Method for measuring the density of a liquid alloy in electrostatic suspension in a space station

Abstract

The present application relates to a kind of for the determination method of liquid alloy density in electrostatic suspension state in space station, after obtaining the photo of liquid alloy in electrostatic suspension state in space station, first photo is converted into pixel matrix and carries out gray scale processing, then deduct photo background by matrix operation.Further, the best binary grayscale threshold of pixel matrix is calculated, and pixel matrix is carried out binary processing.Then all boundary coordinates in pixel matrix are obtained by matrix operation, and each connected boundary is distinguished.Excluding bright spot boundary, the boundary coordinates and center coordinates of liquid alloy are obtained.Then the boundary equation of liquid alloy is obtained by using multi-order Legendre equation fitting liquid alloy boundary coordinates.Finally, the volume of liquid alloy is obtained by rotating integral to the boundary equation of liquid alloy, and then the density value of liquid alloy is obtained.The density value of the present application can be obtained by processing liquid alloy photo, and can be used for the determination of containerless liquid alloy density in electrostatic suspension in space station.

Description

Technical Field

[0001] This invention pertains to alloy density determination methods, specifically a method for determining the density of liquid alloys in an electrostatically suspended state within a space station. Background Technology

[0002] Reference 1, "Zhou Z, Mukherjee S, Rhim W K. Measurement of thermophysical properties of molten silicon using an upgraded electrostatic levitator[J]. Journal of Crystal Growth, 2003, 257(3-4):350-358," discloses a method for determining the density of liquid alloys based on electrostatic levitation on the ground. It uses liquid alloys with a diameter less than 2 mm for electrostatic levitation experiments, and measures their density by photographing them. However, due to the small size of the liquid alloy and the influence of the ground's gravitational field, the sphericity of the liquid alloy is poor, reducing the accuracy of the density measurement. In the microgravity environment of the space station, the liquid alloy exhibits a perfect spherical shape under the influence of surface tension; it is also larger in size than the liquid alloy electrostatically levied on the ground; and the levitation voltage is lower, resulting in less deformation of the liquid alloy caused by the electrostatic field. Therefore, the density measurement of liquid alloys under electrostatic levitation in the space station is more accurate.

[0003] Reference 2, "B, Paul Franois Paradis A, et al. Materials properties measurements and particle beam interactions studies using electrostatic levitation[J]. Materials Science and Engineering:R:Reports,2014,76(feb.):1-53," discloses a method for determining the density of electrostatic levitation by adding a background light source. It uses a high-brightness monochromatic background light source, a filter, and a CMOS camera to acquire clear images of the liquid alloy outline. However, this method requires a high-brightness monochromatic background with a high-power light source; the filter needs to be installed and removed for each experiment; and the CMOS camera has a small field of view, making it easy for the suspended liquid alloy to escape the image's field of view. Furthermore, the CMOS camera requires real-time focusing adjustments. Such a complex experimental method is not suitable for determining the density of containerless, electrostatically levitated liquid alloys in a space station without astronaut involvement.

[0004] The electrostatic levitation experimental device, installed in the containerless materials experimental cabinet aboard the core module of the Chinese space station, has been officially put into operation and can be used to determine the density of containerless liquid alloys. Compared with ground-based electrostatic levitation density determination, the space station's electrostatic levitation provides the hardware foundation for accurate liquid alloy determination due to the larger size, better sphericity, and less susceptibility to electrostatic fields of the suspended liquid alloys. However, due to the size limitations and highly integrated layout of the containerless experimental cabinet on the space station, the captured images contain bright spots caused by reflections from the cavity and the surface of the liquid alloy. The accurate boundaries of the liquid alloy in the images cannot be directly obtained, rendering existing ground-based electrostatic levitation density determination image processing methods unsuitable for determining the density of liquid alloys under electrostatic levitation conditions on the space station. Furthermore, due to the special nature of the electrostatic levitation experimental device being mounted on the space station, it is not easy to modify or adjust the experimental device in real time, and the experimental time on the space station is limited and precious. Therefore, it is particularly important to invent a method suitable for determining the density of liquid alloys under electrostatic levitation conditions on the space station. Summary of the Invention

[0005] Technical problems to be solved

[0006] To avoid the shortcomings of existing technologies, this invention proposes a method for determining the density of liquid alloys in an electrostatically suspended state in a space station, overcoming the limitations of existing methods and solving existing problems.

