High-precision color master batch particle size distribution and surface appearance real-time image detection method and system

Abstract

The invention relates to the technical field of color master batch detection, in particular to a high-precision color master batch particle size distribution and surface topography real-time image detection method and a high-precision color master batch particle size distribution and surface topography real-time image detection system. Comprising the following steps: acquiring a marked overlapped color master batch end contour, repairing the overlapped color master batch end contour based on a contour repairing technology, acquiring a size parameter according to the repaired end shape, matching the size parameter with a preset color master batch projection parameter, and determining a projection line segment of the color master batch. Influences caused by overlapping and shielding of the color master batches in traditional image analysis are compensated through the contour repairing technology, the overlapped color master batches are distinguished, the particle sizes of the overlapped color master batches can be more accurately measured, the phenomenon that the overlapped area is misjudged as a single particle is reduced, statistics on the particle size distribution of the color master batches in the later period is facilitated, and the accuracy of the color master batch particle size analysis is improved. And the particle size distribution condition of the color master batch can be intuitively known by dividing the particle size of the color master batch and generating a particle size distribution scatter diagram.

Description

Technical Field

[0001] The present invention relates to the technical field of masterbatch detection, and specifically, to a real-time image detection method and system for high-precision particle size distribution and surface morphology of masterbatch. Background Art

[0002] Masterbatch is a common plastic masterbatch, mainly used to provide specific color effects and improve the performance of plastic products. The uniformity of the particle size of masterbatch affects the quality and performance of plastic products, as well as the processing efficiency of plastic products. Therefore, the detection of the particle size uniformity of masterbatch is a crucial step in the production process of masterbatch.

[0003] Currently, Chinese Patent Publication No. CN118794845A proposes a method and system for detecting the particle size uniformity of BPE masterbatch, including: obtaining the basic information of the particle size uniformity detection device and the tray mass and screen gray-scale image in each aperture adjustment experiment; calculating the mass of the dropped masterbatch according to the tray mass, and further obtaining the degree of mass increase; calculating the degree of increase in the adjusted particle size; obtaining the partition pulling distance according to the degree of increase in the adjusted particle size and the aperture of the screen mesh; further calculating the gear rotation angle; analyzing the masterbatch area of the screen gray-scale image; iteratively adjusting the screening aperture length until the masterbatch screening is completed; further determining the corrected screening aperture length; drawing a scatter curve of the aperture adjustment experiment to obtain the determination result of the masterbatch.

[0004] Since the masterbatch is cylindrical, when its bottom surface is vertically upward, its projection in the top view direction is circular, and when it is in a flat state, the projection of the masterbatch in the top view direction is rectangular, which are the projections in the normal state. On the contrary, when the cylindrical masterbatch is tilted, at this time, according to the different tilting angles of the masterbatch, the figures presented by its bottom surface and side surface in the top view perspective are different, and the bottom surface of the masterbatch no longer presents a complete circle. Due to the tilt, a part of the bottom surface will be "compressed" or "stretched", resulting in the projection contour presenting an ellipse.

[0005] In this way, when two masterbatches are in the states shown in Figure 3 、 Figure 4 , the upper masterbatch forms an occlusion (overlap) on the lower masterbatch, causing only a partial contour of the lower masterbatch to be exposed. In this case, when using image processing technology to statistically analyze the particle size distribution of the masterbatch, the overlapping area will be misjudged as a single particle, thus affecting the accuracy of the statistical analysis of the particle size of the masterbatch in this state. Summary of the Invention

[0006] The present invention provides a real-time image detection method for the particle size distribution and surface morphology of high-precision color masterbatches. Through contour repair technology, it compensates for the influence caused by the overlapping and occlusion of color masterbatches in traditional image analysis, distinguishes overlapping color masterbatches, enables the particle size of overlapping color masterbatches to be measured more precisely, reduces the phenomenon that the overlapping area is misjudged as a single particle, and facilitates the later statistics of the particle size distribution of color masterbatches, thus solving the problems raised in the above-mentioned background technology;

