JP2020094852A - Particle image analyzer, particle image analysis method, and particle image analysis program - Google Patents

Abstract

To shorten the time needed for reanalysis and improve usability.SOLUTION: Provided is a particle image analyzer 10 for analyzing a particle image in which particles of powder are imaged, comprising: an analysis mode selection unit 33 for selecting an analysis mode from a regular analysis mode and a high-speed analysis mode; and an analysis object image selection unit 34 for selecting the object of image analysis from all of the particle images in accordance with the analysis mode. The analysis object image selection unit 34 selects all of the particle images as the object of image analysis when the analysis mode is the regular analysis mode, and selects some of all of the particle images as the object of image analysis when the analysis mode is the high-speed analysis mode.SELECTED DRAWING: Figure 2

Description

The present invention relates to a particle image analysis device, a particle image analysis method, and a particle image analysis program.

Various methods for analyzing the particle properties of industrial product powders such as pigments, cosmetic powders, toners, particulate catalysts, abrasives, powdered medicines, synthetic resin powders, and fine ceramic particles are known. As one of these methods, a method is known in which a particle image obtained by capturing an image of a particle is subjected to image analysis and the particle property is analyzed based on the result of the image analysis (for example, refer to Patent Document 1). In image analysis, in general, a particle image is cut out from a particle image and the particle property is obtained based on the particle image.

JP, 2003-156427, A

Generally, in one analysis, a large number of particle images are image-analyzed, and the particle property analysis results are obtained based on the respective analysis results. As the number of particle images increases, the analysis results are obtained. Will take longer. Further, when the data amount of the particle image is increased due to higher resolution and the like, the time for image analysis processing per particle image also increases. Due to these factors, it may take a huge amount of time to obtain an analysis result in one analysis.

By the way, in powder analysis, various conditions such as the conditions for cutting out a particle image from a particle image are appropriately changed to obtain a plurality of analysis results. What is happening is to search for certain conditions.
Every time the analysis is performed again under different conditions, the image analysis for the particle image is performed again, but if a long processing time is required for the second image analysis (hereinafter referred to as "reanalysis"), the analysis result can be obtained. It takes an enormous amount of time, and there is a problem in terms of usability.

An object of the present invention is to provide a particle image analysis device, a particle image analysis method, and a particle image analysis program that can shorten the time required for reanalysis and improve usability.

A first aspect of the present invention is a particle image analysis device for performing image analysis of a particle image in which particles of powder are reflected, and an analysis mode selection for selecting one of a first analysis mode and a second analysis mode. And an analysis target image selection unit that selects a target of the image analysis from all the particle images according to the analysis mode, the analysis target image selection unit is configured such that the analysis mode is the first analysis mode. Is selected, all the particle images are selected as the target of the image analysis, and when the analysis mode is the second analysis mode, a part of all the particle images is selected as the target of the image analysis. It is characterized by:

2nd invention is provided with the analysis part which calculates|requires the evaluation value of the some analysis parameter which shows the property of the said particle|grains by the said image analysis in 1st invention, Comprising: The said analysis object image selection part WHEREIN: The said analysis mode is the said In the case of the 2 analysis mode, a predetermined analysis parameter of interest is determined from the analysis parameters, and the particle image is selected so that there is no bias in the evaluation value obtained in the first analysis mode for the analysis parameter of interest. It is characterized by doing.

In a third aspect based on the second aspect, when the analysis mode is the second analysis mode, the analysis target image selection unit selects from among the respective analysis parameters obtained in the first analysis mode. It is characterized in that the analysis parameter of interest is determined based on the variation of the evaluation value.

4th invention is a 2nd or 3rd invention, When the said analysis mode is the said 2nd analysis mode, the said analysis object image selection part makes the said particle image the deviation of the evaluation value of the said analysis parameter of interest. The particle images are classified based on the classification, and the particle image is selected from each classification destination.

5th invention is a any one of 2nd-4th invention, Comprising: The particle image cut-out part which cuts out a particle image from the said particle image is provided, The said analysis part is a plurality of said image analysis of the said particle image. It is characterized in that the evaluation value of the analysis parameter is obtained, and the user can change the analysis processing condition that affects the extraction result of the particle image.

