Particle analysis method of automatic cleanliness analysis system

Abstract

The invention relates to a particle analysis method of an automatic cleanliness analysis system, which is mainly used for automatically analyzing particles on a clean surface so as to realize accurate cleanliness evaluation. The method comprises the following steps: firstly, acquiring image data of particles on a cleaned surface through cleanliness analysis equipment, then, preprocessing the image data to improve the definition of particle boundaries, then, identifying the shape, size and classification information of the particles by adopting an image identification algorithm, and then, according to the shape, size and distribution information of the particles, identifying the shape, size and distribution information of the particles. And finally, a particle analysis report including the particle number, the particle concentration, a morphological analysis result and cleanliness evaluation is generated through automatic software, and the working state of the cleaning equipment is adjusted through a feedback mechanism so as to optimize the cleaning process. The method can realize accurate detection and concentration analysis of clean surface particles, has a wide application prospect, and is especially suitable for the fields of industrial cleaning, precision manufacturing and the like.

Description

Technical Field

[0001] The present invention relates to a particle analysis method of an automatic cleanliness analysis system, which is particularly suitable for a technology for rapid and accurate analysis of particles on a clean surface in the field of industrial production and precision manufacturing. Specifically, the present invention analyzes particles on a clean surface through an image processing algorithm to evaluate the cleanliness, and adjusts the working parameters of the cleaning equipment through a feedback mechanism, thereby achieving automatic optimization and fine control of the cleaning process. Background Art

[0002] With the improvement of industrial automation level, cleanliness control has become one of the important factors affecting product quality, especially in precision manufacturing, electronic component production and high cleanliness environment, surface cleanliness directly affects product performance and quality. Traditional cleanliness detection methods mostly rely on manual detection or simple physical measurement, which are not only inefficient, but also easily interfered by human factors, and cannot meet the needs of high precision and high efficiency.

[0003] At present, there are some cleanliness analysis technologies based on image processing, which use high-definition cameras to image the surface and perform post-analysis on the image to detect the type and quantity of particles. However, these methods often have the following shortcomings: first, image processing methods are sensitive to the shape and size of particles and are easily affected by noise, leading to misjudgment; second, traditional particle analysis methods cannot effectively process large amounts of particle data and cannot generate cleanliness reports in real time; finally, existing technologies lack a feedback mechanism for real-time adjustment of cleaning equipment, resulting in instability in cleaning results.

[0004] Therefore, there is an urgent need for a more accurate and automated cleanliness analysis method that can effectively analyze particle characteristics, calculate particle concentration, and adjust the working parameters of the cleaning equipment through a feedback mechanism, thereby improving the accuracy of cleanliness analysis and cleaning results. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides the following technical solution: a particle analysis method of an automatic cleanliness analysis system, comprising the following steps:

[0006] Step 1: Obtaining particle image data on the clean surface through a cleanliness analysis device;

[0007] Step 2: preprocessing the particle image data, including grayscale, denoising, Gaussian filtering and edge enhancement, to improve the clarity of particle boundaries;

[0008] Step 3: Analyze the processed particle image through image recognition algorithm, and use threshold segmentation, morphological analysis and deep learning model to identify the shape, size and classification information of the particles;

[0009] Step 4: Calculate the number of particles and particle concentration based on the shape, size and distribution information of the particles;

[0010] Step 5: Generate a particle analysis report, which includes the particle quantity, particle concentration, particle morphology analysis results and cleanliness assessment results;

[0011] Step 6. Based on the particle analysis report, the cleaning equipment is adjusted through the feedback mechanism to optimize the cleaning process and further improve the cleanliness.

[0012] Preferably, in step one, the particle image data on the clean surface is acquired by a high-definition camera or microscope configured on the cleanliness analysis device. The cleanliness analysis device can be applied to various surfaces on industrial production lines, including but not limited to metal, glass or plastic surfaces. In order to ensure the accuracy and efficiency of image acquisition, the device is equipped with automatic scanning and positioning functions, which can automatically scan the entire area of ​​the target surface and accurately locate the area to be detected. The acquired particle image data is a two-dimensional image or a high-resolution image. The resolution of the image is high enough to capture the morphology and size of tiny particles, and can clearly display the distribution, shape and boundary characteristics of the particles. Once the image data is acquired, the image data will be transmitted to the analysis and calculation unit through the interface inside the device (the acquired image data is further processed and analyzed).

