A method and device for measuring the size of micron-level particles in the process of impacting a wall surface under high-speed photography conditions

Abstract

This invention discloses a method and device for measuring the particle size of micron-sized particles during impact with a wall under high-speed photography conditions, belonging to the fields of image processing and microscopic particle size detection technology. Based on traditional particle shadow velocimetry, this invention utilizes a particle transport device to create identical motion conditions for both standard and test particles. A high-speed camera captures images of the moving particles against a similar background. The acquired images are then processed to obtain the measured particle size. By analyzing the mathematical relationship between the actual particle size of the standard particle and its measured value, the actual particle size of the test particle is finally derived. This invention establishes a particle image processing workflow that can rapidly process large numbers of image samples, effectively achieving high-speed and high-precision measurement of micron-sized particle size during impact with a wall compared to traditional particle shadow velocimetry.

Description

Technical Field

[0001] This invention belongs to the field of image processing and micro-particle size detection technology, and more specifically, relates to the particle size of micron-level particles during the impact of a wall surface under high-speed photography conditions. Background Technology

[0002] Ash accumulation and slagging on boiler heating surfaces have long plagued boiler operation. To delve deeper into their formation mechanism and control strategies, some researchers have begun studying the rebound and adhesion characteristics of fly ash particles colliding with the boiler wall. The formation of ash deposits on boiler heating surfaces is primarily caused by the inertial collisions of fly ash particles. The average particle size of the fly ash produced in boilers is mostly 40-50 μm, which impacts the heat exchanger surface at a certain velocity under the carrying action of high-temperature flue gas. To study this process under laboratory conditions—that is, to track the impact of individual particles on the wall—advanced measurement techniques are required. Among these, Particle Shadow Velocimetry (PSV) is the most commonly used method. This method only requires placing a light source behind the subject, and the motion characteristics of the particles can be captured using a macro lens. The implementation principle of the above basic technology can be found in the invention patent application publication number CN08957026A.

[0003] Based on the aforementioned patent's measurement foundation, actual measurement revealed that although PSV technology can capture motion images of fly ash particles, conventional image measurement methods struggle to achieve high-precision and efficient measurement of particle diameter. Specifically, the main problems are as follows: First, the image capture results of fly ash particles are significantly affected by environmental factors. For example, when particles are near a nozzle or impact platform, the image size changes considerably due to variations in light intensity. Furthermore, particles typically pass through areas of varying background brightness during their movement, resulting in significant fluctuations and variations in the image size, which are difficult to completely avoid. Second, when photographing fine particles, the measurement accuracy is limited by the resolution and pixel size of the high-speed camera used. When the fly ash particle image captured by the macro lens occupies only a few pixels, conventional image measurement methods for estimating the diameter of the particle shadow generally result in an error of at least 0.5 pixels, leading to measurement results that fail to meet experimental requirements. Thirdly, due to the demands of experimental work, the number of particle images captured by high-speed cameras is enormous, and the imaging effects between images vary, making processing cumbersome and lacking efficient batch processing methods. In general, on the one hand, measurement results are usually affected by factors such as the particle environment and the imaging equipment, placing higher demands on the measurement method; on the other hand, experimental requirements necessitate a large amount of particle image data to support measurement accuracy. Therefore, this patent improves the image measurement method based on existing hardware measurement of particle parameter information and further proposes a workflow for batch processing experimental images.

[0004] A search revealed Chinese patent application publication number CN104390895A, entitled "A Method for Measuring Particle Size Based on Microscopic Imaging Image Gray Scale." The disclosed method includes the following steps: 1) acquiring images of particles in a diluted solution using a microscope; 2) performing gray-scale scanning in the x / y directions to determine the center position of the particles; and 3) analyzing and determining the equivalent diameter D of the particles. e ;4) The known diameter is D. s The equivalent diameter D of the particle e , the diameter ratio D e / D s With D s 5) Perform curve fitting to obtain characteristic curves for particles of different diameters; finally, measure the equivalent diameter D of the particle to be tested. eThe true diameter can be obtained by searching. This patent obtains the equivalent diameter by performing grayscale analysis on the particle image, compares it with known data, and thus obtains the true diameter. It establishes the relationship between the image shadow diameter and the true diameter of the particle using particles with known diameters. It uses this relationship to predict the true diameter of the sample particle based on the image shadow diameter. It is suitable for measuring particles with stable imaging backgrounds, low movement speeds, and a large number of pixels in the particle shadow imaging, but its applicability is very limited.