A sixth invention is the fifth invention, wherein the particle image cutout part cuts out a particle image from the particle image based on particle image cutout threshold data indicating a threshold value of a region of the particle image in the particle image, The analysis processing condition includes at least a threshold value indicated by the particle image cutout threshold value data.

In a seventh aspect based on the fifth or sixth aspect, an image correction unit that performs an image correction process on the particle image to be analyzed before cutting out the particle image is provided, and the analysis processing condition is at least the image correction. It is characterized in that it includes a correction value for processing.

An eighth aspect of the present invention is a particle image analysis method that causes a computer to perform an image analysis of a particle image in which particles of powder are reflected, and a user can select one of the first analysis mode and the second analysis mode. A first step for selecting, a second step for selecting an object of the image analysis from all the particle images according to the analysis mode, and a third step for performing image analysis of each of the particle images selected in the second step. In the second step, when the analysis mode is the first analysis mode, all the particle images are selected as targets of the image analysis, and the analysis mode is the second analysis mode. In this case, a part of all the particle images is selected as a target of the image analysis.

A ninth aspect of the present invention is a computer for performing image analysis of a particle image in which particles of powder are imaged, an analysis mode selection unit for selecting one of the first analysis mode and the second analysis mode, and all of them. The target of image analysis from the particle image of the function as an analysis target image selection unit for selecting according to the analysis mode, the analysis target image selection unit, if the analysis mode is the first analysis mode, When the particle image is selected as a target of the image analysis and the analysis mode is the second analysis mode, a part of all the particle images is selected as a target of the image analysis. A particle image analysis program for

According to the first aspect, in the second analysis mode, the number of particle images to be image-analyzed is reduced, so that the image analysis processing time can be shortened as compared with the first analysis mode. As a result, the user selects the second analysis mode during reanalysis, thereby improving the usability during reanalysis.
According to the second aspect of the invention, in the second analysis mode, a group of particle images whose evaluation values are not biased with respect to the analysis parameter of interest is targeted for image analysis. The result of accurate image analysis can be eliminated.
According to the third invention, in the second analysis mode, the analysis parameter of interest is determined based on the magnitude of the variation of the evaluation value obtained in the first analysis mode. With this second analysis mode, accurate image analysis results can be obtained in a short time for analysis parameters with large variations.
According to the fourth aspect, in the second analysis mode, a particle image to be image-analyzed is selected from each deviation, and a group of particle images having no deviation in the evaluation value of the analysis parameter of interest can be reliably formed.
According to the fifth aspect, in the second analysis mode, the user can accurately perform the image analysis in a short time while changing the analysis processing condition and changing the cutout result of the particle image. As a result, it is possible to accurately and quickly search for an analysis processing condition in which the variation in the evaluation value of the parameter of interest falls within an appropriate range.
In a sixth aspect of the invention, in the second analysis mode, the user executes the image analysis while changing the threshold of the particle image cut-out threshold data as the analysis processing condition, and the variation in the evaluation value of the parameter of interest is appropriately within the appropriate range. It is possible to accurately search for a threshold value in a short time.
According to the seventh aspect, in the second analysis mode, the user executes the image analysis while changing the correction value of the image correction processing as the analysis processing condition, and the variation in the evaluation value of the parameter of interest is appropriately within the appropriate range. It is possible to find accurate correction values accurately and in a short time.
According to the eighth and ninth inventions, as in the first invention, it is possible to improve usability during reanalysis by selecting the second analysis mode by the user during reanalysis.

It is a figure which shows the structure of the particle property analysis system which concerns on embodiment of this invention. It is a figure which shows the functional structure of a particle image analysis apparatus. It is a flowchart which shows the normal analysis mode process in a particle image analysis apparatus. It is a flow chart which shows high-speed reanalysis mode processing in a particle image analysis device. It is a figure which shows an example of the variation of an analysis parameter. It is a figure which shows an example of the number distribution of the particle image for every deviation.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a diagram showing the configuration of a particle property analysis system 1 according to this embodiment.
The particle property analysis system 1 is a system that performs an image analysis on a particle image 4 showing particles of an industrial product powder and analyzes the particle properties of the powder based on the result of the image analysis. As shown in FIG. 1, the particle property analysis system 1 of the present embodiment includes a particle image analysis device 10 and an image input unit 12 that inputs the particle image 4 to the particle image analysis device 10.