[0013] Preferably, in the step 2, the acquired image data is preprocessed to ensure the accuracy and efficiency of the subsequent particle analysis. First, the acquired two-dimensional image or high-resolution image is grayed. Since the particles on the clean surface may have different color characteristics, and these color information are usually affected by factors such as lighting conditions and background interference, in order to remove color interference and simplify the subsequent image processing process, the color image is converted into a grayscale image, and then the image is subjected to Gaussian filtering and denoising. Since the image acquisition process may be affected by factors such as changes in ambient lighting and device jitter, the image contains a certain degree of noise, which will affect the clarity of the particle boundary and the accuracy of the particle morphology. Finally, an edge enhancement algorithm is used to highlight the edges of the particle image so that the shape and boundary characteristics of the particles can be more easily identified in subsequent steps.

[0014] Preferably, in step 3, the image is first segmented using a threshold segmentation algorithm, and the particle area is effectively distinguished from the background area through statistics of pixel values. The segmented image is further optimized by morphological analysis of the particle area, including expansion, corrosion, and opening operations. These processing steps can repair the irregularity of the particle boundary, eliminate small isolated noise, and highlight the overall shape of the particle. Subsequently, the key geometric features of the particle (such as area, aspect ratio, and circularity, etc.) are calculated to provide accurate basic data for classification. The particle classification is based on a deep learning model (convolutional neural network, CNN), which identifies specific categories of particles, such as dust particles, oil droplets, etc., by combining the extracted geometric features with the classification algorithm. Through the training of large-scale samples, it can be quickly classified according to morphological features, providing accurate and detailed particle type statistics for cleanliness analysis.

[0015] Preferably, in step 4, by analyzing the shape, size and spatial distribution of the particles and combining the calculation formula, the concentration distribution of the particles is accurately evaluated to provide a scientific basis for the subsequent cleanliness evaluation. First, the system will count the total number N of particles through the particle image data extracted in the above steps, and determine the geometric parameters (area, boundary, etc.) of each particle. These parameters directly reflect the distribution characteristics of the particles on the clean surface and provide basic data for concentration calculation. The calculation formula for particle concentration is:

[0016]

[0017] Where C represents the particle concentration, A i is the area of ​​the ith particle, A r is the total area of ​​the analysis region, d i is the distance from the ith particle to the center of the reference area, and σ is the standard deviation of the Gaussian function, which is used to control the influence of the distance between the particle and the center of the area on the concentration calculation. This calculation method utilizes the relative position of the particle and the center of the area, and can more accurately reflect the distribution density of the particles.

[0018] Preferably, in step five, the acquired particle data is mainly counted and processed by analysis software to generate a detailed particle analysis report, the report content includes the number of particles, particle concentration and particle morphological analysis results (aspect ratio, circularity, etc.), and combined with the cleanliness evaluation results to provide a comprehensive cleaning performance analysis. The system first counts the total number of particles and the particle concentration per unit area, analyzes the particle distribution in combination with the particle shape characteristics (aspect ratio and circularity), and classifies the particle types to generate classification results. Through the particle distribution diagram, particle size distribution histogram and particle type classification diagram in the report, the spatial distribution, size characteristics and category proportion of the particles are intuitively presented in a visual form, helping users to quickly understand the particle distribution law and pollution characteristics of the clean surface. The report also combines the cleanliness evaluation index to quantitatively evaluate the overall cleaning effect of the clean surface, and provides improvement suggestions, providing a scientific basis for users to optimize cleaning operations and adjust cleaning equipment parameters.

[0019] Preferably, in step six, this step mainly dynamically adjusts the working parameters of the cleaning equipment through a feedback mechanism to improve the cleaning efficiency and effect. The particle analysis report provides particle concentration, particle distribution and cleanliness evaluation results. When the system detects that the particle concentration exceeds the preset threshold or is unevenly distributed, the feedback mechanism will trigger the cleaning equipment to adjust the working mode, such as switching to a higher intensity cleaning mode, increasing the cleaning intensity or increasing the cleaning frequency. In addition, the system can also optimize the working path of the cleaning equipment according to the spatial characteristics of the particle distribution (such as a significantly high particle concentration in a certain area), and concentrate on cleaning areas with more serious pollution. The feedback mechanism can achieve rapid response based on preset threshold rules, and support optimization combined with dynamically adjusted learning algorithms to improve cleaning efficiency. The system can also record the adjusted cleaning parameters and their corresponding cleaning effects to provide data support for further optimization of subsequent cleaning tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flowchart of step 1 provided for this application;