[0007] To achieve the above object, the real-time image detection method for the particle size distribution and surface morphology of high-precision color masterbatches includes:

[0008] S1. Through an optical device and based on microscope imaging technology, the real-time acquired color masterbatch image is transmitted to a processing unit, and the processing unit preprocesses the image;

[0009] S2. Obtain the preprocessed image, acquire the contours of the region-overlapping color masterbatches based on image analysis technology, and mark the region-overlapping color masterbatches;

[0010] S3. Obtain the end contours of the marked overlapping color masterbatches, repair the end contours of the overlapping color masterbatches based on contour repair technology, then obtain its size parameters according to the repaired end shape, and match the size parameters with the preset color masterbatch projection parameters to determine the projection line segments of the color masterbatches;

[0011] S4. Based on the matching result, measure the particle size of each color masterbatch, divide the particle size of the color masterbatch and generate a particle size distribution scatter plot.

[0012] In the above technical solution, mainly through contour repair technology, it compensates for the influence caused by the overlapping and occlusion of color masterbatches in traditional image analysis, distinguishes overlapping color masterbatches, enables the particle size of overlapping color masterbatches to be measured more precisely, reduces the phenomenon that the overlapping area is misjudged as a single particle, and facilitates the later statistics of the particle size distribution of color masterbatches. And by dividing the particle size of the color masterbatch and generating a particle size distribution scatter plot, the particle size distribution of the color masterbatch can be intuitively understood.

[0013] The second object of the invention is to provide a real-time image detection system for the particle size distribution and surface morphology of high-precision color masterbatches, including an image acquisition and processing module, a region-overlapping contour marking module, a projection line segment determination module for color masterbatches, and a particle size distribution display module for color masterbatches;

[0014] Including an image acquisition and processing module: Through an optical device and based on microscope imaging technology, the real-time acquired color masterbatch image is transmitted to a processing unit, and the processing unit preprocesses the image;

[0015] Region overlapping contour marking module: Obtain the preprocessed image, obtain the contours of the region-overlapping color masterbatches based on image analysis technology, and mark the color masterbatches with overlapping regions;

[0016] Projection line segment determination module of color masterbatch: Obtain the end contours of the marked overlapping color masterbatches, repair the end contours of the overlapping color masterbatches based on contour repair technology, then obtain their size parameters according to the repaired end shapes, match the size parameters with the preset color masterbatch projection parameters, and determine the projection line segments of the color masterbatches;

[0017] Color masterbatch particle size distribution display module: Based on the matching results, measure the particle sizes of each color masterbatch, divide the particle sizes of the color masterbatches and generate a scatter plot of the particle size distribution.

[0018] Compared with the prior art, the beneficial effects of the present invention are:

[0019] Use the curve contour of the non-overlapping part to fit the geometric model, map the missing part to one side to supplement the missing elliptical line segment at the end of the color masterbatch, and can accurately measure the end shape of each color masterbatch, and then obtain its size parameters. Based on the matching of the size parameters with the preset color masterbatch projection parameters, the projection line segments and their morphological contours of the color masterbatches can be quickly determined. The contour repair technology compensates for the influence caused by the overlapping and occlusion of the color masterbatches in traditional image analysis, distinguishes the overlapping color masterbatches, enables the overlapping color masterbatches to be more accurately measured for their particle sizes, reduces the phenomenon that the overlapping area is misjudged as a single particle, facilitates the later statistics of the particle size distribution of the color masterbatches, and by dividing the particle sizes of the color masterbatches and generating a scatter plot of the particle size distribution, the particle size distribution of the color masterbatches can be intuitively understood. This method can clearly display the particle size distribution characteristics of the color masterbatches, facilitating subsequent analysis and application. Brief Description of the Drawings

[0020] Figure 1 It is a flowchart of the method of the present invention;

[0021] Figure 2 It is a system block diagram of the present invention;

[0022] Figure 3 It is a diagram of the inclined states of the upper and lower color masterbatches of the present invention;

[0023] Figure 4 It is a diagram of the state where the lower color masterbatch is blocked by the upper color masterbatch of the present invention.