[0005] For example, Chinese patent application publication number CN1084967A, entitled "A Direct Measurement Method and Device for Particle Size," discloses a method comprising the following steps: 1) preparing the sample to be tested as a glass slide; 2) adjusting the field of view by moving the stage horizontally / vertically; 3) moving the reticle along the particle size distribution, and finally calculating the particle size using a formula. This invention uses a stepper motor to control the basic device to measure the particle diameter. This patent is suitable for measuring static particle samples and is difficult to effectively use for measuring the diameter of particles in motion.

[0006] For example, Chinese patent application publication number CN113769797A, entitled "A Method for Determining the Diameter of Microscale Particles in Fluid-Solid Two-Phase Transport," discloses a method comprising the following steps: 1) preparing a polydisperse viscoelastic fluid system; 2) injecting the final polydisperse viscoelastic fluid system into a microchannel from the inlet end using a pressure source; 3) using specific technical means to form a particle chain with progressively increasing particle diameter along the flow direction at the centerline position of the microchannel outlet end, thereby measuring the particle diameter. The main feature of this patent is that it achieves the determination of different particle diameters by controlling the rheological properties of the fluid and the particle diameter within the fluid. However, when using this patent's technical solution for measurement, the fluid preparation is cumbersome, and the standard particle diameter and the rheological properties of the fluid need to be adjusted multiple times during the measurement process, making it unsuitable for measuring the diameter of moving particles at high temperatures.

[0007] Furthermore, Chinese patent application publication number CN103424080A, entitled "A Nanoparticle Diameter Measurement Device and Method," discloses a method comprising the following steps: a light source emits a beam of outgoing light; a concave lens diverges the outgoing light to obtain diverged light; a sample cell scatters the diverged light to obtain first scattered light; a pinhole component images the first scattered light passing through the pinhole to obtain second scattered light; a filter filters the second scattered light to obtain third scattered light; a conversion unit images the third scattered light to obtain an optical image and converts the optical image into a digital signal; and a calculation unit calculates the diameter of the nanoparticle based on the digital signal. The main feature of this patent is that it utilizes the optical properties of nanoparticles to conduct a series of experiments and analyses to obtain the particle diameter. However, this method has high requirements for the surrounding environment, and the optical instruments used are relatively expensive. It is also difficult to adapt to the measurement of the diameter of boiler flue dust particles, which are large in size and have a wide range of size fluctuations. Summary of the Invention

[0008] 1. The problem to be solved

[0009] The purpose of this invention is to overcome the shortcomings of existing technologies for measuring the particle size of fly ash particles during impact with a wall, which suffer from large image acquisition errors, complex and inefficient image processing, and poor processing accuracy. This invention provides a method and device for measuring the particle size of micron-level particles during impact with a wall under high-speed photography conditions. The technical solution of this invention effectively solves the above problems, not only reducing the complexity of image post-processing and improving processing efficiency, but also significantly improving the accuracy of particle size measurement results.

[0010] 2. Technical Solution

[0011] To solve the above problems, the technical solution adopted by the present invention is as follows:

[0012] This invention discloses a particle size measurement device for micron-level particles impacting a wall surface under high-speed photography conditions. The device includes a particle transport device, an impact platform, and a high-speed camera. The particle transport device has a first channel and a second channel, each independently used to transport different airflows. The first channel carries airflow carrying particles to be measured, while the second channel carries airflow carrying standard particles. The inlet ends of the first and second channels are connected to their corresponding airflow supply devices, and the outlet ends of both channels are connected to feeding nozzles. These nozzles simultaneously eject standard particles and particles to be measured. It should be noted that, to reduce subsequent processing errors, the particles to be measured in this invention are first sieved using a sample sieve to select particles whose particle size is close to that of standard particles with known true particle sizes before measurement. This improves measurement accuracy.

[0013] The feed nozzle is also equipped with a structure for the movement and ejection of the two types of particles. The two types of particles move along their own channels in the feed nozzle and are ejected from their own nozzles. The movement of the two types of particles does not interfere with each other. By setting the feed nozzle, the particles to be tested and the standard particles can be ejected together, and the running speed after ejection is consistent.

[0014] The impact platform is located directly below the feeding nozzle, and the impact wall is only for the particles to be tested to impact. The high-speed camera is used to continuously photograph the process of the standard particles and the particles to be tested falling, which is a set of images. The set of images is taken by the high-speed camera during the entire process of a certain particle to be tested being ejected from the feeding nozzle and moving to the impact wall.