[0021] Figure 2 A flow chart of step 2 provided for this application;

[0022] Figure 3 A flowchart of step 3 provided for this application;

[0023] Figure 4 A flowchart of step 4 provided for this application;

[0024] Figure 5 A flowchart of step 5 provided for this application;

[0025] Figure 6A flowchart of step six provided for this application;

[0026] Figure 7 A flow chart of the method steps provided for this application. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0028] As described in the background art, conventional cleaning analysis technologies have the disadvantages of low efficiency, low precision, and insufficient accuracy of cleanliness analysis.

[0029] In order to solve this technical problem, a particle analysis method of an automatic cleanliness analysis system is proposed.

[0030] Specifically, please refer to Figure 1-7 , the particle analysis method of the automatic cleanliness analysis system comprises the following steps:

[0031] Step 1: Obtaining particle image data on the clean surface through a cleanliness analysis device;

[0032] Step 2: preprocessing the particle image data, including grayscale, denoising, Gaussian filtering and edge enhancement, to improve the clarity of particle boundaries;

[0033] Step 3: Analyze the processed particle image through image recognition algorithm, and use threshold segmentation, morphological analysis and deep learning model to identify the shape, size and classification information of the particles;

[0034] Step 4: Calculate the number of particles and particle concentration based on the shape, size and distribution information of the particles;

[0035] Step 5: Generate a particle analysis report, which includes the particle quantity, particle concentration, particle morphology analysis results and cleanliness assessment results;

[0036] Step 6. Based on the particle analysis report, the cleaning equipment is adjusted through the feedback mechanism to optimize the cleaning process and further improve the cleanliness.

[0037] In the step 1, the particle image data on the clean surface is acquired by a high-definition camera or microscope configured on the cleanliness analysis device. The cleanliness analysis device can be applied to various surfaces on industrial production lines, including but not limited to metal, glass or plastic surfaces. In order to ensure the accuracy and efficiency of image acquisition, the device is equipped with automatic scanning and positioning functions, which can automatically scan the entire area of ​​the target surface and accurately locate the area to be detected. The acquired particle image data is a two-dimensional image or a high-resolution image. The resolution of the image is high enough to capture the morphology and size of tiny particles, and can clearly display the distribution, shape and boundary characteristics of the particles. Once the image data is acquired, the image data will be transmitted to the analysis and calculation unit through the interface inside the device (the acquired image data will be further processed and analyzed).

[0038] In the step 2, the acquired image data is preprocessed to ensure the accuracy and efficiency of subsequent particle analysis. First, the acquired two-dimensional image or high-resolution image is grayed. Since the particles on the clean surface may have different color characteristics, and these color information are usually affected by factors such as lighting conditions and background interference, in order to remove color interference and simplify the subsequent image processing process, the color image is converted into a grayscale image. Subsequently, the image is subjected to Gaussian filtering and denoising. Since the image acquisition process may be affected by factors such as changes in ambient lighting and equipment jitter, the image contains a certain degree of noise, which will affect the clarity of the particle boundary and the accuracy of the particle morphology. Finally, the edge enhancement algorithm is used to highlight the edge of the particle image so that the shape and boundary characteristics of the particles can be more easily identified in the subsequent steps.

[0039] In the step three, the image is first segmented by a threshold segmentation algorithm, and the particle area is effectively distinguished from the background area by counting pixel values. The segmented image is further optimized by morphological analysis of the particle area, including dilation, erosion, and opening operations. These processing steps can repair the irregularity of the particle boundary, eliminate small isolated noise, and highlight the overall shape of the particle. Subsequently, the key geometric features of the particles (such as area, aspect ratio, and circularity) are calculated to provide accurate basic data for classification. Particle classification is based on a deep learning model (convolutional neural network, CNN). The specific categories of particles, such as dust particles, oil droplets, etc., are identified by combining the extracted geometric features with the classification algorithm. Through large-scale sample training, it can be quickly classified according to morphological features, providing accurate and detailed particle type statistical information for cleanliness analysis.