[0007] Technical solution

[0008] A method for determining the density of a liquid alloy in an electrostatically suspended state in a space station, characterized by the following steps:

[0009] Step 1: Take a photo of the alloy sample with mass m, P0, and a background photo, P. B And N photos of liquid alloy P n (n=1,2,3…N) is converted into an X×Y pixel matrix Q0, Q B and Q n (n=1,2,3…N), the diameter of the spherical standard of the alloy sample is d0;

[0010] Step 2: For pixel matrix Q0, Q B and Q n (n=1,2,3…N) are converted to grayscale to obtain grayscale pixel matrices G0, G... B and G n (n = 1, 2, 3…N):

[0011] G(x,y)=aRed(x,y)+bGreen(x,y)+cBlue(x,y)

[0012] Where x is the number of rows in the pixel matrix, y is the number of columns in the pixel matrix, Red is the red color component matrix in the pixel matrix, Green is the green color component matrix in the pixel matrix, Blue is the blue color component matrix in the pixel matrix, a is the red color component coefficient, b is the green color component coefficient, and c is the blue color component coefficient.

[0013] Step 3: Compare the grayscale pixel matrix G0 of the standard sample with the grayscale pixel matrices G of the N liquid alloy samples. n Background subtraction is performed on (n = 1, 2, 3…N) to obtain pixel matrices M0 and M... n (n = 1, 2, 3…N):

[0014] M n =G n -G B (n = 0, 1, 2…N)

[0015] Step 4: Calculate the grayscale pixel matrix M n Optimal grayscale threshold w n n = 0, 1, 2…N:

[0016]

[0017] σ 2 (h)=p f (h)(μ f (h)-μ G (h)) 2 +p b (h)(μ b (h)-μ G (h)) 2

[0018] Where: h is the grayscale variable, L is the number of grayscale levels, and z l Let p be the number of elements with grayscale value l in the pixel matrix, S be the total number of elements in the pixel matrix, and p be the number of elements with grayscale value l in the pixel matrix. f (h) represents the proportion of foreground pixels to the total number of pixels, p b (h) represents the proportion of background pixels to the total number of pixels, μ f (h) represents the mean gray level of the foreground, μ b (h) represents the average gray level of the background, μ G (h) represents the average gray level of the pixel matrix. M is the nth pixel matrix n Using h as the grayscale threshold, the inter-class variance is used to obtain the grayscale variable h corresponding to the maximum inter-class variance, which is the nth pixel matrix M. n Optimal grayscale threshold w n ;

[0019] Step 5: Utilize the optimal grayscale threshold wn (n = 0, 1, 2…N) for the pixel matrix M n Binarize (n = 0, 1, 2…N):

[0020]

[0021] Where M n (x,y) is the pixel matrix M n The value of the element at the x-th row and y-th column in the pixel matrix; x is the row number of the pixel matrix, and y is the column number of the pixel matrix;

[0022] Step 6: Use the standard pixel matrix M0 processed in Step 5 to obtain the unit pixel distance u:

[0023]

[0024] Step 7: Calculate the pixel matrix M after processing in Step 6. n The boundary matrix E of (n = 1, 2, 3…N) n (n = 1, 2, 3…N):

[0025]

[0026]

[0027]

[0028]

[0029] Where: m n (x, y) is a pixel matrix M n A 3×3 submatrix centered at the element in row x and column y, where S1 is the horizontal boundary operator and S2 is the vertical boundary operator. For the Hadamard product, ∑ represents the summation of all elements of the matrix, E n (x,y) is the boundary matrix E n The value of the element in row x and column y;

[0030] Step 8: For the boundary matrix E respectively n All connected boundaries in (n = 1, 2, 3…N) are labeled to obtain the set of boundary coordinates of the liquid alloy to be measured. and center coordinates (x0, y0) n (n = 1, 2, 3…N):

[0031]

[0032]

[0033]

[0034]

[0035] in: Boundary matrix E n The set of coordinates of the v-th connected boundary in the data. For set The number of elements, The set with the most elements is the boundary matrix E. n The corresponding set of boundary coordinates for the liquid alloy, (x0, y0). n Boundary matrix E n The corresponding center coordinates of the liquid alloy;