[0024] The meanings of the reference numerals in the figure are as follows:

[0025] 100. Image acquisition and processing module; 200. Region overlapping contour marking module; 300. Projection line segment determination module of color masterbatch; 400. Color masterbatch particle size distribution display module. Detailed implementation mode

[0026] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0027] In view of the fact that the masterbatch located at the upper end forms an occlusion (overlap) with the masterbatch at the lower end, so that only a partial contour of the lower masterbatch is exposed. In this way, when using image processing technology to statistically analyze the particle size distribution of the masterbatch, the overlapping area will be misjudged as a single particle, thus affecting the accuracy of the statistical analysis of the particle size of the masterbatch in this state. The present invention provides a high-precision real-time image detection method for the particle size distribution and surface morphology of masterbatch. See Figure 1 as shown, including the following method steps:

[0028] S1. Through an optical device and based on microscope imaging technology, the real-time acquired masterbatch image is transmitted to a processing unit, and the processing unit preprocesses the image;

[0029] First, when statistically analyzing the particle size distribution of masterbatch on a plane, it is necessary to first obtain the image of the masterbatch on the plane. That is, the preliminary preparation work is divided into a hardware unit, a transmission unit, and a processing unit. Among them, in the hardware unit, by equipping a high-magnification objective lens (such as a 100× oil immersion lens) and a CCD / CMOS camera, real-time imaging capture of the masterbatch is realized. Secondly, during the capture process, a ring-shaped LED light source or coaxial light can be used to ensure uniform illumination of the masterbatch surface and reduce shadow interference;

[0030] Then, the transmission unit converts the digital signal simulated by the real-time imaging of the masterbatch into a data stream recognizable by a computer, that is, the transmission unit imports the acquired image data into the computer for processing. Therefore, the processing unit is used to preprocess the image after receiving the image data, including the segmentation operation of the image. The segmentation operation converts the received image data into a grayscale image based on image edge detection technology, and then calculates the gradient amplitude of the grayscale value of the grayscale image respectively through the Sobel operator principle to detect the edge. Two 3×3 convolution kernels are used to detect the grayscale value gradients in the horizontal direction and the vertical direction respectively, and the grayscale value gradient in the horizontal direction is synthesized with the grayscale value gradient in the vertical direction to form a synthesized gradient amplitude value, which is compared with a preset amplitude value to retain the strong edges of the image, and the image is binarized through adaptive threshold segmentation (such as the Otsu algorithm) to highlight the masterbatch area.

[0031] Here, if the processing unit receives a color image, it needs to be converted into a grayscale image first. If the received image is already a grayscale image, it is directly skipped, as shown below:

[0032] The grayscale value gradient in the horizontal direction and the grayscale value gradient in the vertical direction are respectively H = G(I, G x ), Y = G(I, G y ), where I is the grayscale image, and G ≈ |G x | + |G y |. Assuming a simple 3×3 grayscale image I is H is Y is

[0033] The result is obtained through calculation: Compare G with a preset amplitude (assuming this value is 50). All values greater than 50 are retained. That is, the middle pixel value of the original image is 50, and the surrounding is 10. For the edge image, the edge of the middle pixel is detected, and the gradient amplitude value is 60. In this way, through the gradient calculation in the horizontal and vertical directions, the edges of the image are efficiently detected, providing a basis for the later analysis of the geometric characteristics of masterbatch particles (such as particle size, shape, and contour).

[0034] After the image processing is completed, the next step is carried out. S2: Obtain the preprocessed image, obtain the contours of the overlapping masterbatch particles based on image analysis technology, and mark the overlapping masterbatch particles; and this step is for the staggered overlapping masterbatch particles:

[0035] In S2, the steps to obtain the contours of the overlapping masterbatch particles based on image analysis technology are as follows:

[0036] S210: Obtain the image after the segmentation operation, extract the masterbatch particle contours according to the image, and extract the projection line segments in the contours based on the masterbatch particle contours;

[0037] S211: Establish a coordinate system with the horizontal direction of the image as the x-axis and the vertical direction as the y-axis, extract two parallel lines of the projection line segments of the same masterbatch particle contour, and respectively obtain the coordinates of the endpoints of the parallel line segments;

[0038] S212: Use the geometric method to calculate whether the projection line segments of two masterbatch particles intersect through the coordinates of the line segment endpoints to determine the coincidence situation of the masterbatch particles;

[0039] If the line segments intersect, it proves that the masterbatch particles overlap. If the line segments do not intersect, it proves that the masterbatch particles do not overlap;

[0040] S213: Mark the overlapping masterbatch particles.