The image input unit 12 includes an imaging camera 14 that images the sample powder 6 in real time, and inputs the imaging data of the imaging camera 14 as a particle image 4 to the particle image analysis device 10.
The particle property analysis system 1 of the present embodiment analyzes the particle property of the sample powder 6 using a dynamic image analysis method. That is, the image input unit 12 continuously photographs the particles in the fluid in which the sample powder 6 is dispersed by the image pickup camera 14, and sequentially inputs the image pickup data obtained by the image pickup as the particle image 4 to the particle image analyzer 10. To do.

A static image analysis method may be used instead of the dynamic image analysis method. In this case, the imaging camera 14 images the range observed by an observation device such as an optical microscope or a scanning electron microscope that observes particles, and inputs the imaged data as the particle image 4 to the particle image analysis device 10.
Further, the image input unit 12 includes a scanner device for capturing an image of the sample powder 6 output on a paper such as a photograph, and inputs the image data captured by the scanner device as the particle image 4 to the particle image analysis device 10. You can also

FIG. 2 is a diagram showing a functional configuration of the particle image analysis device 10.
As shown in FIG. 2, the particle image analysis device 10 includes an analysis unit 20, an analysis unit 21, a display unit 22, and an operation unit 24. The particle image analysis device 10 includes a computer. A computer is an electronic device including a processor such as a CPU or MPU, a memory device such as a ROM or a RAM, a storage device such as an HDD or an SSD, and an interface circuit for connecting peripheral devices. Various functions shown in FIG. 2 are realized by executing a program (particle image analysis program) stored in the memory device or the storage device.

The analysis unit 20 performs image analysis of the particle image 5 reflected in the particle image 4, outputs the result of the image analysis to the analysis unit 21, and the analysis unit 21 analyzes the particle property based on the result of the image analysis. The details of image analysis and particle property analysis will be described later.

The display unit 22 displays various kinds of information such as an analysis result of particle properties, and includes an appropriate display device such as a flat panel display.
The operation unit 24 is used by the user to input various instructions and settings, and is provided with appropriate input devices such as operation buttons, a touch panel, a keyboard, and a pointing device.

In the particle image analysis device 10, the analysis unit 20, the analysis unit 21, the display unit 22, and the operation unit 24 may be housed in the same device casing and integrally configured, or may be appropriately configured. The part may be configured separately.

Next, the analysis unit 20 and the analysis unit 21 will be described in detail.

The analysis unit 20 of the present embodiment is provided with two analysis modes, a normal analysis mode (first analysis mode) and a high-speed reanalysis mode (second analysis mode), which can be selected by the user.
The normal analysis mode is a mode in which image analysis is performed on all particle images 4 to be analyzed, and is used at least during the first analysis.
The high-speed re-analysis mode is a mode that can be selected at the time of re-analysis. Instead of performing image analysis of all particle images 4 to be analyzed, the particle images 4 that meet a predetermined condition are targeted. This is a mode for executing image analysis. In this high-speed reanalysis mode, the number of particle images 4 to be image-analyzed is smaller than in the normal analysis mode, so the time until the image analysis processing is completed is shorter than in the normal analysis mode.

As shown in FIG. 2, the analysis unit 20 includes an image acquisition unit 30, an image storage unit 32, an analysis mode selection unit 33, an analysis target image selection unit 34, an image correction unit 36, and a particle image cutout. A unit 38 and an analysis calculation unit 40 are provided.

The image acquisition unit 30 acquires the particle image 4 from the image input unit 12 through an interface circuit included in the computer, and outputs the particle image 4 to the image storage unit 32. The image acquisition unit 30 is not limited to the image input unit 12, and may acquire the particle image 4 from another computer device via a recording medium such as a semiconductor memory or an optical disk, or an electric communication line.

The image storage unit 32 stores the particle images 4 for the number of sheets used for at least one analysis in the above-mentioned memory device or storage device.