[0040] In step 4, by analyzing the shape, size and spatial distribution of the particles and combining the calculation formula, the concentration distribution of the particles is accurately evaluated to provide a scientific basis for the subsequent cleanliness evaluation. First, through the particle image data extracted in the previous step, the system will count the total number of particles N and determine the geometric parameters (area, boundary, etc.) of each particle. These parameters directly reflect the distribution characteristics of the particles on the clean surface and provide basic data for concentration calculation. The calculation formula for particle concentration is:

[0041]

[0042] Where C represents the particle concentration, A i is the area of ​​the ith particle, A r is the total area of ​​the analysis region, d i is the distance from the ith particle to the center of the reference area, and σ is the standard deviation of the Gaussian function, which is used to control the influence of the distance between the particle and the center of the area on the concentration calculation. This calculation method utilizes the relative position of the particle and the center of the area, and can more accurately reflect the distribution density of the particles.

[0043] In step five, the acquired particle data is mainly counted and processed by analysis software to generate a detailed particle analysis report. The report includes the number of particles, particle concentration, and particle morphology analysis results (aspect ratio, circularity, etc.), and provides a comprehensive cleaning performance analysis in combination with the cleanliness assessment results. The system first counts the total number of particles and the particle concentration per unit area, analyzes the particle distribution in combination with the particle shape characteristics (aspect ratio and circularity), and classifies the particle types to generate classification results. Through the particle distribution diagram, particle size distribution histogram, and particle type classification diagram in the report, the spatial distribution, size characteristics, and category proportions of the particles are intuitively presented in a visual form, helping users to quickly understand the particle distribution law and contamination characteristics of the clean surface. The report also combines the cleanliness assessment indicators to quantitatively evaluate the overall cleaning effect of the clean surface, and provides improvement suggestions, providing a scientific basis for users to optimize cleaning operations and adjust cleaning equipment parameters.

[0044] In step six, this step mainly uses a feedback mechanism to dynamically adjust the working parameters of the cleaning equipment to improve cleaning efficiency and effectiveness. The particle analysis report provides particle concentration, particle distribution and cleanliness evaluation results. When the system detects that the particle concentration exceeds the preset threshold or is unevenly distributed, the feedback mechanism will trigger the cleaning equipment to adjust the working mode, such as switching to a higher intensity cleaning mode, increasing the cleaning intensity or increasing the cleaning frequency. In addition, the system can also optimize the working path of the cleaning equipment according to the spatial characteristics of the particle distribution (such as a significantly high particle concentration in a certain area), and concentrate on cleaning areas with more serious pollution. The feedback mechanism can achieve rapid response based on preset threshold rules, and support optimization combined with dynamically adjusted learning algorithms to improve cleaning efficiency. The system can also record the adjusted cleaning parameters and their corresponding cleaning effects to provide data support for further optimization of subsequent cleaning tasks.

[0045] The particle analysis method of the automatic cleanliness analysis system provided by the present invention has the following beneficial effects:

[0046] Efficient particle detection and analysis: Through high-definition cameras or microscopes combined with automatic scanning and positioning functions, the system can quickly and accurately obtain particle image data, greatly improving the efficiency of particle detection on clean surfaces, and is suitable for a variety of industrial scenarios.

[0047] Accurate image processing and classification: Using advanced image grayscale processing, Gaussian filtering denoising, edge enhancement, morphological analysis and deep learning algorithms to achieve accurate extraction and classification of particle morphological features, ensuring the accuracy and robustness of particle identification.

[0048] Innovative concentration calculation method: Based on the concentration calculation formula of particle shape, size and spatial distribution, the relative position of particles and regional center is introduced into the concentration calculation model, which reflects the particle distribution density more intuitively and comprehensively, and supports accurate cleanliness evaluation.

[0049] Comprehensive analysis reporting and visualization: The system generates particle analysis reports that provide detailed information on particle count, concentration, morphological characteristics and distribution maps. The analysis results are clearly presented through visualization tools to help users quickly evaluate cleaning effects and develop optimization measures.

[0050] Intelligent feedback adjustment function: Based on the cleanliness assessment results, the system optimizes the cleaning effect by dynamically adjusting the working parameters of the cleaning equipment (such as mode, strength, frequency, etc.), and combines learning algorithms to achieve adaptive adjustment, significantly reducing the need for manual intervention.

[0051] Improve cleaning efficiency and cost-effectiveness: Through real-time feedback and optimization of cleaning strategies, the present invention effectively reduces the waste of cleaning resources and reduces equipment operating costs, while ensuring the efficiency and sustainability of the cleaning process.