[0041] After obtaining the image after the segmentation operation, a contour detection algorithm is used to extract the boundary contours of the color masterbatch. Among them, a single color masterbatch corresponds to a closed projected contour, while overlapping color masterbatches form a projected contour composed of the interlaced projected contours of the two. Suppose there are two overlapping color masterbatches, Color Masterbatch ID = 1 and Color Masterbatch ID = 2. First, extract the projected parallel lines of the color masterbatch and obtain the left coordinates of the endpoints of the parallel line segments:

[0042] For example: Color Masterbatch ID = 1: The endpoint coordinates of the extracted parallel line 1 are (0, 0) and (2, 0); the endpoint coordinates of parallel line 2 are (0, 1) and (2, 1);

[0043] Color Masterbatch ID = 2: The endpoint coordinates of the extracted parallel line 3 are (1, 0) and (1, 2); the endpoint coordinates of parallel line 4 are (2, 0) and (2, 2); Through analysis, it can be seen that parallel line 1 and parallel line 2 of Color Masterbatch ID = 1 intersect with parallel line 3 and parallel line 4 of Color Masterbatch ID = 2 on both the x-axis and the y-axis. Hereby, Color Masterbatch ID = 1 and Color Masterbatch ID = 2 are marked. Similarly, other overlapping color masterbatches are operated according to the same steps above.

[0044] Secondly, when the parallel lines of the overlapping color masterbatches intersect, only the overlapping area of the two color masterbatches can be determined, and it cannot be determined whether the parallel lines of the two color masterbatches are collinear. That is to say, in order to make the parallel lines of the two color masterbatches coincide with each other, the following conditions need to be met:

[0045] In S212, the size of the overlapping area of the color masterbatch is affected by the inclination state of the color masterbatch. Therefore, when determining the overlapping state of the color masterbatch, obtain the endpoint coordinates of the parallel lines of the color masterbatch, and based on the endpoint coordinates, use the vector parallelism judgment algorithm to determine whether the parallel lines of the two color masterbatches are collinear. The vector parallel conditions are as follows:

[0046] (x2 - x1)(y4 - y3) = (x4 - x3)(y2 - y1);

[0047] Among them, (x2 - x1, y2 - y1) and (x4 - x3, y4 - y3) are the direction vectors of the parallel lines of the two color masterbatches respectively;

[0048] The following is shown by an example:

[0049] For the line segment of Color Masterbatch ID = 1, its direction vector AB = (x2 - x1, y2 - y1) = (3 - 1, 4 - 2);

[0050] For the line segment of Color Masterbatch ID = 2, its direction vector CD = (x4 - x3, y4 - y3) = (7 - 5, 8 - 6);

[0051] That is, AB = (2, 2); CD = (2, 2); By calculation, it is concluded that the two line segments are parallel. Since only judging whether one of the parallel lines of the two masterbatches is collinear cannot meet the condition that the parallel lines of the two masterbatches coincide with each other. Therefore, based on the above conditions, the collinearity of the parallel lines of the masterbatches is judged respectively. When the two parallel lines of the overlapping masterbatches meet the conditions of being collinear respectively and the parallel lines intersecting, the parallel lines of the two masterbatches can be determined to coincide with each other.