[0036] Step 9: Fit the boundary coordinates of the liquid alloy to be measured density using the multi-order Legendre equations. Obtain the boundary equations

[0037]

[0038] in: For a k-th order Legendre polynomial, c k The coefficients of the k-th order Legendre polynomial, To fit the boundary equations, the polynomial coefficients are obtained by fitting the equation as follows:

[0039]

[0040] r s Let be the actual distance from the s-th boundary coordinate to the center coordinate. Let c be the angle of the s-th boundary coordinate in polar coordinates. Then, fit c using the least squares method. k F(x0,y0,c0,c1,c2c) k Let ) be the objective function;

[0041] Step 10: Apply boundary equations The volume V of the liquid alloy with the density to be measured is obtained by rotational integration. n (n = 1, 2, 3…N):

[0042]

[0043] Step 11: Calculate the density ρ of the liquid alloy n (n=1,2,3…N), corresponding to temperature, thus obtaining the temperature-density curve ρ(T) of the liquid alloy:

[0044]

[0045] The proportion p of the number of foreground pixels to the total number of pixelsf (h) is:

[0046] The proportion p of the number of background pixels to the total number of pixels b (h) is:

[0047] The foreground grayscale mean μ f (h) is:

[0048] The average gray level of the background μ b (h) is:

[0049] The average gray level of the pixel matrix μ G (h) is:

[0050] Beneficial effects

[0051] This invention proposes a method for determining the density of liquid alloys in an electrostatically levitated state in a space station. After obtaining a photograph of the liquid alloy in this state, the method first converts the photograph into a pixel matrix and performs grayscale processing. Then, the background is subtracted through matrix operations. Further, the optimal binarization grayscale threshold for the pixel matrix is ​​calculated, and the pixel matrix is ​​binarized. Then, matrix operations are used to obtain the coordinates of all boundaries in the pixel matrix, and each connected boundary is distinguished. After excluding bright boundary spots, the boundary coordinates and center coordinates of the liquid alloy are obtained. Then, the boundary equation is obtained by fitting the liquid alloy boundary coordinates using a multi-order Legendre equation. Finally, the liquid alloy volume is obtained by performing a rotational integral on the liquid alloy boundary equation, and thus the density value of the liquid alloy is obtained.

[0052] The beneficial effects of this invention are as follows: Step four solves the problem that the background of the cavity in the photograph affects the density measurement of the electrostatically suspended liquid alloy in the space station; Step nine solves the problem that bright spots on the surface of the liquid alloy in the photograph affect the density measurement of the electrostatically suspended liquid alloy in the space station. Therefore, there is no need to add a background light source and filter, or perform unnecessary adjustments and focusing operations in the space station. Only a fixed density camera needs to be used to take a photograph of the liquid alloy with the density to be measured. Using this invention, the clear boundary coordinates of the liquid alloy can be obtained through photograph grayscale conversion, background subtraction, boundary calculation, and liquid alloy boundary recognition. The volume can then be calculated, and thus the density data of the liquid alloy can be obtained. Attached Figure Description

[0053] Figure 1 This is a grayscale image of a zirconium-vanadium liquid alloy in electrostatic levitation state on the space station at a temperature of 1640K.

[0054] Figure 2This is a grayscale image of the zirconium-vanadium liquid alloy after removing the background.

[0055] Figure 3 This is a binarized photograph of the zirconium-vanadium liquid alloy.

[0056] Figure 4 Through calculation Figure 3 All boundary diagrams obtained.

[0057] Figure 5 It is a distinction Figure 4 A schematic diagram after each connected boundary.

[0058] Figure 6 This is a schematic diagram of the boundary of the zirconium-vanadium liquid alloy after eliminating the interference of bright spots.

[0059] Figure 7 The temperature-volume curve of the zirconium-vanadium liquid alloy was measured using the electrostatic levitation device on the space station.

[0060] Figure 8 The temperature-density curve of the zirconium-vanadium liquid alloy was measured using the electrostatic levitation device on the space station. Detailed Implementation

[0061] The present invention will now be further described in conjunction with the embodiments and accompanying drawings:

[0062] The solution adopted by this invention to solve its technical problem is: a method for determining the density of liquid alloys in an electrostatically suspended state in a space station, characterized by including the following steps:

[0063] Step 1: The mass of the alloy sample whose density is to be measured is m; the diameter of the spherical standard used to determine the unit pixel distance of the photo is d0.