[0015] Furthermore, the particle transport device of the apparatus is manufactured into a cylindrical structure with a first channel at its center. This first channel is cylindrical, and a second channel is formed circumferentially around its outer edge. The feeding nozzle has a funnel-shaped structure, with its central region connected to the first channel and its edge region connected to the second channel. The central and edge regions are not connected. This design ensures that the particles to be tested are ejected from the center of the device, while standard particles are ejected from the periphery of the central region. This guarantees a high degree of similarity in the background environment of the trajectories of the two types of particles upon incident, and that the designed velocities of the two types of particles are also consistent. This facilitates the subsequent synchronous imaging of the motion trajectories of a series of standard and test particles by a high-speed camera. This significantly reduces the large imaging differences caused by environmental changes, thereby reducing errors in predicting the sample particle diameter.

[0016] Furthermore, to ensure a high degree of similarity between the background environment of the area through which the standard particles and the particles to be tested fall, and to reduce the uncontrollable influence of airflow disturbance on the particle trajectory, this invention sets the distance between the impact wall and the feeding nozzle on the impact platform to be 3-5 mm, and the diameter D of the impact wall is... f With the length L of the impact platform f The ratio is 1:(10-15). On the other hand, in this invention, at the discharge port of the feeding nozzle, the inner diameter of the port for standard particles to be ejected is D, the flow rate of the standard particle airflow is Qs, the inner diameter of the port for particles to be tested to be ejected is Dt, the wall thickness of the port for particles to be tested to be ejected is d, the flow rate of the particle airflow is Qt, and the following relationship is satisfied:

[0017]

[0018] This invention, through the aforementioned design and by matching the outlet cross-section of the feeding nozzle with the dimensions of the impact wall, effectively reduces the disturbance of the flow field around the impact surface, better controls the falling trajectory of the standard particles, and better controls the two types of particles within an area of ​​consistent background brightness, further improving the accuracy of the standard particle image size measurement standard. By optimizing the structure of the particle transport device used for measurement, this invention creates a highly similar imaging background environment for both the standard particles and the particles to be measured, significantly reducing the large imaging differences caused by environmental changes, thereby reducing errors in sample particle diameter prediction and improving image acquisition accuracy. Using the particle transport device of this invention to acquire images not only effectively reduces the workload of post-processing but also significantly reduces the complexity of image data processing, improving its processing efficiency.

[0019] Simultaneously, this invention also designs a particle image processing workflow. When processing the image set acquired using the aforementioned particle transport device, the image processing workflow is optimized to achieve efficient batch processing of the image set with high accuracy. This invention employs optimized image processing technology to measure the particle size of moving particles at the micrometer level. This measurement process can be completed by writing a program on the MATLAB software platform. The processing mainly includes particle image diameter identification and particle size estimation. The particle image diameter identification method includes the following steps:

[0020] Step S1: Read the image set captured by the high-speed camera in sequence according to the shooting time, and convert the RGB images into grayscale images to obtain the matrix distribution of grayscale values ​​of the grain images;

[0021] Step S2: Capture the image capture window. This involves narrowing the image analysis area based on the approximate location of the particle's trajectory and defining a suitable analysis field of view. This reduces the influence of shadows at the edges of bright areas during image analysis, ensuring a more uniform background grayscale value within the capture window, resulting in higher image quality and fewer interfering images. This leads to a clearer contrast between the foreground and background of the particle image. The actual size of the capture window can be pre-defined based on specific image characteristics. Simultaneously, since particles pass through different locations within the shooting area during their fall, even slight changes in background light intensity can affect the particle's image size in high-speed photography. Therefore, it's necessary to determine the most stable imaging area based on the actual lighting brightness and the contrast between the particle and background, and then analyze and process the particle image in that area. In practice, all particle images within a set of images captured during the impact of a standard particle can be pre-processed to obtain the standard particle's trajectory and spatial distribution of particle size. A suitable window area can then be derived by referring to the actual particle size of the standard particle. The above settings are mainly based on the fact that in the image processing process, especially the processing of grain images, there is often interference from other particles in the environment. At the same time, the size of the image of the same particle often changes under different background brightness. Therefore, the window-based image processing method can remove the interference of environmental particles while locking the grain image under a specific background brightness for processing, so that the obtained image grain size value is more stable and the result is more accurate. This further reduces the complexity of the image post-processing program and improves the accuracy of the relevant data.

