Method for accurately calculating size of fuzzy cloud and mist particles

Abstract

The invention discloses a fuzzy cloud particle size accurate calculation method, which comprises a calibration module and a detection module, and is characterized in that the calibration module is used for placing standard plates with different diameters at different Z positions for imaging to obtain a series of standard plate dot images, and dividing the standard plate dot images into three categories according to the diameters of the standard plate dot images, respectively calculating the pixel diameter and ambiguity of each image in the three categories; obtaining three data sets respectively corresponding to the three types of images; fitting is carried out on the three data sets to obtain the corresponding relation among the particle ambiguity, the pixel diameter and the real diameter in each category; and the detection module is used for calculating the physical diameter of the particles by using the corresponding relation among the ambiguity, the pixel diameter and the real diameter of the particles in each category obtained by the calibration module. By adopting the embodiment to calculate the fuzzy cloud particle size, the calculation precision of the particle size can be improved, a guarantee is provided for subsequent parameter calculation, meanwhile, the sampling space is indirectly increased, and the actual particle field distribution is truly reflected.

Description

Technical Field

[0001] This invention belongs to the field of cloud and fog particle size calculation, and specifically relates to a method for accurate calculation of fuzzy cloud and fog particle size. Background Technology

[0002] The study of particle spectrum distribution in clouds and fog is of great significance in meteorological forecasting and weather modification, and is a key physical quantity for studying the microscopic physical structure of clouds and fog. Accurate calculation of cloud and fog particle diameter is a crucial technique for calculating particle spectrum distribution, and is the basis for retrieving parameters such as inversion number concentration (NC), liquid water content (LWC), median diameter (MVD), and equivalent diameter (ED).

[0003] Due to limitations of microscope lenses and optical imaging systems, micron-sized cloud particles can be clearly imaged within the lens's depth of field. Beyond this depth, the image becomes increasingly blurry. This makes it difficult for image processing algorithms to extract the true boundary information of the particles, leading to significant errors in the calculation of cloud particle size.

[0004] The traditional approach to dealing with this problem is to discard these blurry particles directly and only process and count the particles that are clearly imaged. This results in a significant reduction in the number and types of particles in the sampling space, and the collected and statistically calculated particle samples cannot reflect the actual distribution of cloud and fog particles. Summary of the Invention

[0005] In view of the aforementioned shortcomings of existing methods for calculating the size of fuzzy cloud particles, the purpose of this invention is to propose a more accurate method for calculating the size of fuzzy cloud particles.

[0006] To achieve the aforementioned objectives, the technical solution adopted by this invention is: a method for accurately calculating the size of fuzzy cloud particles, comprising: a calibration module and a detection module; The calibration module is used for: Step 1-1, placing standard plates of different diameters at different Z positions for imaging to obtain a series of standard plate dot images from clear to blurry; Steps 1-2: Divide all collected standard plate dot images into three categories based on the diameter of the standard plate dot images: Class A: [2um, 10um] Class B: [10um, 100um] Class C: [100um, 500um]; Steps 1-3: Calculate the pixel diameter and blur level of each image in the three categories respectively; Steps 1-4: Based on the true diameter of each particle category and the pixel diameter and blurriness obtained in Steps 1-3, obtain three datasets corresponding to the three categories of images respectively; Steps 1-5: Fit the three datasets obtained in Steps 1-4 with a quadratic polynomial to obtain the correspondence between particle blur, pixel diameter and true diameter in each category; The detection module is used for: step 2-1, binarizing the particle image using a fixed threshold; Step 2-2: Use the 8-connected-component search algorithm to find connected components; Steps 2-3: Calculate the area of ​​the connected components and use Hough circle fitting to calculate their pixel diameter d; Steps 2-4: Calculate the fuzziness f corresponding to the connected components; Steps 2-5: Calculate the particle physical diameter using the correspondence between particle ambiguity, pixel diameter and true diameter obtained in steps 1-5 of the calibration module.

[0007] Furthermore, in steps 1-3, the calculation methods for pixel diameter and blurriness are as follows: Step 1-3.1: Binarize the particle image using a fixed threshold; Step 1-3.2: Use the 8-connected-component search algorithm to find connected components and calculate the width c_width, height c_height, and center coordinates (c_x, c_y) of each connected component. Step 1-3.3: Perform Hough circle fitting on each connected component to obtain the pixel diameter of the particle.