The analysis mode selection unit 33 receives from the operation unit 24 a user selection operation for selecting one of the normal analysis mode and the fast reanalysis mode.
The analysis target image selection unit 34 selects the analysis target particle image 4 from the images stored in the image storage unit 32 according to the analysis mode selected by the user, and outputs the particle image 4 to the image correction unit 36.
Specifically, when the analysis mode is the normal analysis mode, the analysis target image selection unit 34 selects all the particle images 4 accumulated for the analysis, and the image correction unit 36 selects those particle images 4 from each other. Output to.
On the other hand, when the analysis mode is the high speed reanalysis mode, the analysis target image selection unit 34 selects the particle images 4 based on a predetermined condition and outputs the particle images 4 to the image correction unit 36. In the present embodiment, the predetermined condition is a condition set based on the analysis result in the normal analysis mode, and the details will be described later.

The image correction unit 36 performs image correction processing on the particle image 4 and outputs the image to the particle image cutout unit 38. This image correction process is a correction process that affects the appearance of the particle image 5 (FIG. 1) in the particle image 4 (that is, the clarity of the particle image 5). There are also brightness adjustment and contour correction.
The particle image cutout unit 38 cuts out the particle image 5 from the particle image 4 after the image correction processing based on the particle image cutout threshold value data 39, and outputs the particle image 5 to the analysis calculation unit 40. The particle image cut-out threshold value data 39 is data indicating a threshold value for dividing the area corresponding to the particle image 5 and the other area (so-called background area) in the particle image 4. As the threshold value, for example, the brightness value or chromaticity of the pixel of the particle image 4 or a combination thereof is used. As a method of cutting out the particle image 5 from the particle image 4 based on the particle image cutting threshold value data 39, an appropriate image processing technique such as contour detection can be used.

Here, the image correction unit 36 stores in advance, in the memory device or the storage device, the correction value data 37 indicating the initial value of the correction value of each of the image correction items, and the image correction processing is performed based on the correction value data 37. Execute. Similarly, the particle image cutout unit 38 also stores the initial value of the particle image cutout threshold data 39 in the memory device or the storage device in advance.
When the values of the correction value data 37 and the particle image cutout threshold value data 39 change, the clarity of the particle image 5 by the image correction processing and the cutout of the particle image 5 by the particle image cutout portion 38 are affected. become. That is, the values of the correction value data 37 and the particle image cutout threshold value data 39 can be said to be analysis processing conditions that affect the cutout result of the particle image 5.
In the present embodiment, each correction value of the correction value data 37 and the particle image cutout threshold data 39 can be changed by the user by operating the operation unit 24, and the user can correct the correction value data 37 and the particle image. By appropriately changing the cut-out threshold data 39, image correction processing is performed on the particle image 4 with an appropriate correction value, and a region of the particle image 4 that is regarded as the particle image 5 is appropriately changed to perform analysis. It can be executed repeatedly.

The analysis calculation unit 40 performs image analysis of the particle image 5, obtains evaluation values of various analysis parameters indicating particle properties, and outputs the evaluation values of each analysis parameter to the analysis unit 21 as a result of image analysis.
In the present embodiment, the analysis calculation unit 40 obtains the evaluation value of the analysis parameter related to the evaluation of the particle shape. Such analysis parameters include circle equivalent diameter, area, maximum length (major axis), maximum length vertical length (minor axis), perimeter, envelope perimeter, circumscribing rectangle major axis, circumscribing rectangle minor axis, circularity, aspect ratio, unevenness. There are degrees.
The equivalent circle diameter is the diameter of a circle having an area equal to the projected area of the particle, and the area is the projected area of the particle. Also, the maximum length (major axis) is the maximum distance between two points on the projected contour line of the particle, and the maximum length vertical length (minor axis) is when the particle is sandwiched by two straight lines parallel to the maximum length (major axis). Is the distance. The perimeter is the length of the projected contour line of the particle, and the envelope perimeter is the length of the circumference connecting the convex parts of the particle in the shortest distance. The circumscribed rectangle major axis is the length of the long side of the smallest rectangle that can contain the contour of the particle, and the circumscribed rectangle minor axis is the length of the short side of the smallest rectangle that can contain the contour of the particle. The circularity is calculated by the square of 4π×projected area of particles/perimeter, the aspect ratio is calculated by the maximum length (major axis)/maximum vertical length (minor axis), and the irregularity is the envelope perimeter/perimeter. Required by the chief.