[0052] However, the projection line segments of the two overlapping masterbatches cannot be recognized, which often causes the overlapping masterbatches to be ignored when counting the particle size distribution of the masterbatches, that is:

[0053] Since the masterbatch is cylindrical, when its bottom surface is vertically upward, its projection in the top view direction is circular, and when it is in a tiled state, the projection of the masterbatch in the top view direction is rectangular. These belong to the projections under normal conditions. On the contrary, when the cylindrical masterbatch is tilted, at this time, according to the different tilting angles of the masterbatch, the figures presented by its bottom surface and side surface in the top view are different, and the bottom surface of the masterbatch no longer presents a complete circle. Due to the tilt, a part of the bottom surface will be "compressed" or "stretched", resulting in the projection contour presenting an ellipse.

[0054] In this way, when the two masterbatches are in the states shown in Figure 3 、 Figure 4 , the masterbatch at the upper end blocks (overlaps) the masterbatch at the lower end, thus affecting the statistics of the particle size of the masterbatch in this state.

[0055] Therefore, after determining that the parallel lines of the two masterbatches coincide with each other, the next step S3 is carried out: Obtain the end contours of the marked overlapping masterbatches, repair the end contours of the overlapping masterbatches based on the contour repair technology, then obtain its dimension parameters according to the repaired end shape, and match the dimension parameters with the preset masterbatch projection parameters to determine the projection line segments of the masterbatch;

[0056] The preset masterbatch projection parameters are from the standard sample library of industry specifications, and its content includes the projection parameters of masterbatches with known sizes and shapes at different tilting angles, which can be obtained by direct query and matching.

[0057] Among them, in S3, the dimension parameters include the major axis and minor axis of the ellipse. The major axis is the diameter of the masterbatch, and the length of the major axis of the masterbatch remains unchanged in the projection state. The minor axis changes in length according to the tilting angle of the masterbatch. The minor axis and major axis of the end ellipse of each masterbatch projection correspond to a masterbatch projection parallel line with a specified length.

[0058] First, in the above-mentioned S3, the steps to determine the masterbatch projection parallel lines are as follows:

[0059] S310. Locate the ends of the masterbatch particles from the binary image, extract the end contours and morphology, fit a geometric model using the non-overlapping curve contours, and map the missing part to one side to supplement the missing elliptical segments at the ends of the masterbatch particles.

[0060] S311. Match the minor axis and major axis of the ellipse with the preset projection parameters of the masterbatch particles, and based on the matching results, determine the length of the parallel lines of the occluded masterbatch particles.

[0061] S312. Draw the determined parallel lines in the image in the form of dotted lines, and keep the parallel lines perpendicular to the major axis. Then, supplement the occluded ellipse at the other end of the parallel lines according to the already supplemented elliptical segments.

[0062] In this way, the contour segments of the masterbatch particles occluded at the lower end are supplemented on the basis of the existing segments and combined with the contour repair technology, providing a basis for calculating the projected particle size of the masterbatch particles in the later stage. Further, S4. Based on the matching results, measure the particle sizes of each masterbatch particle, divide the particle sizes of the masterbatch particles, and generate a scatter plot of the particle size distribution.

[0063] Taking the major axis length as the actual particle size of the masterbatch particles, there are the following masterbatch particle size data:

[0064]

[0065] Directly taking the major axis length as the particle size, the particle size list is obtained: [22.5, 18.3, 26.7, 20.1, 24.9]. Traverse the particle size data and count the frequency of each interval:

[0066] [15 - 20): 1 (18.3);

[0067] [20 - 25): 3 (22.5, 20.1, 24.9);

[0068] [25 - 30): 1 (26.7);

[0069] Draw a scatter plot: x-axis: midpoint of the particle size interval (such as 17.5, 22.5, 27.5); y-axis: frequency; scatter points: each interval corresponds to a point, and the frequency value is marked.

[0070] The curve profile of the non-overlapping part is used to fit the geometric model, and the missing part is mapped to one side to supplement the missing elliptical segment at the end of the masterbatch, so as to accurately measure the end shape of each masterbatch, and then obtain its size parameters. Based on the matching of the size parameters with the preset projection parameters of the masterbatch, the projection line segment and its morphological profile of the masterbatch can be quickly determined. The contour repair technology compensates for the influence caused by the overlap and occlusion of the masterbatch in traditional image analysis, differentiates the overlapping masterbatches, enables the overlapping masterbatches to be more accurately measured for their particle sizes, reduces the phenomenon that the overlapping area is misjudged as a single particle, facilitates the subsequent statistics of the particle size distribution of the masterbatch, and by dividing the particle size of the masterbatch and generating a scatter plot of the particle size distribution, the particle size distribution of the masterbatch can be intuitively understood. This method can clearly display the particle size distribution characteristics of the masterbatch, facilitating subsequent analysis and application.