[0064] Step 2: Take background cavity photos P of the sample-free state, obtained by the density camera in the electrostatic levitation device of the space station. B 1. A spherical standard sample photograph P0 and N photographs of the liquid alloy whose density is to be measured P n (n = 1, 2, 3…N) is converted into an X×Y pixel matrix Q. B and Q n (n = 0, 1, 2, ... N).

[0065] Step 3: For the pixel matrix Q... B and Q n (n=0,1,2,…N) is converted to grayscale to obtain the grayscale pixel matrix G. B and G n (n = 0, 1, 2, ... N).

[0066] G(x,y)=aRed(x,y)+bGreen(x,y)+cBlue(x,y)(1)

[0067] Where x is the number of rows in the pixel matrix, y is the number of columns in the pixel matrix, Red is the red color component matrix in the pixel matrix, Green is the green color component matrix in the pixel matrix, Blue is the blue color component matrix in the pixel matrix, a is the red color component coefficient, b is the green color component coefficient, and c is the blue color component coefficient.

[0068] Step 4: Perform tests on the grayscale pixel matrix G. n Background subtraction is performed on (n = 0, 1, 2, ..., N) to obtain the pixel matrix M. n (n = 0, 1, 2, ... N).

[0069] M n =G n -G B n = 0, 1, 2…N(2)

[0070] Step 5: Calculate the grayscale pixel matrix M respectively. n The optimal grayscale threshold w for (n = 0, 1, 2, ..., N) n (n = 0, 1, 2, ... N).

[0071]

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078] Where h is the grayscale variable, L is the number of grayscale levels, and z l Let p be the number of elements with grayscale value l in the pixel matrix, S be the total number of elements in the pixel matrix, and p be the number of elements with grayscale value l in the pixel matrix. f (h) represents the proportion of foreground pixels to the total number of pixels, p b (h) represents the proportion of background pixels to the total number of pixels, μ f (h) represents the mean gray level of the foreground, μ b (h) represents the average gray level of the background, μ G (h) represents the average gray level of the pixel matrix. M is the nth pixel matrix n Using h as the grayscale threshold, the inter-class variance is used to obtain the grayscale variable h corresponding to the maximum inter-class variance, which is the nth pixel matrix M.n Optimal grayscale threshold w n .

[0079] Step 6: Utilize the optimal grayscale threshold w n (n = 0, 1, 2…N) for the pixel matrix M n (n=0,1,2…N) are binarized.

[0080]

[0081] Where M n (x,y) is the pixel matrix M n The value of the element at the x-th row and y-th column.

[0082] Step 7: Use the standard pixel matrix M0 processed in Step 6 to obtain the unit pixel distance u.

[0083]

[0084] Step 8: Calculate the pixel matrix M after processing in Step 6. n The boundary matrix E of (n = 1, 2, 3…N) n (n = 1, 2, 3…N).

[0085]

[0086]

[0087]

[0088]

[0089] Where m n (x, y) is a pixel matrix M n A 3×3 submatrix centered at the element in row x and column y, where S1 is the horizontal boundary operator and S2 is the vertical boundary operator. For the Hadamard product, ∑ represents the summation of all elements of the matrix, E n (x,y) is the boundary matrix E n The value of the element in row x and column y.

[0090] Step 9: For the boundary matrix E respectively n All connected boundaries in (n = 1, 2, 3…N) are labeled to obtain the set of boundary coordinates of the liquid alloy to be measured. and center coordinates (x0, y0) n (n = 1, 2, 3…N).

[0091]

[0092]

[0093]

[0094]

[0095] in Boundary matrix E n The set of coordinates of the v-th connected boundary in the data. For set The number of elements, The set with the most elements is the boundary matrix E. n The corresponding set of boundary coordinates for the liquid alloy, (x0, y0). n Boundary matrix E n The corresponding center coordinates of the liquid alloy.