The analysis unit 21 statistically processes the evaluation value of the analysis parameter (result of image analysis) obtained for each particle image 5 using an appropriate statistical method to generate various statistical graphs. These statistical graphs are displayed on the display unit 22 and displayed as analysis results.
In addition to the statistical graph, the analysis unit 21 may generate appropriate data (for example, image data of the particle image 5) as an analysis result and display it on the display unit 22.

Next, as the operation of the particle image analysis device 10, the analysis operation in each of the normal analysis mode and the fast reanalysis mode will be described. In the description of this operation, it is assumed that all the particle images 4 used in the analysis have already been input from the image input unit 12 and accumulated in the image accumulation unit 32 in the particle image analysis device 10.

FIG. 3 is a flowchart of the normal analysis mode process.
As described above, the normal analysis mode is an analysis mode executed at least during the first analysis. When the analysis mode selection unit 33 receives a user selection operation in the normal analysis mode as the analysis mode (step Sa1), the analysis target image selection unit 34 analyzes all the particle images 4 for the analysis from the image storage unit 32. And sequentially output to the image correction unit 36 (step Sa2). The image correction unit 36 performs an image correction process on the particle image 4 in accordance with the correction value data 37, and outputs it to the particle image cutout unit 38 (step Sa3). Every time the particle image 4 is input, the particle image cutout unit 38 cuts out all the particle images 5 shown in the particle image 4 based on the particle image cutout threshold data 39, and obtains the respective particle images 5. The data is output to the analysis calculation unit 40 (step Sa4). Every time the particle image 5 is input, the analysis operation unit 40 obtains the evaluation value of the analysis parameter indicating the particle property by image analysis of the particle image 5 and outputs the evaluation value of the analysis parameter to the analysis unit 21 (step Sa5). Then, the analysis unit 21 statistically processes the evaluation value of the analysis parameter for each particle image 5 to analyze the particle property, and outputs the analysis result to the display unit 22 (step Sa6).

After the analysis in the normal analysis mode, when the user changes the correction value data 37 of the image correction processing for the particle image 4 and the particle image cutout threshold value data 39 to change the cutting result of the particle image 5 and performs the analysis again, In addition to the normal analysis mode described above, a high-speed reanalysis mode can be selected and executed.

FIG. 4 is a flowchart of the fast reanalysis mode process. In the figure, the same processes as those in FIG. 3 are designated by the same reference numerals, and the description thereof will be omitted.
As shown in FIG. 4, when the analysis mode selection unit 33 receives a user selection operation of the high speed reanalysis mode as the analysis mode (step Sb1), the analysis target image selection unit 34 analyzes the particle image 4 corresponding to the predetermined condition. The target image is selected from the image storage unit 32 (step Sb2).

Specifically, in step Sb2, the analysis target image selection unit 34 determines one analysis parameter (hereinafter, referred to as “focused analysis parameter”) from various analysis parameters of the evaluation target obtained by image analysis. .. Then, a certain proportion of the particle images 4 are selected from all the particle images 4 so as not to cause a bias in the evaluation value of the analysis parameter of interest.

By such selection, the image analysis (reanalysis) is performed again for the group of particle images 4 in which the evaluation values of the analysis parameter of interest are not biased. As a result, it is possible to eliminate the bias of the evaluation values due to the group selected as the image analysis target and obtain the accurate image analysis result. Therefore, when the correction value data 37 of the image correction process and the particle image cutout threshold value data 39 when cutting out the particle image 5 are changed and re-analyzed, the correction value data 37 and the particle image cutout threshold value are changed. The change in the data 39 is accurately reflected in the change in the evaluation value of the analysis parameter of interest. As a result, the user can accurately search for the correction value data 37 and the particle image cutout threshold value data 39 that can obtain a proper evaluation value for the analysis parameter of interest, based on the change in the evaluation value at the time of reanalysis. Further, as described above, at the time of reanalysis, only a part of the particle images 4 is subjected to image analysis, so that the image analysis processing time at the time of reanalysis can be shortened, thereby improving the usability at the time of reanalysis. it can.

