original image is an image obtained by imaging with a scanning electron microscope. The image processing system according to any one of Appendices 1 to 5. (Appendix 7) An image processing method executed by an image processing system including at least one processor, obtaining a first original image showing a plurality of aggregated particles respectively formed by aggregation of a plurality of primary particles; performing a first binarization process on the first original image to generate a first binary image that distinguishes a foreground region composed of the plurality of aggregated particles from a background region outside the foreground region; generating a second original image showing some of the plurality of aggregated particles based on the first original image and the first binary image, wherein a pixel value of each pixel of the second original image is a pixel value of each pixel of the first original image corresponding to the each pixel of the second original image; performing a second binarization process on the second original image to generate a second binary image that distinguishes a foreground region composed of the partial aggregated particles from a background region outside the foreground region; An image processing method including the above steps. (Appendix 8) obtaining a first original image showing a plurality of aggregated particles respectively formed by aggregation of a plurality of primary particles; Executing a first binarization process on the first original image to generate a first binary image that distinguishes a foreground region composed of the plurality of aggregated particles from a background region outside the foreground region; Based on the first original image and the first binary image, generating a second original image that shows some of the plurality of aggregated particles, wherein the pixel value of each pixel of the second original image is the pixel value of each pixel of the first original image corresponding to each pixel of the second original image; Executing a second binarization process on the second original image to generate a second binary image that distinguishes a foreground region composed of some of the aggregated particles from a background region outside the foreground region; An image processing program for causing a computer to execute the above steps.

[0073] According to Supplementary Notes 1, 7, and 8, a first binary image showing a plurality of aggregated particles is obtained by the first binarization process. Then, based on the first original image and the first binary image, a second original image showing some of the plurality of aggregated particles, wherein the pixel value of each pixel is the pixel value of each pixel of the first original image, is generated. A second binarization process is executed on this second original image to generate a second binary image for some of the aggregated particles. In the first binarization process, the appearance of individual aggregated particles may affect each other among the plurality of aggregated particles, and such an influence may be reflected in the first binary image. By focusing on some of the aggregated particles and performing the second binarization process, the influence of such an appearance can be suppressed, and the shape of the aggregated particles can be specified more accurately. By using this second binary image, it becomes possible to evaluate the shape of the primary particles forming some of the aggregated particles with high accuracy.

[0074] According to Supplementary Note 2, a second original image, which is also a grayscale image, is generated based on the first original image that is a grayscale image or the first original image that is a grayscale image converted from a color image. By performing a first binarization process based on a mixture Gaussian model on the first original image that is a grayscale image, a first binary image in which a foreground region composed of a plurality of aggregated particles and a background region other than the foreground region are accurately distinguished is generated. By using this first binary image, a second original image is generated after accurately discriminating some of the aggregated particles among the plurality of aggregated particles. Then, by performing a second binarization process based on a mixture Gaussian model on the second original image that is a grayscale image, a second binary image in which a foreground region composed of some of the aggregated particles and a background region other than the foreground region are more accurately distinguished is generated. By using this second binary image, it becomes possible to evaluate the shape of the primary particles with higher precision.

[0075] According to Supplementary Note 3, since in the first binary image, a foreground region composed of a plurality of aggregated particles and a background region other than the foreground region are distinguished, a partial image can be easily generated from the first binary image. By converting the pixel value of each pixel of this partial image, the second original image can be easily generated.

[0076] According to Supplementary Note 4, during the second binarization process, since the influence of other aggregated particles does not affect the appearance of one aggregated particle, the shape of the one aggregated particle can be specified more accurately. As a result, it becomes possible to evaluate the shape of the primary particles with higher precision.

[0077] According to Supplementary Note 5, based on the perimeter length of some of the aggregated particles whose shapes are accurately specified in the second binary image, the average particle diameter of the plurality of primary particles forming the some of the aggregated particles can be accurately calculated.

[0078] According to Supplementary Note 6, since the contrast of each of the first original image and the second original image is high, a first binary image and a second binary image in which the shape of the aggregated particles is specified more accurately can be generated. As a result, it becomes possible to evaluate the shape of the primary particles with higher precision.

Explanation of Signs

[0079] 10… Image processing system, 11… Image acquisition unit, 12… Binarization processing unit, 13… Generation unit, 14… Evaluation unit, 211, 212… First original image, 221, 222… First binary image, 231, 232… Second original image, 241, 242… Second binary image, 251… Partial image.

Claims

1. comprising at least one processor, wherein the at least one processor, obtains a first original image showing a plurality of aggregated particles each formed by aggregation of a plurality of primary particles, performs a first binarization process on the first original image to generate a first binary image that distinguishes a foreground region composed of the plurality of aggregated particles from a background region outside the foreground region, based on the first original image and the first binary image, generates a second original image showing some of the plurality of aggregated particles, wherein a pixel value of each pixel of the second original image is a pixel value of each pixel of the first original image corresponding to each pixel of the second original image, performs a second binarization process on the second original image to generate a second binary image that distinguishes a foreground region composed of some of the aggregated particles from a background region outside the foreground region, an image processing system.

2. wherein the at least one processor, performs the first binarization process based on a mixture Gaussian model, performs the second binarization process based on the mixture Gaussian model, the image processing system according to claim 1.

3. wherein the at least one processor, generates a partial image from the first binary image, the partial image being a part of the first binary image and showing some of the plurality of aggregated particles, converts a pixel value of each pixel of the partial image into a pixel value of each pixel of the first original image corresponding to each pixel of the partial image to generate the second original image, the image processing system according to claim 1 or 2.

4. wherein the at least one processor generates the second original image showing one of the plurality of aggregated particles as some of the aggregated particles, the image processing system according to claim 1 or 2.

5. For each of the partial aggregated particles in the second binary image, the at least one processor calculates an average particle size of the plurality of primary particles forming the aggregated particle based on a perimeter length of the aggregated particle. The image processing system according to claim 1 or 2.

6. The first original image is an image obtained by imaging with a scanning electron microscope. The image processing system according to claim 1 or 2.

7. An image processing method executed by an image processing system including at least one processor, obtaining a first original image showing a plurality of aggregated particles respectively formed by aggregation of a plurality of primary particles; performing a first binarization process on the first original image to generate a first binary image that distinguishes a foreground region composed of the plurality of aggregated particles from a background region outside the foreground region; generating a second original image showing some of the plurality of aggregated particles based on the first original image and the first binary image, wherein a pixel value of each pixel of the second original image is a pixel value of each pixel of the first original image corresponding to the each pixel of the second original image; performing a second binarization process on the second original image to generate a second binary image that distinguishes a foreground region composed of the some of the aggregated particles from a background region outside the foreground region; An image processing method including the above.

8. obtaining a first original image showing a plurality of aggregated particles respectively formed by aggregation of a plurality of primary particles; performing a first binarization process on the first original image to generate a first binary image that distinguishes a foreground region composed of the plurality of aggregated particles from a background region outside the foreground region; Based on the first original image and the first binary image, generating a second original image showing some of the plurality of aggregated particles, wherein the pixel value of each pixel of the second original image is the pixel value of each pixel of the first original image corresponding to each pixel of the second original image; Executing a second binarization process on the second original image to generate a second binary image that distinguishes a foreground region composed of some of the aggregated particles from a background region outside the foreground region; An image processing program for causing a computer to execute.

Images