Calculation method of neoplasm index for characterizing satellitization particles

Abstract

The invention discloses a method for calculating a neoplasm index for representing satellitization particles, and relates to the technical field of particle image processing. The neoplasm index calculation method for characterizing satellitization particles comprises the following steps: preprocessing a plurality of satellitization multi-particle images; performing image segmentation on each satellite multi-particle image to obtain a plurality of single-particle images; calculating the maximum inscribed circle of the single particle; covering and cutting off the single particle of each single particle image by using the expanded maximum inscribed circle so as to extract a plurality of area images of each single particle image; removing the regional image with the area smaller than the lowest resolution; identifying the neoplasm according to the irregularity of the regional image; counting the number of neoplasm of each single particle image, and calculating neoplasm index distribution of different particles. According to the method, the satellite particle degree of the satellite particle image is represented by using the neoplasm index, the morphology characteristics of the particles are accurately reflected, and a basis is provided for improving the powder quality and optimizing the preparation process.

Description

Technical Field

[0001] The present invention relates to the technical field of particle image processing, and in particular to a method for calculating a vegetation index for characterizing satellite particles. Background Art

[0002] Satellite particles are particles in which one or more smaller particles attach to the surface of a larger particle. For example, in the preparation of metal 3D printing powder, a high-speed gas flow breaks up a stream of molten metal into droplets of varying sizes. During the flight and solidification of these droplets, due to their faster motion and solidification rates, they collide with larger, partially solidified droplets. This leads to fusion between the large and small droplets, resulting in small particles adhering to the surface of the larger particle. After the large droplets fully solidify, this attachment structure remains, forming satellite particles. This can reduce the quality and processing properties of metal powders, affecting their sphericity and flowability, and negatively impacting applications such as metal additive manufacturing. Therefore, quantitative analysis of satellite particles is an urgent challenge in industry. To quantitatively analyze satellite particles, the concept of the Vegetation Index has been introduced. The Vegetation Index is a key metric used in materials science to assess the degree of satellite formation in powders, particularly in advanced manufacturing technologies such as powder metallurgy and 3D printing. The vegetation index can quantify the number of satellite particles attached to the particle surface. Currently, it is necessary to design a vegetation index calculation method to characterize satellite particles, so as to characterize the degree of satellite particles based on satellite particle images, accurately reflect the morphological characteristics of the particles, and provide a basis for improving powder quality and optimizing the preparation process. Summary of the Invention

[0003] The technical problem to be solved by the present invention is how to use the vegetation index to characterize the degree of satellite particles in satellite particle images, accurately reflect the morphological characteristics of the particles, and provide a basis for improving powder quality and optimizing the preparation process. The purpose is to provide a method for calculating the vegetation index for characterizing satellite particles to solve the above technical problems.

[0004] A method for calculating a vegetation index for characterizing satellite particles comprises the following steps:

[0005] Preprocess multiple satellite multi-particle images;

[0006] performing image segmentation on each of the satelliteized multi-particle images to obtain a plurality of single-particle images;

[0007] Calculate the maximum inscribed circle of each single particle in the single particle image;

[0008] Using the enlarged maximum inscribed circle to cover and cut off the single particle in each of the single particle images, so as to extract the remaining multiple region images of each of the single particle images;

[0009] Remove the above-mentioned regional images with an area smaller than the minimum resolution;

[0010] identifying vegetation according to the irregularity of the image of each of the above regions;

[0011] The number of vegetation in each of the above single particle images is counted, and the vegetation index distribution of different particles is calculated.

[0012] The preprocessing of the plurality of satelliteized multi-particle images includes: performing binarization processing on the plurality of satelliteized multi-particle images, and removing particles connected to the image boundaries.

[0013] The step of performing image segmentation on each of the satelliteized multi-particle images to obtain a plurality of single-particle images includes: performing image segmentation on each of the satelliteized multi-particle images using a regular circumscribed rectangle of each particle to obtain a single-particle image.

[0014] The calculating of the maximum inscribed circle of each single particle in the single particle image includes: calculating the maximum inscribed circle of each single particle in the single particle image using the particle contour extracted by an edge detection algorithm.

[0015] The above-mentioned method of using the enlarged maximum inscribed circle to cover and cut off the single particle in each of the above-mentioned single particle images to extract the multiple remaining regional images of each of the above-mentioned single particle images includes: using the center of the maximum inscribed circle as the origin, covering and cutting off the single particle image with the enlarged maximum inscribed circle, and extracting the multiple regional images remaining after the covering and cutting off.