Here, in the present embodiment, the analysis target image selection unit 34 determines the analysis parameter having the largest variation in the evaluation value as the analysis parameter of interest in the analysis result obtained from the image analysis result in the normal analysis mode. .. Further, the analysis target image selection unit 34 evaluates the variation of each analysis parameter based on the CV (coefficient of variation) value. For example, as shown in FIG. 5, each analysis parameter of “equivalent circle diameter”, “area”, and “circularity” is obtained by image analysis, and the CV value of each analysis parameter is “7”, “5”. , And “3”. The larger the CV value, the larger the variation in the evaluation value for the analysis parameter. Therefore, in the illustrated example, the analysis target image selection unit 34 determines the “circle equivalent diameter” as the analysis parameter of interest.

Further, in the present embodiment, the analysis target image selection unit 34 selects a group of particle images 4 that does not cause bias in the evaluation value of the analysis parameter of interest based on the deviation of the analysis parameter of interest of the particle image 4.
Specifically, the analysis target image selection unit 34 calculates the deviations of the analysis parameters of interest for all the particle images 4 and, as shown in FIG. 6, sets the particle images 4 in a predetermined deviation range A1, A2, A3. .. are classified, and for each predetermined range A1, A2, A3,... Of the classification destination, a certain percentage (50% in the illustrated example) of particle images 4 is selected from all the particle images 4 belonging thereto. In this case, the deviation of the analysis parameter of interest is obtained by the difference between the evaluation value (image analysis result) of the analysis parameter of interest for the particle image 5 shown in the particle image 4 and the average value of the analysis parameter of interest. .. In this case, when a plurality of particle images 5 are shown in one particle image 4, the average value of the deviations of the particle images 5 is the deviation of the particle image 4.

As described above, at the time of reanalysis, the particle image 4 is selected from each deviation (a predetermined range of deviation in the present embodiment), thereby forming a group of particle images 4 in which the evaluation value of the analysis parameter of interest is not biased. .. Further, by setting the number of particle images 4 selected from each deviation to be a fixed ratio of the total number of particle images 4 belonging to the deviation, the number of particle images 4 to be image-analyzed can be reduced while suppressing the deviation of the evaluation value. Therefore, the image analysis processing time at the time of re-analysis can be efficiently shortened.

Returning to FIG. 4 above, after the particle image 4 is selected in the high-speed reanalysis mode as described above (step Sb1), the image correction process in the image correction unit 36 (step Sa2) as in the normal analysis mode. The cutout process of the particle image 5 in the particle image cutout unit 38 (step Sa3), the image analysis process in the analysis calculation unit 40 (step Sa4), and the analysis process of the particle property in the analysis unit 21 (step Sa5) are sequentially performed. To be executed.

As described above, in the fast reanalysis mode, an analysis parameter having a large variation in the image analysis result in the normal analysis mode is set as a focused analysis parameter, and a group of particle images 4 whose evaluation values are not biased with respect to the focused analysis parameter is set. The image analysis is performed again for the target. Through this re-analysis, the user can accurately correct the correction value data 37 and the particle image cut-out threshold data 39, which can obtain an evaluation value (image analysis result) in which the variation of the analysis parameter of interest falls within an appropriate range. You can find it in time. Then, the user can obtain the analysis result in which the variation of the evaluation value in each analysis parameter is suppressed by performing the image analysis in the normal analysis mode using the correction value data 37 and the particle image cutout threshold value data 39. it can. The correction value data 37 and the particle image cutout threshold value data 39 found in this way may be used as initial values and used in the next analysis.

According to this embodiment, the following effects are produced.

In the particle image analysis device 10 of the present embodiment, the analysis mode selection unit 33 that selects one of the normal analysis mode (first analysis mode) and the high-speed reanalysis mode (second analysis mode), and all of them. And an analysis target image selection unit 34 that selects an image analysis target from the particle image 4 according to the analysis mode. When the analysis mode is the normal analysis mode, the analysis target image selection unit 34 selects all the particle images 4 as the target of the image analysis, and when the analysis mode is the high speed reanalysis mode, all the particle images 4 are analyzed. Select a part of the above for image analysis.
In the high-speed reanalysis mode, the number of particle images 4 to be image-analyzed decreases, so the user selects this high-speed reanalysis mode during reanalysis, and the image analysis processing time during reanalysis is shorter than in the normal analysis mode. Therefore, usability at the time of reanalysis can be improved.