[0022] Furthermore, it should be further explained that the use of standard particles for collision analysis in this invention is to obtain areas with high imaging quality within the captured image beforehand (during the collision process, the particles mainly move vertically during the incident phase and move significantly horizontally during the rebound phase, thus enabling a relatively comprehensive evaluation of the imaging characteristics of different locations within the camera's shooting area). This is because the delineation of the window and the determination of a reasonable area require a basic range basis during the actual processing operation.

[0023] Currently, image analysis algorithms are limited to a few fixed types. Even when the optimal processing method is selected, it is difficult to accurately obtain the true particle size. This is because imaging errors exist during image capture. These errors depend on hardware conditions such as the capturing equipment and are influenced by various factors. The measurement device of this invention is based on this. However, even so, there are still other uncontrollable factors during measurement. This invention addresses these undesirable results caused by these factors by starting with the imaging results (image processing). (These influencing factors can be judged by the position of the particle in the image). For example, window cropping and field of view division are performed during image processing. Thus, during implementation, a suitable area needs to be pre-determined in the image. Typically, the particle imaging size within this area is close to the true size, and the imaging is relatively stable, thereby improving the processing accuracy.

[0024] Step S3: Determine whether any particles have entered the recognition area through the capture window. The area must contain both the particle to be tested and the standard particle. If any particles enter the recognition area, proceed to the next step; otherwise, end the processing, output the result, and automatically read the next image.

[0025] Step S4: Through the capture window, based on the particle's falling process and the sequence of image capture, preliminarily determine whether the particle's position is reasonable. In principle, the sequence of particle image reading should be consistent with the camera's shooting order, that is, the reading order should match the particle's falling process. At the same time, considering that the falling trajectories of the standard particle and the particle to be tested may intersect during actual operation, a certain distance must be ensured between them. Therefore, the reasonableness of the height and horizontal position of the particle image can be determined based on the particle's position in the capture window. If it is reasonable, proceed to the next step of processing; otherwise, end the processing process.

[0026] Step S5: Perform sub-pixel processing on the captured image window, that is, re-divide the pixels. The density of pixel distribution in the processed particle image increases, and the distribution of gray values ​​at the particle edges is refined, which helps to improve the accuracy of image binarization. Then, the actual size of the processed image is calibrated according to the diameter of the impact wall.

[0027] Step S6: To improve the accuracy of particle image binarization and reduce interference from background content, a judgment window is further extracted based on the recognition window. The judgment window contains the image of the particle to be judged, and the background area is relatively small with a stable grayscale distribution. Since different image segmentation thresholds affect the particle size measurement results, different existing binarization algorithms are first used to obtain segmentation thresholds for the standard particle image. For each threshold, the processed image particle size is compared with the actual particle size of the standard particle. The binarization threshold algorithm corresponding to the closest diameter is selected. Then, the selected threshold segmentation method is used to process the particle image to be tested in the same window to identify its image particle size. Because the environmental conditions of the standard particle and the particle to be tested in the same image are highly consistent, the optimal threshold algorithm for binarizing the standard particle image has good applicability to the processing of nearby particle images to be tested. The mathematical relationship between the standard particle image particle size and the actual particle size can be well used to estimate the actual particle size of the particle to be tested. Therefore, by using this threshold to process the particle image to be tested, the diameter of the particle to be tested in high-speed motion can be accurately derived.

[0028] Step S7: Output the particle size results of the test particle and the standard particle images, and store them in the database.

[0029] Furthermore, the size of the particle to be tested is estimated by constructing a mathematical relationship between the particle size in a standard particle image and its actual particle size, including the following steps:

[0030] Step S1: Read the particle size information of standard particles from the database;

[0031] Step S2: Calculate the average particle size D of a standard particle image. e D e Take the squared average and calculate the true particle size D of the standard particles. s With D e By calculating the quotient, we can obtain the ratio 'a' between the two; where:

[0032]

[0033] D1, D2, ... D n Let n be the particle size of each image obtained from the standard particle, and n be the number of images of the standard particle.

[0034] Step S3: Since different particle movement speeds will affect the size of the particle image, the parameter information of the standard particles is distinguished according to different feeding design speeds. For the same design speed, the particle size D of the standard particle image is constructed. e A graph showing the ratio 'a';

[0035] Step S4: Calculate the average particle size D' of a certain particle image using the method in step S2. e ;

[0036] Step S5, via D e -a diagram and D' e The true particle size D' of the particle to be tested was calculated. s That is, D' s =a*D' e .

[0037] Compared with existing technologies, the technical solution of this invention, in the study of hot particle adhesion problems, for high-speed moving micron particles, can create the same motion conditions for standard particles and particles to be tested by using the particle conveying device of this invention, thereby greatly improving the shooting accuracy of particle images, effectively reducing the difficulty of subsequent particle size data processing, and solving the problem of particle size measurement error caused by factors such as changes in background environment and hardware limitations during shooting.