[0096] Step 10: Fit the boundary coordinates of the liquid alloy to be measured density using the multi-order Legendre equations. Obtain the boundary equations

[0097]

[0098] in For a k-th order Legendre polynomial, c k The coefficients of the k-th order Legendre polynomial, To fit the boundary equations, the polynomial coefficients are obtained by fitting the equation as follows:

[0099]

[0100] r s Let be the actual distance from the s-th boundary coordinate to the center coordinate. Let c be the angle of the s-th boundary coordinate in polar coordinates. Then, fit c using the least squares method. k F(x0,y0,c0,c1,c2c) k ) is the objective function.

[0101] Step 11: Apply boundary equations The volume V of the liquid alloy with the density to be measured is obtained by rotational integration. n (n = 1, 2, 3…N).

[0102]

[0103] Step 12: Calculate the density ρ of the liquid alloy. n (n=1,2,3…N), corresponding to the temperature, the temperature-density curve ρ(T) of the liquid alloy can be obtained.

[0104]

[0105] The present invention will now be illustrated using the density determination of zirconium-vanadium liquid alloy as an example.

[0106] Step 1: The mass of the zirconium-vanadium alloy sample to be tested is m = 0.0705 grams; the diameter of the spherical standard used to determine the unit pixel distance of the photograph is d0 = 0.2673 centimeters.

[0107] Step 2: Take background cavity photos P of the sample-free state, obtained by the density camera in the electrostatic levitation device of the space station. B Photograph P0 of the spherical standard sample and 1716 photographs of the zirconium-vanadium liquid alloy to be tested for density P n (n = 1, 2, 3… 1716) is converted to a 512 × 512 pixel matrix Q. B and Q n (n = 0, 1, 2, ..., 1716).

[0108] Step 3: For the pixel matrix Q... B and Q n (n = 0, 1, 2… 1716) are converted to grayscale to obtain the grayscale pixel matrix G. B and G n (n=0,1,2…1716), converting G1 into an image yields... Figure 1 .

[0109] G(x,y)=aRed(x,y)+bGreen(x,y)+cBlue(x,y) (24)

[0110] Where x is the number of rows in the pixel matrix, y is the number of columns in the pixel matrix, Red is the red color component matrix in the pixel matrix, Green is the green color component matrix in the pixel matrix, Blue is the blue color component matrix in the pixel matrix, a is the red color component coefficient with a value of 0.299, b is the green color component coefficient with a value of 0.587, and c is the blue color component coefficient with a value of 0.114.

[0111] Step 4: Perform tests on the grayscale pixel matrix G. n Background subtraction is performed on (n = 0, 1, 2… 1716) to obtain the pixel matrix M. n (n=0,1,2…1716), converting M1 into an image yields... Figure 2 .

[0112] M n =G n -G B n = 0, 1, 2…1716 (25)

[0113] Step 5: Calculate the grayscale pixel matrix M respectively.n The optimal grayscale threshold w for (n = 0, 1, 2… 1716) n (n=0,1,2…1716), where w1=151.011.

[0114]

[0115]

[0116]

[0117]

[0118]

[0119]

[0120]

[0121] Where h is the grayscale variable, L = 256 is the number of grayscale levels, and z l Let p be the number of elements with grayscale value l in the pixel matrix, and S = 512 × 512 be the total number of elements in the pixel matrix. f (h) represents the proportion of foreground pixels to the total number of pixels, p b (h) represents the proportion of background pixels to the total number of pixels, μ f (h) represents the mean gray level of the foreground, μ b (h) represents the average gray level of the background, μ G (h) represents the average gray level of the pixel matrix. M is the nth pixel matrix n Using h as the grayscale threshold, the inter-class variance is used to obtain the grayscale variable h corresponding to the maximum inter-class variance, which is the nth pixel matrix M. n Optimal grayscale threshold w n .

[0122] Step 6: Utilize the optimal grayscale threshold w n (n = 0, 1, 2…1716) respectively for pixel matrix M n (n=0,1,2…1716) is binarized, and the binarized M1 is converted into an image to obtain Figure 3 .

[0123]

[0124] Where M n (x,y) is the pixel matrix M n The value of the element at the x-th row and y-th column.

[0125] Step 7: Using the standard pixel matrix M0 processed in Step 6, obtain the unit pixel distance u = 0.00166 cm / pixel.