[0008] Steps 1-3.4: Crop the connected component regions to obtain the image. The clipping boundary parameters are: Steps 1-3.5: For image I crop Perform a first-order difference operation to obtain image I. diff : Steps 1-3.6, at the location , along I diff In the x-direction, with the window size as With a step size of 1, calculate the difference d between the maximum and minimum grayscale values ​​in each window. i The blurriness of the particle image is then: .

[0009] Furthermore, in steps 1-4, the three datasets corresponding to the three types of samples are as follows: .

[0010] Furthermore, the fitting formulas for fitting the three datasets obtained in steps 1-4 using quadratic polynomials in steps 1-5 are as follows: .

[0011] Furthermore, the formula for calculating the physical diameter of the particle in steps 2-5 is as follows: .

[0012] The aforementioned method for accurately calculating the particle size of fuzzy clouds and fog can improve the calculation accuracy of particle size, provide a guarantee for the calculation of subsequent index parameters, and indirectly increase the sampling space, so as to truly reflect the actual particle field distribution.

[0013] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0014] Figure 1 This is a schematic diagram illustrating the principle of microscopic imaging in this embodiment; Figure 2 This is a schematic diagram of the calibration module process in this embodiment; Figure 3 This is a schematic diagram of standard plate dot images acquired at different Z positions in the calibration module of this embodiment; Figure 4 This is a schematic diagram of the average fitting error for the three categories in this embodiment; Figure 5 This is a schematic diagram of the detection module process in this embodiment. Detailed Implementation

[0015] The features and exemplary embodiments of various aspects of the present invention will now be described in detail. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that the invention may be practiced without requiring some of these specific details. The following description of embodiments is merely intended to provide a better understanding of the invention by illustrating examples of the invention.

[0016] This embodiment provides a method for accurately calculating the particle size of fuzzy clouds and fog, including: a calibration module and a detection module; The calibration module is used for: Step 1-1, placing standard plates of different diameters at different Z positions for imaging to obtain a series of standard plate dot images from clear to blurry; For details, please refer to Figure 1This embodiment is based on the Mie scattering principle of particles. A microscope is used to perform bright field microscopic imaging of micron-sized cloud and fog particles. Then, an image processing algorithm is used to calculate parameters such as the diameter, distribution, and phase of cloud and fog particles per unit time and volume. Finally, based on the obtained parameters, the spectral distribution characteristic parameters of cloud and fog particles are calculated through an inversion algorithm.

[0017] Please see Figure 2 In this embodiment, the calibration module is performed in an optical laboratory, using "microwires and standard plates for particle volume calibration" (see Table 1). In this embodiment, standard plates of different diameters are placed at different Z-positions for imaging, with a Z-direction sampling interval of 20 μm, resulting in a series of standard plate dot images ranging from clear to blurred. (See Table 1 for details.) Figure 3 In this embodiment, standard plate dot images with a diameter of 10 μm are used as an example, and standard plate dot images are collected at different Z positions.

[0018] Table 1 Sampling range of standard plate dot image Steps 1-2: Divide all collected standard plate dot images into three categories based on the diameter of the standard plate dot images: Class A: [2um, 10um] Class B: [10um, 100um] Class C:[100um,500um].

[0019] Steps 1-3: Calculate the pixel diameter and blur level of each image in the three categories respectively; Specifically, the calculation methods for pixel diameter and blur are as follows: Step 1-3.1: Binarize the particle image using a fixed threshold; Step 1-3.2: Use the eight-connected-component search algorithm to find connected components and calculate the width c_width, height c_height, and center coordinates (c_x, c_y) of the connected components; Step 1-3.3: Perform Hough circle fitting on each connected component to obtain the pixel diameter d of the particle; Steps 1-3.4: Crop the connected component regions to obtain the image. The clipping boundary parameters are: Steps 1-3.5: For image I crop Perform a first-order difference operation to obtain image I. diff : Steps 1-3.6, at the location Along image I diff In the x-direction, with the window size as With a step size of 1, calculate the difference d between the maximum and minimum grayscale values ​​in each window. i The blurriness of the particle image is then: .

[0020] Steps 1-4: Calculate the three datasets corresponding to the three types of samples based on the true diameter D of each type of particle, the pixel diameter d obtained in step 3, and the blurriness f. Specifically, the three datasets corresponding to the three types of samples are as follows: .