[0038] Most importantly, based on high-speed photography, this invention proposes a comprehensive and efficient analysis and processing scheme for measuring micron-level dynamic particle size parameters. By optimizing the image processing flow, especially the S2-S2 processing operation in the image particle size identification process, it can quickly achieve efficient batch processing of image sets with high processing accuracy. The technical solution of this invention provides a convenient shortcut for rapidly obtaining the massive amount of information required for the study of particle adhesion and rebound phenomena under hot conditions. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the particle size measuring device in this invention;

[0040] Figure 2 This is a flowchart of particle image processing in this invention.

[0041] Figure 3 This is a diagram showing the motion trajectory of the standard particle in this invention;

[0042] Figure 4 The distribution of particle size along the horizontal direction in the particle image;

[0043] Figure 5 The distribution of particle size along the vertical direction in the particle image.

[0044] In the picture:

[0045] 1. Airflow of particles to be tested; 2. First channel; 3. Feed nozzle; 4. Outgoing airflow; 5. Impact wall; 6. Impact platform; 7. Particles to be tested; 8. Standard particles; 9. Second channel; 10. Airflow of standard particles. Detailed Implementation

[0046] Since particle diameter significantly influences its rebound characteristics upon impact with a wall, accurate measurement of particle diameter is crucial under limited conditions, especially for particles with only a few pixels in the imaging sensor. To address this critical issue of particle size measurement, this invention employs high-speed imaging as the fundamental method for measuring particle impact dynamics. This method, based on PSV technology, utilizes imaging equipment and an illumination source to record the entire particle motion process, and uses existing supporting computer software for camera control and post-processing analysis of the image data.

[0047] In actual particle collision experiments, particles are continuously fed into a high-temperature heating pipe by a feeder. The particles are then ejected through an accelerating nozzle after passing through the heating pipe, falling vertically and colliding with the platform. The entire process occurs within a relatively small high-temperature environment. Therefore, to more accurately analyze the adhesion and rebound characteristics of particles, a scheme capable of precisely measuring the size of micron-level moving particles needs to be designed. Furthermore, based on existing hardware measurement technologies, this scheme can achieve batch processing of particle images with high accuracy in a short time, thereby obtaining a large amount of particle data.

[0048] Because the particles are small in size, have high speed, and the collision process is short, a high-speed camera equipped with a macro lens is used for imaging, and a xenon light source that continuously provides cold light is used for illumination. Based on this method, after acquiring the particle image, the actual length represented by each pixel is obtained by calibrating with a reference object, thereby estimating the true size of the particle and related motion parameters and other basic information.

[0049] The present invention will be further described below with reference to specific embodiments.

[0050] Example 1

[0051] like Figure 1As shown, this invention discloses a particle size measurement device for micron-level particles impacting a wall surface under high-speed photography conditions. The device includes a particle transport device, an impact platform 6, a high-speed camera, and a light source. The particle transport device has a first channel 2 and a second channel 9. The first channel 2 allows the flow of the particle airflow 1 to be measured, and the second channel 9 allows the flow of the standard particle airflow 10. The outlet ends of both the first channel 2 and the second channel 9 are connected to a feeding nozzle 3. The feeding nozzle 3 is equipped with heating or accelerating components to heat or accelerate the particles. This configuration ensures that the particle 7 to be measured and the standard particle 8 are ejected simultaneously from the feeding nozzle 3 at the same speed. In this embodiment, a xenon light source that continuously provides cold light is used as the light source.

[0052] It should be noted that, to ensure the accuracy of the measurement results, in this embodiment, before measurement, the particle 7 to be measured is first screened using a sample sieve to select particles 7 whose particle size is closer to that of the standard particle 8 with a known true particle size. This effectively reduces the measurement error caused by large differences in particle size in the image. By controlling the falling speed of the two types of particles, the similarity of the background of the two particles in the same image can be controlled, significantly reducing the particle image background brightness variation and particle lighting deviation caused by surrounding objects blocking the light. This invention, through the setting of the feeding nozzle 3, effectively controls the similarity of the movement conditions of the particle 7 to be measured and the standard particle 8, greatly reducing the measurement error caused by the difference in particle speed, thereby ensuring measurement accuracy.