[0016] Before removing the regional images with an area smaller than the minimum resolution, the method further includes: comparing particle parameters of the regional images at different resolutions, and taking the resolution at which the particle parameters have obvious differences as the minimum resolution.

[0017] The above-mentioned identification of vegetation based on the irregularity of each of the above-mentioned regional images includes: calculating the maximum inscribed circle and the minimum circumscribed circle of each of the above-mentioned regional images, and calculating the irregularity of each of the above-mentioned regional images; judging whether the irregularity of each of the above-mentioned regional images meets the preset indicators of vegetation; when satisfied, determining that the regional image is the above-mentioned vegetation.

[0018] An electronic device includes a memory, a processor, and a computer program running on the processor. When the processor executes the computer program, the steps of the above-mentioned method for calculating a vegetation index for characterizing satellite particles are implemented.

[0019] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for calculating a vegetation index for characterizing satellite particles.

[0020] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0021] The present invention provides a method for calculating a vegetation index for characterizing satellite particles, comprising: preprocessing multiple satelliteized multi-particle images; segmenting each satelliteized multi-particle image to obtain multiple single-particle images; calculating the maximum inscribed circle of a single particle in each single-particle image; using the enlarged maximum inscribed circle to perform overlay excision on the single particle in each single-particle image to extract multiple regional images of each single-particle image; removing regional images with an area smaller than the minimum resolution; identifying vegetation based on the irregularity of each regional image; counting the number of vegetation in each single-particle image, and calculating the vegetation index distribution for different particles. The present invention can characterize the degree of satelliteization of particles in satellite particle images, accurately reflecting the morphological characteristics of the particles, and providing a basis for improving powder quality and optimizing the preparation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the examples. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be considered as limiting the scope. A person of ordinary skill in the art can also derive other relevant drawings based on these drawings without inventive effort. In the drawings:

[0023] Figure 1 This is a flow chart of a method for calculating a vegetation index for characterizing satellite particles according to an embodiment of the present application;

[0024] Figure 2 Schematic diagram of a satellite-based multi-particle image according to an embodiment of the present application;

[0025] Figure 3 For the embodiment of this application Figure 2 Multiple single particle images after segmentation;

[0026] Figure 4 For the embodiment of this application Figure 3 The 24th particle covers the image after resection;

[0027] Figure 5 These are regional images at different resolutions according to the embodiment of the present application;

[0028] Figure 6 This is a distribution diagram of the vegetation index of 5 different types of nickel-based alloy particles in the examples of this application. DETAILED DESCRIPTION

[0029] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with examples and drawings. The exemplary embodiments of the present invention and their descriptions are only used to explain the present invention and are not intended to limit the present invention.

[0030] Example

[0031] like Figure 1 As shown, the embodiment of the present application provides a method for calculating the vegetation index for characterizing satellite particles, comprising the following steps:

[0032] Preprocess multiple satellite multi-particle images;

[0033] performing image segmentation on each of the satelliteized multi-particle images to obtain a plurality of single-particle images;

[0034] Calculate the maximum inscribed circle of each single particle in the single particle image;

[0035] Performing overlay excision on the single particle in each of the single particle images using a maximum inscribed circle to extract multiple region images of each of the single particle images;

[0036] Remove the above-mentioned regional images with an area smaller than the minimum resolution;

[0037] identifying vegetation according to the irregularity of the image of each of the above regions;

[0038] The number of vegetation in each of the above single particle images is counted, and the vegetation index distribution of different particles is calculated.

[0039] The preprocessing of the plurality of satelliteized multi-particle images includes: performing binarization processing on the plurality of satelliteized multi-particle images, and removing particles connected to the image boundaries.

[0040] The step of performing image segmentation on each of the satelliteized multi-particle images to obtain a plurality of single-particle images includes: performing image segmentation on each of the satelliteized multi-particle images using a regular circumscribed rectangle of each particle to obtain a single-particle image.

[0041] The calculating of the maximum inscribed circle of each single particle in the single particle image includes: calculating the maximum inscribed circle of each single particle in the single particle image using the particle contour extracted by an edge detection algorithm.

[0042] The above-mentioned method of using the enlarged maximum inscribed circle to cover and cut off the single particle in each of the above-mentioned single particle images to extract the multiple remaining regional images of each of the above-mentioned single particle images includes: using the center of the maximum inscribed circle as the origin, covering and cutting off the single particle image with the enlarged maximum inscribed circle, and extracting the multiple regional images remaining after the covering and cutting off.