The particle image analysis apparatus 10 of the present embodiment includes an analysis unit 20 that obtains evaluation values of a plurality of analysis parameters indicating the properties of particles by image analysis. Then, when the analysis mode is the high speed reanalysis mode, the analysis target image selection unit 34 determines a predetermined analysis parameter of interest from the analysis parameters, and sets the evaluation value obtained for the analysis parameter of interest in the normal analysis mode. The particle image 4 is selected so that there is no bias.
As a result, in the high-speed reanalysis mode, the group of the particle images 4 having no bias in the evaluation values for the focused analysis parameter is the target of image analysis, so that the bias caused by the group selected as the image analysis target is eliminated and the The result of various image analysis can be obtained.

In the particle image analysis device 10 of the present exemplary embodiment, the analysis target image selection unit 34, when the analysis mode is the high-speed reanalysis mode, has a large variation in the evaluation value among the analysis parameters obtained in the normal analysis mode. Based on this, the analysis parameter of interest is determined.
With this high-speed reanalysis mode, reanalysis can be executed in a short time for analysis parameters with large variations in evaluation values, and accurate image analysis results can be obtained.

In the particle image analysis device 10 of the present exemplary embodiment, the analysis target image selection unit 34 classifies the particle image 4 based on the deviation of the evaluation value of the focused analysis parameter when the analysis mode is the high speed reanalysis mode, and each classification is performed. Select the particle image from the beginning.
As a result, at the time of reanalysis, the particle image 4 to be image-analyzed is selected from each deviation, and it is possible to reliably form a group of particle images 4 having no deviation in the evaluation value of the analysis parameter of interest.

The particle image analysis device 10 of the present embodiment includes a particle image cutout unit 38 that cuts out the particle image 5 from the particle image 4, and the analysis unit 20 obtains evaluation values of a plurality of analysis parameters by image analysis of the particle image 5. .. Then, the user can change the analysis processing condition that affects the extraction result of the particle image 5.
As a result, the user executes the above-mentioned re-analysis while changing the analysis processing condition and changing the cutout result of the particle image 5, and sets the appropriate analysis processing condition in which the variation in the evaluation value of the parameter of interest is within a reasonable range. , Can be searched accurately and in a short time.

In the particle image analysis device 10 of the present embodiment, the particle image cutout unit 38 extracts the particle image 5 from the particle image 4 based on the particle image cutout threshold data 39 indicating the threshold value of the region of the particle image 5 in the particle image 4. The cutout and analysis processing conditions include at least the threshold indicated by the particle image cutout threshold data 39.
Thereby, the user executes the above-mentioned re-analysis while changing the threshold of the particle image cut-out threshold data 39 as the analysis processing condition, and accurately and in a short time an appropriate threshold for suppressing the variation in the evaluation value of the parameter of interest. You can find out.

The particle image analysis apparatus 10 of the present embodiment includes an image correction unit 36 that performs image correction processing on the particle image 4 to be analyzed before cutting out the particle image 5, and the analysis processing condition includes at least a correction value of the image correction processing. I did it.
As a result, the user can perform the above-mentioned re-analysis while changing the correction value of the image correction processing as the analysis processing condition, and find an appropriate correction value that suppresses the variation of the evaluation value of the parameter of interest accurately and in a short time. You can

The above-described embodiment can be arbitrarily modified and applied without departing from the spirit of the present invention.

In the above-described embodiment, the correction value of the image correction processing in the image correction unit 36 is exemplified as the analysis processing condition that affects the extraction result of the particle image 5, and this can be changed by the user. However, the present invention is not limited to this, and various setting values used for cutting out the particle image 5 in the particle image cutting unit 38, an image processing algorithm, and other various conditions may be user-changeable as analysis processing conditions.

In the above-described embodiment, the analysis parameter of interest may be user-designable in the fast reanalysis mode.