[0052] Wide applicability: The present invention is applicable to particle detection and cleanliness assessment on the surfaces of various materials (such as metal, glass, plastic, etc.) in industrial production lines, and has wide industrial application value.

[0053] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral one; it can be a mechanical connection, an electrical connection, or communication with each other; it can be a direct connection, or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0054] Obviously, the embodiments described above are only some embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are given in the accompanying drawings, but they do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific embodiments, or to perform equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of the present invention, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present invention.

Claims

1. A particle analysis method for an automatic cleanliness analysis system, characterized in that: The following steps are involved: Step 1: Obtaining particle image data on the clean surface through a cleanliness analysis device; Step 2: preprocessing the particle image data, including grayscale, denoising, Gaussian filtering and edge enhancement, to improve the clarity of particle boundaries; Step 3: Analyze the processed particle image through image recognition algorithm, and use threshold segmentation, morphological analysis and deep learning model to identify the shape, size and classification information of the particles; Step 4: Calculate the number of particles and particle concentration based on the shape, size and distribution information of the particles; Step 5: Generate a particle analysis report, which includes the particle quantity, particle concentration, particle morphology analysis results and cleanliness assessment results; Step 6. Based on the particle analysis report, the cleaning equipment is adjusted through the feedback mechanism to optimize the cleaning process and further improve the cleanliness.

2. The particle analysis method of the automatic cleanliness analysis system according to claim 1, characterized in that: In the step 1, the image data of particles on the clean surface is obtained by configuring a high-definition camera or microscope on the cleanliness analysis device. The device obtains the target surface image through automatic scanning and positioning functions. The image data is a two-dimensional image or a high-resolution image, and is transmitted to the analysis and calculation unit through an interface.

3. The particle analysis method of the automatic cleanliness analysis system according to claim 1, characterized in that: In the step 2, the image data is first grayed to remove color interference and simplify image analysis, and then the gray image is denoised using a Gaussian filtering algorithm to eliminate interference that may be caused by environmental and equipment noise. Finally, an edge enhancement algorithm is used to enhance the edges of the particle image and highlight the boundaries of the particles, making them easier to identify and analyze in subsequent processing.

4. The particle analysis method of the automatic cleanliness analysis system according to claim 1, characterized in that: In the step three, the denoised and enhanced image is first segmented using a threshold segmentation algorithm to extract the boundaries of the particles, and then the particles are further processed through morphological analysis to calculate the particle area, aspect ratio, circularity and other shape features, and finally a deep learning model is used to identify and classify the particle categories and determine the particle type based on the morphological features.

5. The particle analysis method of the automatic cleanliness analysis system according to claim 1, characterized in that: In step 4, firstly, the number N of particles and the concentration of particles per unit area are calculated by the shape, size and spatial distribution of the identified particles. The calculation formula of the particle concentration is: Where C represents the particle concentration, A i is the area of ​​the ith particle, A r is the total area of ​​the analysis region, d i is the distance from the ith particle to the center of the reference area, and σ is the standard deviation of the Gaussian function, which is used to control the influence of the distance between the particle and the center of the area on the concentration calculation. This calculation method utilizes the relative position of the particle and the center of the area, and can more accurately reflect the distribution density of the particles.

6. The particle analysis method of the automatic cleanliness analysis system according to claim 1, characterized in that: In step five, the particle number, particle concentration and particle morphology analysis results are counted and processed by analysis software to generate a detailed particle analysis report, which includes the particle number, particle concentration, particle morphology analysis results, and cleanliness assessment results. The report visually displays the particle distribution diagram, particle size distribution histogram and particle type classification diagram for users to evaluate the cleaning effect and make corresponding improvement decisions.

7. The particle analysis method of the automatic cleanliness analysis system according to claim 1, characterized in that: In step six, the operating parameters of the cleaning equipment are adjusted through a feedback mechanism based on the particle concentration, particle distribution and cleanliness evaluation results provided in the particle analysis report. If the particle concentration is too high or unevenly distributed, the cleaning equipment will automatically adjust the working mode or perform higher intensity cleaning to improve the cleaning effect. This feedback mechanism can be optimized based on preset thresholds or dynamically adjusted learning algorithms to achieve more precise cleaning control.

Images