[0126]

[0127] Step 8: Calculate the pixel matrix M after processing in Step 6. n The boundary matrix E of (n = 1, 2, 3, ..., 1716) n (n=1,2,3…1716), converting E1 into an image yields... Figure 4 .

[0128]

[0129]

[0130]

[0131]

[0132] Where X = 512, Y = 512, m n (x, y) is a pixel matrix M n A 3×3 submatrix centered at the element in row x and column y. For horizontal boundary operators, For vertical boundary operators, For the Hadamard product, ∑ represents the summation of all elements of the matrix, E n (x,y) is the boundary matrix E n The value of the element in row x and column y.

[0133] Step 9: For the boundary matrix E respectively n Label all connected boundaries in (n = 1, 2, 3… 1716), and after labeling E1, we can obtain... Figure 5 The first connected boundary is marked in blue, and the second connected boundary is marked in red. This yields the set of boundary coordinates for the zirconium-vanadium liquid alloy. The center coordinates of the zirconium-vanadium liquid alloy are (x0, y0). n (n = 1, 2, 3… 1716), will After drawing, you can get Figure 6 The center coordinates are (x0, y0)1 = (246, 231.5).

[0134]

[0135]

[0136]

[0137]

[0138] Where X = 512, Y = 512, Boundary matrix E n The set of coordinates of the v-th connected boundary in the Ω region, card(Ω) v ) is the set Ω v The number of elements, The set with the most elements is the boundary matrix E. n The corresponding set of boundary coordinates for the zirconium-vanadium liquid alloy, where s is its boundary coordinate number (x0, y0). n Boundary matrix E n The corresponding center coordinates of the zirconium-vanadium liquid alloy.

[0139] Step 10: Fit the boundary coordinates of the zirconium-vanadium liquid alloy using the 6th order Legendre equation. The boundary equations for zirconium-vanadium liquid alloys were obtained.

[0140]

[0141] in For a k-th order Legendre polynomial, c k The coefficients of the k-th order Legendre polynomial, To fit the boundary equations, the polynomial coefficients are obtained by fitting the equation as follows:

[0142]

[0143] r s Let be the actual distance from the s-th boundary coordinate to the center coordinate. Let c be the angle of the s-th boundary coordinate in polar coordinates. Then, fit c using the least squares method. k F(x0,y0,c0,c1,c2…c6) is the objective function.

[0144] Step 11: Apply the boundary equations of zirconium-vanadium liquid alloys respectively. The volume V of the zirconium-vanadium liquid alloy is obtained by rotational integration. n (n=1,2,3…1716), where V1=0.011551 cubic centimeters. Corresponding to the temperature, the temperature-volume curve V(T) of the zirconium-vanadium liquid alloy can be obtained, as shown below. Figure 7 .

[0145]

[0146] Step 12: Calculate the density ρ of the zirconium-vanadium liquid alloy. n (n=1,2,3…1716), where ρ1=6.125 g / cm³. Corresponding to temperature, the temperature-density curve ρ(T) of the zirconium-vanadium liquid alloy can be obtained, such as… Figure 8 .

[0147]

