Method for detecting mixing uniformity of conductive powder

Abstract

The invention relates to the technical field of conductive powder mixing uniformity detection, and discloses a conductive powder mixing uniformity detection method, which comprises the following steps: acquiring a conductive powder image, and converting the conductive powder image into a grayscale image; setting a gray level according to the gray value of each powder before mixing, and carrying out gray level conversion on the gray image; gray level co-occurrence matrixes of the converted gray level images under different interval parameters and differences between the gray level co-occurrence matrixes under different intervals are calculated; and judging whether the conductive powder is uniformly mixed or not according to the difference between the gray-level co-occurrence matrixes. By adopting the scheme of the invention, the mixing uniformity of the powder can be accurately judged according to the difference between the conductive powder textures obtained under different parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of detecting the mixing uniformity of conductive powder, and particularly to a method for detecting the mixing uniformity of conductive powder. Background Art

[0002] At present, conductive powder is widely used in the fields of electronic manufacturing, conductive coatings, composite materials, printed circuit boards, etc. The quality and uniformity of conductive powder have a direct impact on its performance in applications, especially in occasions requiring high conductivity, such as batteries, sensors, and conductive materials. In various industrial applications, the uniformity of conductive powder is a key factor affecting the performance of the final product. For example, in the manufacture of batteries or conductive coatings, if the powder is not evenly mixed, it will lead to local conductivity differences, thereby affecting the overall performance and stability of the product. Therefore, how to accurately and efficiently detect the mixing uniformity of conductive powder is an important issue in industrial production.

[0003] Traditional methods for detecting the mixing uniformity of conductive powder generally rely on manual sampling, physical measurement, or chemical analysis. These methods often have the disadvantages of complex operation, long time, high cost, and it is difficult to achieve efficient, real-time, and non-destructive detection. Summary of the Invention

[0004] The object of the present invention is: in view of the above problems, the present invention proposes a method for detecting the mixing uniformity of conductive powder, which judges the mixing uniformity of the powder based on the differences between the conductive powder textures obtained under different parameters.

[0005] To achieve the above object, the present invention provides a method for detecting the mixing uniformity of conductive powder, including the following steps: Obtain an image of conductive powder and convert the image of conductive powder into a grayscale image; Set gray levels according to the gray values of each powder before mixing, and perform gray level conversion on the grayscale image; Calculate the gray level co-occurrence matrix of the converted grayscale image at different interval parameters and the differences between the gray level co-occurrence matrices at different intervals; Judge whether the conductive powder is evenly mixed according to the differences between the gray level co-occurrence matrices.

[0006] Wherein, when calculating the gray level co-occurrence matrix, the number of times the pixel values of two pixel points with an interval of i in the target direction are (a, b) is counted as the element value of the a-th row and b-th column in the gray level co-occurrence matrix in the target direction at interval k. Calculate the gray level co-occurrence matrices in different target directions and take the average value as the gray level co-occurrence matrix of the conductive powder at interval i.

[0007] Optionally, when setting the gray level, sort the gray values of each powder before mixing from small to large, and set the gray level according to the magnitudes of the gray values before and after.

[0008] Optionally, when calculating the difference between gray-level co-occurrence matrices at different intervals, calculate the texture feature vector of the conductive powder through the gray-level co-occurrence matrix, and calculate the difference between the texture feature vectors of the conductive powder at different intervals. Among them, the texture feature vector of the conductive powder includes: the entropy, energy value, and angular second moment of the conductive powder. The value range of the difference is from 0 to positive infinity, and the larger the difference value, the greater the difference between the gray-level co-occurrence matrices.

[0009] Optionally, set the interval of the gray-level co-occurrence matrix according to the image size.

[0010] Optionally, the calculation method of the interval is: randomly generate a number between 0 and the image size as the interval; the total number of randomly generated numbers is the number of times the gray-level co-occurrence matrix is calculated.

[0011] Optionally, when the difference between the gray-level co-occurrence matrices is less than the set threshold, the mixing uniformity of the conductive powder is qualified; otherwise, the mixing uniformity of the conductive powder is unqualified.