[0053] The impact platform 6 is located directly below the feeding nozzle 3, and the impact wall 5 is only for the particle 7 to be tested to impact. The distance between the impact wall 5 on the impact platform 6 and the feeding nozzle 3 is 3-5 mm. The high-speed camera is mounted at a certain distance outside the impact platform 6 (this distance must ensure that the particle being photographed is within the adjustment range of the lens, and the specific distance is affected by the actual model of the macro lens used in the camera; generally, the distance between the particle and the lens is about 15-25 cm). The high-speed camera is used to continuously photograph a set of images of a single standard particle 8 and a single particle 7 falling, which is a set of images captured by the high-speed camera during the entire process of a certain particle 7 being ejected from the feeding nozzle 3 and moving to the impact wall 5.

[0054] As a further optimization of this embodiment, the particle conveying device of the present invention is processed into a concentric cylindrical structure with dual channels. A first channel 2 is machined in the central region, and a second channel 9 is formed circumferentially around the outside of the first channel 2. The feeding nozzle 3 has a funnel-shaped structure, with its central region connected to the first channel 2 and its edge region connected to the second channel 9. However, its central and edge regions are not connected. This design ensures that the particle to be tested 7 is ejected from the center of the device, while the standard particle 8 is ejected from the periphery of the central region. This ensures a high degree of similarity in the background environment of the trajectories of the two types of particles during injection, and that the design velocities of the two types of particles are also consistent. This facilitates the subsequent synchronous imaging of a series of standard particles 9 and the particle to be tested 7 by a high-speed camera. This significantly reduces the large imaging differences caused by environmental changes between the two types of particles, thereby reducing the error in predicting the sample particle diameter.

[0055] Specifically, such as Figure 1 As shown, to further ensure that the background environment of the area through which the standard particle 8 and the particle to be tested 7 fall is highly similar, and to reduce the uncontrollable influence of airflow disturbance on the particle trajectory, this invention sets the distance between the impact wall 5 on the impact platform 6 and the feeding nozzle 3 to be 3-5 mm, and the diameter D of the impact wall 5 is... f The length L of the impact platform 6 f The ratio is 1:(10~15), and the diameter D of the impact wall 5 is... f The inner diameter of the port from which the test particle 7 is ejected is slightly larger than Dt, and smaller than the inner diameter of the port from which the standard particle 8 is ejected is D. When the test particle 7 impacts the impact wall 5, the ejected airflow 4 at the position of the standard particle 8 is circumferentially discharged from the impact platform 6, thereby helping to reduce the influence of airflow disturbance on the particle's trajectory. Furthermore, to further reduce the influence of airflow disturbance on the measurement results, the airflow rate is controlled to better match the structural dimensions of the particle transport device. This invention better controls the descent trajectory of the standard particle 8 by minimizing the disturbance in the flow field around the impact surface, and better controls the two types of particles within an area with consistent background brightness. By controlling the flow rate of the standard particle airflow 10 to Qs, the flow rate of the test particle airflow 1 to Qt, and the wall thickness of the port from which the test particle 7 is ejected to be d, the following relationship exists:

[0056]

[0057] By optimizing the structure of the measuring device of the present invention, the accuracy of image acquisition is improved when it is used to acquire particle images, thereby effectively reducing the workload of subsequent image processing, significantly reducing the complexity of image data processing, and improving its processing efficiency.

[0058] The following describes in detail the operation process of image processing of image sets captured by the measuring device of the present invention, and the steps of measuring the particle size of micron-sized particles through the above image processing process, using specific examples.

[0059] First, determine the standard particle size D in this case. s The particle size was 78.9 μm. Test particles with a particle size close to that of the standard particles were sieved using a sample sieve. A photo set was acquired using the measuring device of this invention. Specifically, in this case, the distance between the impact wall 5 and the feeding nozzle 3 was controlled to be 4 mm, and the diameter D of the impact wall 5 was... f The length L of the impact platform 6 is 2mm. f The diameter of the port from which the test particle 7 is ejected is 30mm. The inner diameter D of the port from which the standard particle 8 is ejected is 2mm, and the wall thickness d of the port from which the test particle 7 is ejected is 2mm. The flow rate Qs of the standard particle airflow 10 and the flow rate Qt of the test particle airflow 1 are adjusted according to the formula of the present invention. It should be noted that an example value is given here for different port dimensions of the feeding nozzle 3 (such as Dt, D, d). The value can be determined according to the actual situation. No specific limitation is made in the present invention. After determining the above dimensions of the device, the relative particle airflow flow rate can be determined according to the formula of the present invention, thereby creating a suitable shooting background environment and minimizing the impact of airflow disturbance on the image acquisition accuracy. During the process of the particles falling after being ejected, a high-speed camera is used to continuously shoot and obtain an image set.