Claims

1. A method for determining the density of a liquid alloy in electrostatically levitated state in a space station, characterized by The steps are as follows: Step 1: convert the m alloy sample photo P0, background photo P B and N liquid alloy photos P n (n = 1, 2, 3…N) into X x Y order pixel matrix Q0, Q B and Q n (n = 1, 2, 3…N), the spherical sample diameter of the alloy sample is d0; Step 2: Grayscale processing is performed on the pixel matrices Q0, Q B and Q n (n = 1, 2, 3…N) to obtain the grayscale pixel matrices G0, G B and G n (n = 1, 2, 3…N): G(x, y) = aRed(x, y) + bGreen(x, y) + cBlue(x, y) Wherein x is the row number of the pixel matrix, y is the column number of the pixel matrix, Red is the red color component matrix in the pixel matrix, Green is the green color component matrix in the pixel matrix, Blue is the blue color component matrix in the pixel matrix, a is the red color component coefficient, b is the green color component coefficient, and c is the blue color component coefficient; Step 3: Background deduction is performed on the sample gray pixel matrix G0 and N liquid alloy gray pixel matrices G n (n = 1, 2, 3…N) to obtain pixel matrices M0 and M n (n = 1, 2, 3…N): M n = G n - G B (n = 0, 1, 2... N) Step 4: Calculate the gray scale pixel matrix M n optimal gray scale threshold w n n = 0, 1, 2... N: σ 2 (h) = p f (h) (μ f (h) - μ G (h)) 2 + p b (h) (μ b (h) - μ G (h)) 2 wherein: h is the gray variable, L is the gray level, z l is the number of elements with gray value of l in the pixel matrix, S is the total number of elements in the pixel matrix, p f (h) is the proportion of the number of foreground pixels to the total number of pixels, p b (h) is the proportion of the number of background pixels to the total number of pixels, μ f (h) is the mean value of foreground gray, μ b (h) is the mean value of background gray, μ G (h) is the mean value of the pixel matrix gray, is the nth pixel matrix M n The inter-class variance when h is the gray threshold value, the gray variable h corresponding to the maximum inter-class variance is the optimal gray threshold value w n of the nth pixel matrix M n ; Step 5: using the optimal gray threshold w n (n = 0, 1, 2...N) respectively to the pixel matrix M n (n = 0, 1, 2...N) to carry on the binary processing: wherein M n (x,y) is the value of the element in the xth row and yth column of the pixel matrix M n x is the number of rows of the pixel matrix, and y is the number of columns of the pixel matrix. Step 6: Obtain the unit pixel distance u by using the sample pixel matrix M0 processed in step 5: Step 7: Calculate the pixel matrix M after step six processing respectively n Boundary matrix E of (n = 1, 2, 3...N) n (n = 1, 2, 3...N): a n (x,y) = ∑(m n (x,y) ⊙ S1) b n (x,y) = ∑(m n (x,y) ⊙ S2) wherein: m n (x,y) is the pixel matrix M n the 3x3 sub-matrix centered on the element of the x-th row and y-th column, S1 is the horizontal border operator, S2 is the vertical border operator, is the Hadamard product, ∑ denotes the sum over all element values of a matrix, E n (x,y) is the border matrix E n the value of the element of the x-th row and y-th column. Step 8: Label all the connected boundaries in the boundary matrix E n (n = 1, 2, 3…N) to obtain the liquid alloy boundary coordinate set of the density to be measured and the center coordinates (x0, y0) n (n = 1, 2, 3…N): wherein: is the boundary matrix E n is the vth connected boundary coordinate set in the boundary matrix E is the number of elements in the set is the set with the most elements, i.e. the boundary matrix E is the set with the most elements, i.e. the boundary matrix E n is the corresponding liquid alloy boundary coordinate set, (x0, y0) n is the boundary matrix E n is the center coordinate of the corresponding liquid alloy Step 9: Fitting the liquidus boundary coordinates of the test density with a multi-order Legendre equation Obtaining the boundary equation where: is a kth order Legendre polynomial, c k is a kth order Legendre polynomial coefficient, is a fit boundary equation, polynomial coefficients are fit by: r s is the actual distance of the s-th boundary coordinate to the center coordinate, is the angle of the s-th boundary coordinate in polar coordinates, fitted with least squares k , F(x0, y0, c0, c1, c2…c k ) is the objective function; Step 10: Solving the boundary equation The volume of the liquid alloy V is obtained by integrating the rotation n (n = 1, 2, 3... N): Step 11: Obtain the density of liquid alloy p n (n = 1, 2, 3...N), corresponding to the temperature, that is, the temperature-density curve of the liquid alloy p(T):

2. The method for measuring the density of a liquid alloy in an electrostatically levitated state in a space station according to claim 1, wherein: The proportion p of the number of foreground pixels in the total number of pixels f (h) is:

3. The method for measuring the density of a liquid alloy in an electrostatically levitated state in a space station according to claim 1, wherein: The proportion p of the number of background pixels in the total number of pixels b (h) is:

4. The method for measuring the density of a liquid alloy in an electrostatically levitated state in a space station according to claim 1, wherein: the foreground gray level mean μ f (h) is:

5. The method for measuring the density of liquid alloy in electrostatic levitation in space station according to claim 1, wherein: the background gray level mean μ b (h) is:

6. The method for measuring the density of a liquid alloy in an electrostatically levitated state in a space station according to claim 1, wherein: The mean value of the gray scale of the pixel matrix μ G (h) is:

Images