[0012] Compared with the prior art, the method for detecting the mixing uniformity of a conductive powder in an embodiment of the present invention has the beneficial effect that: according to the gray values of the powders to be mixed, set the gray level of the image of the conductive powder after mixing, and after converting the gray level, calculate the gray-level co-occurrence matrix of the conductive powder under different interval parameters, calculate the difference between the gray-level co-occurrence matrices at different intervals, and judge whether the conductive powder is mixed evenly according to the difference. Description of the Drawings

[0013] Figure 1 is a flowchart of the method from step S1 to step S4 in the method for detecting the mixing uniformity of a conductive powder in an embodiment of the present invention. Detailed Embodiments

[0014] The following combines the drawings and embodiments to further describe in detail the specific embodiments of the present invention. The following embodiments are used to illustrate the present invention but are not used to limit the scope of the present invention.

[0015] As Figure 1 shown, a method for detecting the mixing uniformity of a conductive powder in a preferred embodiment of an embodiment of the present invention includes the following steps: S1: Obtain an image of the conductive powder and convert the image of the conductive powder into a grayscale image; In an embodiment of the present application, an industrial camera can be used to capture an image of the mixed conductive powder, convert the image into a grayscale image, and judge whether the powder is mixed evenly through the grayscale image.

[0016] S2: Set the gray level according to the gray values of each powder before mixing, and perform gray level conversion on the gray image; It should be noted that the conductive powder is composed of different powders mixed together. The characteristics of various powders on the gray image are different, that is, the pixel values are different. When various powders are mixed evenly, the statistical characteristics of the pixel values at different positions / regions of the conductive powder on the image are the same, that is, the texture of each region on the image is the same. Therefore, in order to make the result of the gray level co-occurrence matrix more in line with powder detection, first calculate the gray values of each powder before mixing, and set the gray level according to the gray values of each powder before mixing.

[0017] As an example, the gray values of each powder before mixing can be sorted from small to large. After sorting, set the gray level according to the size of the front and back gray values.

[0018] Specifically, when n powders are mixed, the powder gray values are sorted from small to large as , then the converted gray levels are as follows: Among them, represents the gray value after converting the gray level, represents the gray value before converting the gray level.

[0019] After converting the gray level of the conductive powder gray image, calculating its gray level co-occurrence matrix includes: counting the number of times the pixel values of two pixel points with a target direction interval of i are (a, b), as the element value of the a-th row and b-th column in the gray level co-occurrence matrix in the target direction when the interval is k, and calculating the average value of the gray level co-occurrence matrices in different target directions as the gray level co-occurrence matrix of the conductive powder when the interval is i. Among them, the target direction refers to the direction defined in the image plane for analyzing the relationship between pixels.

[0020] Specifically, first, after performing gray level conversion on the conductive powder gray image, the gray value of each pixel point in the image has been remapped to a specific gray level range. The purpose of this step is to strengthen the feature differences of different powders on the gray image, making the subsequent texture analysis more accurate.

[0021] Next, start calculating the gray-level co-occurrence matrix. The gray-level co-occurrence matrix is a statistical tool used to describe the texture features of an image, which can reflect the correlation between pixel points in the image in terms of gray values. Among them, in the image, the spacing parameter (i) represents the distance between two pixel points. For example, when calculating the gray-level co-occurrence matrix, different spacing distances can be set, such as a spacing of 1, a spacing of 2, etc. A spacing of 1 represents two adjacent pixel points, and a spacing of 2 represents the distance between two pixel points with one pixel point in between. There are multiple directions that can be considered for the image, such as 0° (horizontal to the right), 45°, 90° (vertical downward), 135°, etc. When calculating the gray-level co-occurrence matrix, it is necessary to perform statistics separately for each target direction.

[0022] Then, for the target direction and spacing (i), count the number of times the pixel pairs that meet the conditions (i.e., two pixel points are separated by a distance i in the target direction and their gray values are a and b respectively) appear in the image. For example, in the case where the target direction is 0° and the spacing is 1, if there are two adjacent pixel points (horizontally adjacent) in the image and their gray values are 85 and 170 respectively, then the number of occurrences of the pixel pair (a = 85, b = 170) will be incremented by 1. This counting process is carried out on the entire gray image. For each possible combination of gray values (a, b), the number of times they appear in the target direction and spacing will be recorded.