[0060] Then, the image processing flow of this invention is used to process the above image set to measure the particle size of particles in motion. This mainly consists of two parts: particle image size recognition and particle size estimation. The particle image size recognition process is as follows: Figure 2 As shown, the specific steps include the following:

[0061] 1. Read the above image set sequentially according to the shooting time of the camera, and convert the RGB images into grayscale images to obtain the matrix distribution of the image grayscale values;

[0062] 2. Capture the capture window. Here, all standard particle images in the image set during the standard particle impact process are processed beforehand to obtain the motion trajectory of the standard particles, such as... Figure 3 As shown. (The standard particle impaction analysis here is used for windowing and field-of-view analysis, only to evaluate the field-of-view characteristics within the images captured by the camera, where the X-axis represents the horizontal direction along the impact wall and the Y-axis represents the normal direction along the impact wall), and the distribution of image particle size in the two-dimensional plane (along the X and Y directions respectively) (as shown in the figures). Figure 4 and Figure 5The appropriate window area is obtained by referring to the actual particle size of the standard particle (the actual particle size is 78.9um), and the window area of ​​the image of the particle to be tested is divided according to this rule in order to capture the best capture window.

[0063] 3. Determine whether any particles have entered the recognition area through the capture window, ensuring that the recognition area contains both the particle to be tested and the standard particle, and preliminarily determine whether the particle position is reasonable; if the above two requirements are not met, output the result, end the processing flow, and automatically read the next image;

[0064] 4. The captured image window is further divided into pixels, and the actual size of the processed image is calibrated according to the width of the 2mm diameter impact platform.

[0065] 5. Based on the recognition window, further extract the interpretation window. The interpretation window contains the image of the particles to be interpreted, and the background area is relatively small with a stable gray value distribution.

[0066] 6. Different binarization algorithms are used to obtain segmentation thresholds for the standard particle images. The processing results of the above methods are compared with the actual particle size values ​​of the standard particles, and the binarization threshold segmentation method corresponding to the closest diameter is selected. In this embodiment, the Mean average grayscale thresholding method is used. This thresholding method is used to process the particle image to identify the size of the image particle.

[0067] 7. Output the particle size results of the test particle 7 and the standard particle 8.

[0068] Finally, based on the above image processing results, the size of the particle to be tested is estimated by constructing a mathematical relationship between the particle size of the standard particle image and its actual particle size. The specific process is as follows:

[0069] 1. Read the particle size information of standard particles from the database;

[0070] 2. Calculate the average particle size D of a standard particle image. e D e Take the squared average and calculate the true particle size D of the standard particles. s With D e By calculating the quotient, we can obtain the ratio a (a = D). w / D e ), thus obtaining D e =77.2024um, a=0.9785.

[0071] 3. For the same design speed, construct a standard particle image with particle size D. e A graph showing the ratio 'a';

[0072] 4. Calculate the average particle size D' of a certain particle image using the method in step 2. e =78.0738um;

[0073] 5. Through D e -a diagram and D' e The true particle size D' of the particle to be tested was calculated. s =D' e *a = 76.3952um, here we take a = 0.9785.

[0074] Furthermore, the particle size of the test particle 7 in a static state in this embodiment was measured, and the result was 78.5 μm, which is close to the actual particle size of the test particle measured by the method of the present invention. The measurement method of the present invention can not only effectively measure the particle size of particles in motion, but also has high measurement accuracy. At the same time, the image processing method of the present invention is fast and efficient, which is beneficial to providing a convenient shortcut for quickly obtaining and processing the massive amount of information required for the study of particle adhesion and rebound phenomena under hot conditions.