In the above-described embodiment, the particle image analysis device 10 obtains the parameter relating to the shape as the property of the particle by image analysis, but not limited to this, the parameter relating to the property other than the shape may be obtained.

The functional blocks shown in FIG. 2 are schematic diagrams in which the constituent elements of the particle image analysis apparatus 10 are classified and shown according to the main processing contents in order to facilitate understanding of the present invention, and according to the processing contents. , And can be classified into more components. Also, one component can be classified so as to perform more processing.

DESCRIPTION OF SYMBOLS 1 Particle property analysis system 4 Particle image 5 Particle image 6 Sample powder 10 Particle image analyzer 12 Image input unit 20 Analysis unit 21 Analysis unit 24 Operation unit 33 Analysis mode selection unit 34 Analysis target image selection unit 36 Image correction unit 37 Correction Value data 38 Particle image cutout part 39 Particle image cutout threshold data 40 Analysis calculation part

Claims (9)

A particle image analysis device for image-analyzing a particle image showing particles of powder,
An analysis mode selection unit for selecting one of the first analysis mode and the second analysis mode;
An analysis target image selection unit that selects the target of the image analysis from all the particle images according to the analysis mode,
The analysis target image selection unit,
When the analysis mode is the first analysis mode, all the particle images are selected as targets of the image analysis,
When the analysis mode is the second analysis mode, a part of all the particle images is selected as a target of the image analysis,
A particle image analysis device characterized by the above. An analysis unit that obtains evaluation values of a plurality of analysis parameters indicating the properties of the particles by the image analysis,
The analysis target image selection unit, when the analysis mode is the second analysis mode,
From the analysis parameters to determine a predetermined analysis parameter of interest,
The particle image analysis apparatus according to claim 1, wherein the particle image is selected so that there is no bias in the evaluation value obtained in the first analysis mode for the focused analysis parameter. The analysis target image selection unit, when the analysis mode is the second analysis mode,
The particle image analysis apparatus according to claim 2, wherein the focused analysis parameter is determined based on the variation in the evaluation value from among the analysis parameters obtained in the first analysis mode. The analysis target image selection unit, when the analysis mode is the second analysis mode,
The particle image analysis apparatus according to claim 2 or 3, wherein the particle images are classified based on a deviation of an evaluation value of the focused analysis parameter, and the particle images are selected from each classification destination. A particle image cutout unit that cuts out a particle image from the particle image,
The analysis unit is
Obtaining evaluation values of the plurality of analysis parameters by image analysis of the particle image,
The user can change the analysis processing condition that affects the extraction result of the particle image.
The particle image analysis device according to any one of claims 2 to 4, wherein The particle image cutting section,
Cut out a particle image from the particle image based on the particle image cut-out threshold data indicating the threshold of the area of the particle image in the particle image,
The particle image analysis apparatus according to claim 5, wherein the analysis processing condition includes at least a threshold value indicated by the particle image cutout threshold value data. An image correction unit that performs an image correction process on the particle image to be analyzed before the particle image is cut out,
The particle image analysis apparatus according to claim 5, wherein the analysis processing condition includes at least a correction value of the image correction processing. A particle image analysis method for causing a computer to perform an image analysis of a particle image showing particles of powder,
A first step of allowing the user to select one of the first analysis mode and the second analysis mode;
A second step of selecting a target of the image analysis from all the particle images according to the analysis mode;
A third step of image-analyzing each of the particle images selected in the second step,
In the second step,
When the analysis mode is the first analysis mode, all the particle images are selected as targets of the image analysis,
When the analysis mode is the second analysis mode, a part of all the particle images is selected as a target of the image analysis,
A particle image analysis method characterized by the above. A computer for image analysis of particle images showing particles of powder,
An analysis mode selection unit that selects one of the first analysis mode and the second analysis mode, and
The target of the image analysis from all the particle images is caused to function as an analysis target image selection unit that selects according to the analysis mode,
The analysis target image selection section is
When the analysis mode is the first analysis mode, all the particle images are selected as targets of the image analysis,
When the analysis mode is the second analysis mode, a part of all the particle images is selected as a target of the image analysis,
A particle image analysis program characterized by the above.

Images