[0043] Before removing the regional images with an area smaller than the minimum resolution, the method further includes: comparing particle parameters of the regional images at different resolutions, and taking the resolution at which the particle parameters have obvious differences as the minimum resolution.

[0044] The above-mentioned identification of vegetation based on the irregularity of each of the above-mentioned regional images includes: calculating the maximum inscribed circle and the minimum circumscribed circle of each of the above-mentioned regional images, and calculating the irregularity of each of the above-mentioned regional images; judging whether the irregularity of each of the above-mentioned regional images meets the preset indicators of vegetation; when satisfied, determining that the regional image is the above-mentioned vegetation.

[0045] Five types of 3D printing nickel-based alloy powders were selected for the experiment, including (a) Monel, (b) GH3230, (c) NiCocr, (d) NiCocr-Y, and (e) Fe-Ni. The collected multi-particle images were binarized to remove particles connected to the boundary, and the collected images were segmented using the regular circumscribed rectangle of the particles. Figure 2 Taking (a) Monel's multi-particle image as an example, the segmentation results are as follows Figure 3 Multiple single particle images are shown. Figure 3 Taking the 24th satellite particle in the image as an example, an edge detection algorithm is used to extract the particle outline and calculate the maximum inscribed circle diameter. The definition of satellite particles indicates that small particles attached to large particles are vegetation. To accurately extract vegetation, this embodiment uses a circle with the center of the maximum inscribed circle as the origin and a diameter 1.05 times the maximum inscribed circle to perform overlay excision on the original particle, as shown in the following example. Figure 4 As shown in the figure, small residual areas will result in low-pixel images, and low image resolution will cause deviations in the calculated particle feature parameters, thus affecting the subsequent judgment of vegetation. Figure 4 After the satellite particles are covered and removed, there are 4 remaining regional images. Taking one of the regional images as an example, its images at different resolutions are as follows: Figure 5 As shown, the selected characteristic parameters include the maximum inscribed circle diameter (x imax ), minimum circumscribed circle diameter (x cmin ), Irregularity, Circularity, EllipseRatio and AspectRatio.

[0046] Figure 5 The calculation results of images with different resolutions and their feature parameters are shown in Table 1 below.

[0047]

[0048] Table 1

[0049] As shown in Table 1, when the image resolution is less than 29×26 pixels, its feature parameters begin to differ significantly from those of high-resolution images. Therefore, to ensure the accuracy of the analysis results when determining vegetation, images of regions with an area less than 29×26 pixels after overlay resection are removed. Multiple images of the remaining regions after overlay resection are removed, and images of the remaining regions are extracted after removing the small regions. This embodiment of the present invention proposes a vegetation determination metric: an image with an irregularity greater than 0.12 in the remaining region indicates an attached vegetation. Figure 4 The final two extracted region images have irregularities of 0.81 and 0.65 respectively. Based on this indicator, both region images are vegetation with particles attached. The definition of irregularity is shown in the following equation (1):

[0050]

[0051] Among them, x imax is the maximum inscribed circle diameter, x cmin is the minimum circumscribed circle diameter.

[0052] The embodiment of the present invention first uses the center of the particle's largest inscribed circle as the origin, draws a circle slightly larger than 1.05 times the diameter of the largest inscribed circle, and removes the area less than 29×26 pixels to extract the remaining area image. Secondly, in the extracted area image, the area image is judged to be an outgrowth attached to the particle based on the vegetation judgment index: when the irregularity is greater than 0.12, the area image is judged to be an outgrowth attached to the particle. Finally, the number of outgrowths for each nickel-based alloy particle is counted, and the outgrowth index is calculated according to the definition. The outgrowth index (Og) is defined as shown in the following equation (2), and the value range is between 0 and 1:

[0053]

[0054] Where n represents the number of attached satellite particles. When n=0, it means that there are no small particles attached to the particle surface. The vegetation index O g is 1, when there are 1, 2, 3, 4, 5 and other satellite particles attached to the particle surface, then O g The values ​​are 0.5, 0.33, 0.25, 0.2 and 0.17 respectively, and so on. The smaller the n value, the better the O g The closer the value is to 1, the lower the degree of satelliteization and the closer the particles are to spherical. The larger the n value is, the smaller the O g The closer the value is to 0, the higher the degree of satelliteization and the more irregular the particles.