[0021] Steps 1-5: Fit the three datasets obtained in step 4 with a quadratic polynomial to obtain the correspondence between particle blur, pixel diameter and true diameter in each category; Specifically, the fitting formula is: Among them, C A0 C A1 C A2 C A3 C A4 C A5 C B0 C B1 C B2 C B3 C B4 C B5 C CO C C1 C C2 C C3 C C4 C C5 For the coefficients to be determined, the average fitting error for the three categories is as follows: Figure 4 As shown.

[0022] Please see Figure 5 The detection module is used for: step 2-1, binarizing the particle image using a fixed threshold; Step 2-2: Use the 8-connected-component search algorithm to find connected components; Steps 2-3: Calculate the area of ​​the connected component Area and use Hough circle fitting to calculate its pixel diameter d. The constraint condition for the area of ​​the connected component is: 4≤Area≤490000. Steps 2-4: Calculate the ambiguity f corresponding to the connected components using steps 1-3.4 to 1-3.6 of the calibration module; Steps 2-5: Calculate the particle physical diameter using the correspondence between particle ambiguity f, pixel diameter d, and true diameter D obtained in steps 1-5 of the calibration module.

[0023] Furthermore, the formula for calculating the physical diameter of the particle in steps 2-5 is as follows: .

[0024] The method for accurately calculating cloud and fog particle size in this embodiment can improve the accuracy of particle size calculation, provide a guarantee for the calculation of subsequent index parameters, and indirectly increase the sampling space to truly reflect the actual particle field distribution.

[0025] The above description is merely a preferred embodiment of the present invention. Any simple modifications, equivalent changes, and alterations made by those skilled in the art to the above embodiments without departing from the scope of the present invention and based on the technical essence of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for accurately calculating the particle size of fuzzy clouds and fog, characterized in that, include: Calibration module, detection module; The calibration module is used for: Step 1-1, placing standard plates of different diameters at different Z positions for imaging to obtain a series of standard plate dot images from clear to blurry; Steps 1-2: Divide all collected standard plate dot images into three categories based on the diameter of the standard plate dot images: Class A: [2um, 10um] Class B: [10um, 100um] Class C: [100um, 500um]; Steps 1-3: Calculate the pixel diameter and blur level of each image in the three categories respectively; Steps 1-4: Based on the true diameter of the particles in each category and the pixel diameter and blurriness obtained in Steps 1-3, obtain the three datasets corresponding to the images of the three categories respectively; Steps 1-5: Fit the three datasets obtained in step 4 with a quadratic polynomial to obtain the correspondence between particle blur, pixel diameter and true diameter in each category; The detection module is used for: step 2-1, binarizing the particle image using a fixed threshold; Step 2-2: Use the 8-connected-component search algorithm to find connected components; Steps 2-3: Calculate the area of ​​the connected components and use Hough circle fitting to calculate its diameter d; Steps 2-4: Calculate the fuzziness f corresponding to the connected components; Steps 2-5: Calculate the particle physical diameter using the correspondence between particle ambiguity, pixel diameter and true diameter obtained in steps 1-5 of the calibration module.

2. The method for accurately calculating the particle size of fuzzy clouds and fog according to claim 1, characterized in that, The calculation methods for pixel diameter and blurriness in steps 1-3 are as follows: Step 1-3.1: Binarize the particle image using a fixed threshold; Step 1-3.2: Use the 8-connected-component search algorithm to find connected components and calculate the width c_width, height c_height, and center coordinates (c_x, c_y) of each connected component. Step 1-3.3: Perform Hough circle fitting on each connected component to obtain the pixel diameter of the particle; Steps 1-3.4: Crop the connected component regions to obtain the image. The clipping boundary parameters are: Steps 1-3.5: For image I crop Perform a first-order difference operation to obtain image I. diff : Steps 1-3.6, at the location , along I diff In the x-direction, with the window size as With a step size of 1, calculate the difference d between the maximum and minimum grayscale values ​​in each window. i The blurriness of the particle image is then: .

3. The method for accurately calculating the particle size of fuzzy clouds and fog according to claim 1, characterized in that, In steps 1-4, the three datasets corresponding to the three categories of images are as follows: 。 4. The method for accurately calculating the particle size of fuzzy clouds and fog according to claim 1, characterized in that, The fitting formulas used in steps 1-5 to fit the three datasets obtained in steps 1-4 using quadratic polynomials are as follows: 。 5. The method for accurately calculating the particle size of fuzzy clouds and fog according to claim 1, characterized in that, The formula for calculating the physical diameter of the particle in steps 2-5 is: 。

Images