[0075] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A particle size measurement device for micron-sized particles impacting a wall surface under high-speed photography conditions, characterized in that: The system includes a particle conveying device, an impact platform (6), and a high-speed camera. The particle conveying device has a first channel (2) and a second channel (9). The first channel (2) is for the flow of the particle to be tested (1), and the second channel (9) is for the flow of the standard particle (10). The outlet ends of the first channel (2) and the second channel (9) are connected to the feeding nozzle (3). The feeding nozzle (3) is used to simultaneously spray out the standard particle (8) and the particle to be tested (7). The impact platform (6) is located directly below the feeding nozzle (3). The distance between the impact wall (5) on the impact platform (6) and the feeding nozzle (3) is 3~5mm. The impact wall (5) is only for the particle to be tested (7) to impact. The high-speed camera is used to continuously photograph the falling process of the standard particle (8) and the particle to be tested (7). The inner diameter of the port at the outlet of the feeding nozzle (3) from which standard particles are ejected is [missing information]. D The flow rate of the standard particulate airflow (10) is Qs The inner diameter of the port from which the particle to be tested (7) is ejected is Dt The wall thickness of the port from which the particle (7) is ejected is d The flow rate of the particulate airflow (1) to be measured is Qt : 。 2. The particle size measurement device for micron-level particles impacting a wall surface under high-speed photography conditions as described in claim 1, characterized in that: The particle conveying device is processed into a concentric cylindrical structure with a first channel (2) at its center and a second channel (9) formed circumferentially around the outside of the first channel (2); the feeding nozzle (3) is a funnel-shaped structure with its central area connected to the first channel (2) and its edge area connected to the second channel (9).

3. The particle size measurement device for micron-sized particles impacting a wall surface under high-speed photography conditions as described in claim 1, characterized in that: The diameter of the impact wall (5) D f Length of impact platform (6) L f The ratio is 1:(10~15).

4. A method for measuring the particle size of micron-sized particles during impact with a wall surface under high-speed photography conditions, characterized in that: The measuring device according to any one of claims 1-3 is used to acquire images of particles before they impact the wall, and image processing technology is used to process the images captured by the high-speed camera. The identification of particle size in the images specifically includes the following steps: Step S1: Read the image set captured by the high-speed camera in sequence according to the shooting time, and convert the RGB images into grayscale images to obtain the matrix distribution of grayscale values ​​of the grain images; Step S2: Capture the capture window, ensuring that the background grayscale value distribution within the capture window is uniform and there are no interfering images, so that there is a clear contrast between the foreground and background of the grain image. Step S3: Determine whether any particles have entered the recognition area through the capture window. The area must contain the particle to be tested (7) and the standard particle (8). If a particle enters the recognition area, proceed to the next step. If no particle enters the recognition area, end the processing, output the result, and automatically read the next image. The feed nozzle (3) is also equipped with a structure for the movement and ejection of the two particles. The two particles move along their own channels in the feed nozzle (3) and are ejected from their own nozzles. The movement of the two particles does not interfere with each other. By setting the feed nozzle (3), the particle to be tested (7) and the standard particle (8) are ejected together, and the running speed after ejection is consistent. Step S4: Determine whether the position of the particles within the capture window is reasonable. If reasonable, proceed to the next step of processing; otherwise, end the processing. Step S5: Perform sub-pixel processing on the capture window, and then calibrate the actual size of the processed image according to the diameter of the impact wall (5). Step S6: Based on the recognition window, further extract the judgment window, which contains images of two types of particles; use the binarization threshold segmentation algorithm to process the image of the particle to be tested (7) to identify the particle size of the standard particle (8) and the particle to be tested (7); Step S7: Output the particle size results of the test particle and the standard particle images, and store them in the database.

5. The method for measuring the particle size of micron-sized particles during impact with a wall under high-speed photography conditions according to claim 4, characterized in that: It also includes the estimation of the particle size of the particle to be tested (7).

6. The method for measuring the particle size of micron-sized particles during impact with a wall under high-speed photography conditions, as described in claim 5, is characterized in that: In step S6, different binarization processing algorithms are first used to obtain the segmentation threshold for the standard particle image. The particle size result of the image processed above is compared with the actual particle size value of the standard particle (8) for each threshold. The binarization threshold segmentation algorithm used when the diameter is closest is selected. Then, the threshold segmentation method is used to process the image of the particle to be tested (7) in the same window to identify the particle size of the image of the particle to be tested (7).

7. A method for measuring the particle size of micron-sized particles during impact with a wall surface under high-speed photography conditions, as described in claim 5 or 6, characterized in that: The method for estimating the particle size of the particle to be measured is as follows: Step S1: Read the particle size information of the standard particles (8) in the information database; Step S2: Calculate the average particle size of a standard particle image. D e The actual particle size of the standard particles (8) D s and D e By calculating the quotient, we can obtain the ratio 'a' between the two; where: D 1, D 2, ... D n The particle size of each image was measured for the standard particle (8). n The standard number of particle images; Step S3: Construct standard particle size images for the same design speed. D e A graph showing the ratio 'a'; Step S4: Calculate the average particle size of a certain particle image using the method in step S2. ; Step S5, through D e -a diagram and The true particle size of the particle to be tested (7) was calculated. ,Right now =a .

Images