[0071] In summary, this embodiment also provides a real-time image detection system for the particle size distribution and surface morphology of high-precision masterbatch, as Figure 2 shown, including an image acquisition and processing module 100, a region overlapping contour marking module 200, a projection line segment determination module 300 for masterbatch, and a masterbatch particle size distribution display module 400;

[0072] It includes an image acquisition and processing module 100: Through optical equipment and based on microscope imaging technology, the real-time acquired masterbatch image is transmitted to the processing unit, and the processing unit preprocesses the image;

[0073] The region overlapping contour marking module 200: Obtains the preprocessed image, acquires the contours of the region overlapping masterbatches based on image analysis technology, and marks the region overlapping masterbatches;

[0074] The projection line segment determination module 300 for masterbatch: Obtains the end contours of the marked overlapping masterbatches, repairs the end contours of the overlapping masterbatches based on contour repair technology, then obtains its size parameters according to the repaired end shape, and matches the size parameters with the preset projection parameters of the masterbatch to determine the projection line segment of the masterbatch;

[0075] The masterbatch particle size distribution display module 400: Based on the matching result, measures the particle size of each masterbatch, divides the particle size of the masterbatch and generates a scatter plot of the particle size distribution.

[0076] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and the descriptions in the specification are only preferred examples of the present invention, which are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.

Claims

1. A real-time image detection method for the particle size distribution and surface morphology of high-precision color masterbatch, characterized in that: It includes the following method steps: S1. Through an optical device and based on microscopy imaging technology, the image of the masterbatch obtained in real time is transmitted to a processing unit, and the processing unit preprocesses the image; S2. Obtain the preprocessed image, obtain the contours of the overlapping masterbatch based on image analysis technology, and mark the overlapping masterbatch; S3. Obtain the end contours of the marked overlapping masterbatch, repair the end contours of the overlapping masterbatch based on contour repair technology, then obtain its size parameters according to the repaired end shape, and match the size parameters with the preset masterbatch projection parameters to determine the projection line segments of the masterbatch; S4. Based on the matching result, measure the particle size of each masterbatch, divide the particle size of the masterbatch, and generate a scatter plot of the particle size distribution.

2. The real-time image detection method for the particle size distribution and surface morphology of high-precision color masterbatch according to claim 1, wherein: After receiving the image data, the processing unit is used to preprocess the image, including performing a segmentation operation on the image. The segmentation operation converts the received image data into a grayscale image based on image edge detection technology, and then calculates the gradient amplitude of the grayscale value of the grayscale image respectively through the principle of the Sobel operator to detect edges.

3. The real-time image detection method for the particle size distribution and surface morphology of high-precision color masterbatch according to claim 2, wherein: The method for calculating the gradient amplitude of the grayscale value of the grayscale image to detect edges is shown as follows: Two 3×3 convolution kernels are used to detect the grayscale value gradients in the horizontal and vertical directions respectively. The grayscale value gradient in the horizontal direction is synthesized with the grayscale value gradient in the vertical direction. The formed synthetic gradient amplitude value is compared with a preset amplitude value, the strong edges of the image are retained, and the image is binarized through adaptive threshold segmentation to highlight the masterbatch area.