[0055] When measuring different types of nickel-based alloy particles, according to the proposed vegetation index calculation method, the vegetation index distribution diagram of each type of nickel-based alloy particles is obtained as shown in the figure: Figure 6 shown. Figure 6The horizontal axis is the vegetation index, and the vertical axis is the percentage of vegetation. The bar chart shows the distribution of vegetation counts for six types of particles: (a) Monel, (b) GH3230, (c) NiCocr, (d) NiCocr-Y, and (e) Fe-Ni. The vegetation index calculation formula (2) shows that Monel and GH3230 alloy particles have more vegetation attached to their surfaces, lower vegetation indices, and higher degrees of satellite formation. NiCocr, NiCocr-Y, and Fe-Ni alloy particles have higher vegetation indices, lower degrees of satellite formation, and are nearly spherical. The vegetation index of different nickel-based alloy particles varies significantly, which is related to their chemical composition and preparation process. A lower vegetation index indicates a higher degree of satellite formation on the particle surface, which may affect the powder's fluidity, adhesion, and other properties. The statistical results of the vegetation index are of great significance for optimizing the preparation process of metal alloy powders and improving their performance in industrial applications.

[0056] In summary, the present embodiment provides a method for calculating a vegetation index for characterizing satellite particles: the method preprocesses multiple satelliteized multi-particle images; performs image segmentation on each satellite particle image to obtain a single particle image; calculates the maximum inscribed circle of the single particle; uses the enlarged maximum inscribed circle to perform overlay excision on the single particle image to extract the remaining multiple regional images; removes regional images with an area smaller than the minimum resolution; identifies vegetation based on the irregularity of each regional image; counts the number of vegetation in each single particle image, and calculates the vegetation index distribution for different particles. This method can characterize the degree of satelliteization of particles in satellite particle images, accurately reflecting the particle morphological characteristics, and providing a basis for improving powder quality and optimizing production processes.

[0057] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for calculating the vegetation index for characterizing satellite particles, characterized in that: The steps include: Preprocess multiple satellite multi-particle images; performing image segmentation on each of the satelliteized multi-particle images to obtain a plurality of single-particle images; Calculating the maximum inscribed circle of each single particle in the single particle image; Using the enlarged maximum inscribed circle to cover and cut off the single particle in each of the single particle images, so as to extract a plurality of remaining region images of each of the single particle images; Removing the image of the region with an area smaller than the minimum resolution; identifying vegetation according to the irregularity of the image of each region; The number of vegetation in each single particle image is counted, and the vegetation index distribution of different particles is calculated.

2. The method for calculating the vegetation index for characterizing satellite particles according to claim 1, wherein: The preprocessing of the plurality of satelliteized multi-particle images includes: performing binarization processing on the plurality of satelliteized multi-particle images, and removing particles connected to the image boundaries.

3. The method for calculating the vegetation index for characterizing satellite particles according to claim 1, wherein: The performing image segmentation on each of the satelliteized multi-particle images to obtain a plurality of single-particle images includes: performing image segmentation on each of the satelliteized multi-particle images using a regular circumscribed rectangle of each particle to obtain a single-particle image.

4. The method for calculating the vegetation index for characterizing satellite particles according to claim 1, wherein: The calculating of the maximum inscribed circle of the single particle in each of the single particle images includes: calculating the maximum inscribed circle of the single particle in each of the single particle images using the particle contour extracted by an edge detection algorithm.

5. The method for calculating the vegetation index for characterizing satellite particles according to claim 1, wherein: The method of using the enlarged maximum inscribed circle to cover and cut off the single particle in each single particle image to extract the multiple remaining regional images of each single particle image includes: using the center of the maximum inscribed circle as the origin, covering and cutting off the single particle image with the enlarged maximum inscribed circle, and extracting the multiple remaining regional images after the covering and cutting off.

6. The method for calculating the vegetation index for characterizing satellite particles according to claim 1, wherein: Before removing the regional images with an area smaller than the lowest resolution, the method further includes: comparing particle parameters of the regional images at different resolutions, and taking the resolution at which the particle parameters have obvious differences as the lowest resolution.

7. The method for calculating the vegetation index for characterizing satellite particles according to claim 1, wherein: The method of identifying vegetation based on the irregularity of each of the regional images includes: calculating the maximum inscribed circle and the minimum circumscribed circle of each of the regional images, and calculating the irregularity of each of the regional images; judging whether the irregularity of each of the regional images meets the preset index of vegetation; and when the index is met, determining that the regional image is a vegetation.

8. An electronic device comprising a memory, a processor, and a computer program running on the processor, characterized in that: When the processor executes the computer program, the steps of the method for calculating the vegetation index for characterizing satellite particles according to any one of claims 1 to 7 are implemented.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for calculating a vegetation index for characterizing satellite particles according to any one of claims 1 to 7 are implemented.

Images