4. The real-time image detection method for the particle size distribution and surface morphology of high-precision color masterbatch according to claim 3, characterized in that: The horizontal direction gray value gradient and the vertical direction gray value gradient are respectively H = G(I, G x ), Y = G(I, G y ), where I is a grayscale image; G≈|G x |+|G y |。 5. The real-time image detection method for the particle size distribution and surface morphology of high-precision color masterbatch according to claim 2, wherein: In the step S2, the steps for obtaining the contours of the overlapping masterbatch based on image analysis technology are as follows: S210. Obtain the image after the segmentation operation, extract the masterbatch contour according to the image, and extract the projection line segments in the contour based on the masterbatch contour; S211. Establish a coordinate system with the horizontal direction of the image as the x-axis and the vertical direction as the y-axis, extract two parallel lines of the projection line segments of the same masterbatch contour, and respectively obtain the coordinates of the endpoints of the parallel line segments; S212. Adopt a geometric method to calculate whether the projection line segments of two masterbatches intersect through the coordinates of the line segment endpoints to determine the overlapping situation of the masterbatches; If the line segments intersect, it proves that the masterbatches overlap. If the line segments do not intersect, it proves that the masterbatches do not overlap; S213. Mark the overlapping masterbatches.

6. The real-time image detection method for the particle size distribution and surface morphology of high-precision color masterbatch according to claim 5, characterized in that: In the step S212, the size of the overlapping area of the masterbatches is affected by the inclination state of the masterbatches. Therefore, when determining the overlapping state of the masterbatches, obtain the coordinates of the endpoints of the parallel lines of the masterbatches, and based on the endpoint coordinates, use the vector parallelism judgment algorithm to determine whether the parallel lines of the two masterbatches are collinear.

7. The real-time image detection method for the particle size distribution and surface morphology of high-precision color masterbatch according to claim 6, characterized in that: The vector parallel condition is as follows: (x2 - x1)(y4 - y3) = (x4 - x3)(y2 - y1); Among them, (x2 - x1, y2 - y1) and (x4 - x3, y4 - y3) are the direction vectors of the parallel lines of the two masterbatches respectively.

8. The real-time image detection method for the particle size distribution and surface topography of high-precision color masterbatch according to claim 1, wherein: In S3, the dimensional parameters include the major axis and minor axis of the ellipse. The major axis is the diameter of the masterbatch. The length of the major axis of the masterbatch remains unchanged in the projection state. The minor axis varies in length according to the inclination angle of the masterbatch. The minor axis and major axis of the end ellipse of each masterbatch projection correspond to a masterbatch projection parallel line with a specified length.

9. The real-time image detection method for the particle size distribution and surface topography of high-precision color masterbatch according to claim 8, wherein: In S3, the steps to determine the masterbatch projection parallel line are as follows: S310: Locate the end of the masterbatch in the binary image, extract the end contour and morphology, fit a geometric model using the non-overlapping curve contour, and map the missing part to one side to supplement the missing elliptical segment at the end of the masterbatch. S311: Match the minor axis and major axis of the ellipse with the preset masterbatch projection parameters. Based on the matching result, determine the length of the parallel line of the occluded masterbatch. S312: Draw the determined parallel line length in the image in the form of a dotted line, and keep the parallel line perpendicular to the major axis. Then, supplement the occluded ellipse at the other end of the parallel line according to the already supplemented elliptical segment.

10. A system for the real-time image detection method of the particle size distribution and surface topography of the high-precision masterbatch described in claim 1, characterized in that, It includes an image acquisition and processing module (100), a region overlapping contour marking module (200), a projection line segment determination module (300) for masterbatch, and a masterbatch particle size distribution display module (400). It includes an image acquisition and processing module (100): Through an optical device and based on microscope imaging technology, transmit the real-time acquired masterbatch image to the processing unit, and the processing unit preprocesses the image. The region overlapping contour marking module (200): Obtain the preprocessed image, acquire the contour of the region overlapping masterbatch based on image analysis technology, and mark the region overlapping masterbatch. The projection line segment determination module (300) for masterbatch: Obtain the end contour of the marked overlapping masterbatch, repair the end contour of the overlapping masterbatch based on contour repair technology, then obtain its dimensional parameters according to the repaired end shape, match the dimensional parameters with the preset masterbatch projection parameters, and determine the projection line segment of the masterbatch. The masterbatch particle size distribution display module (400): Based on the matching result, measure the particle size of each masterbatch, divide the masterbatch particle size, and generate a scatter plot of the particle size distribution.

Images