[0023] Furthermore, take the number of occurrences of the pixel pairs (a, b) obtained from the above statistics as the element value at the corresponding position in the gray-level co-occurrence matrix. Specifically, when the spacing is k, the element value at the a-th row and b-th column of the gray-level co-occurrence matrix in the target direction is the number of times the pixel pair (a, b) appears in the image. For example, if the pixel pair (a = 85, b = 170) appears 10 times in a certain target direction and spacing, then the element value at the 85-th row and 170-th column of the corresponding gray-level co-occurrence matrix is 10. Among them, for each spacing parameter, it is necessary to calculate the gray-level co-occurrence matrices in different target directions. This is because the texture features in different directions may be different, and considering the information in multiple directions can more comprehensively reflect the texture of the image.

[0024] Finally, take the average of the gray-level co-occurrence matrices in different target directions. This step can integrate the texture information in different directions to obtain a comprehensive gray-level co-occurrence matrix, which can better represent the texture features of the conductive powder image at this spacing parameter. Finally, the gray-level co-occurrence matrix of the conductive powder at a spacing of i is obtained.

[0025] It should be noted that when the conductive powder is evenly mixed, the gray-level co-occurrence matrices at different intervals should be the same. Since the gray-level co-occurrence matrix is obtained through statistics, for example, the probability of the pixel pair (a, b) appearing at an interval of 1 and the probability of the pixel pair (a, b) appearing at an interval of 2 should be the same. Because in the case of even mixing, the textures of different regions and different angles of the conductive powder are the same, and the texture features of the conductive powder obtained at different intervals are also the same.

[0026] S3: Calculate the gray-level co-occurrence matrix of the converted gray-scale image under different interval parameters and the difference between the gray-level co-occurrence matrices at different intervals; In the embodiment of the present application, the difference between the gray-level co-occurrence matrices of the conductive powder at different intervals is calculated to determine whether it is evenly mixed. The smaller the difference, the more evenly it is mixed. The difference calculation formula of the gray-level co-occurrence matrix of the conductive powder is as follows: Among them, represents the difference between the i-th gray-level co-occurrence matrix and the j-th gray-level co-occurrence matrix, represents the i-th gray-level co-occurrence matrix, represents the j-th gray-level co-occurrence matrix, represents the inner product of the two gray-level co-occurrence matrices, represents the norm of the i-th gray-level co-occurrence matrix, represents the norm of the i-th gray-level co-occurrence matrix. In this embodiment, the value range of the difference is from -1 to 1. The larger the value, the smaller the difference between the two matrices. The difference is normalized. The larger the normalized value, the greater the difference, and the smaller the value, the smaller the difference.

[0027] In the embodiment of the present application, the method for calculating the difference between the two gray-level co-occurrence matrices further includes: The above calculation method reflects the difference between the two matrices. The gray-level co-occurrence matrix is used to calculate the texture features of the image. Therefore, the texture features of the conductive powder at different intervals can be calculated, and then the difference between the texture features of the conductive powder at different intervals can be calculated. In one embodiment, the entropy, energy value, angular second moment, etc. of the conductive powder are calculated through the gray-level co-occurrence matrix to obtain a vector reflecting the texture features of the conductive powder, and then the difference between the texture feature vectors of the conductive powder at different intervals is calculated. The calculation method is as follows: Among them, represents the difference between the i-th gray-level co-occurrence matrix and the j-th gray-level co-occurrence matrix, represents the k-th element value of the texture feature vector of the conductive powder obtained from the i-th gray-level co-occurrence matrix, It represents the k-th element value of the texture feature vector of the conductive powder obtained from the j-th gray-level co-occurrence matrix. In this embodiment, the value range of the difference is from 0 to positive infinity. The larger this value is, the greater the difference between the two matrices. Normalize it.

[0028] So far, the differences of the gray-level co-occurrence matrices at different intervals are obtained.

[0029] S4: Determine whether the conductive powder is evenly mixed according to the differences between the gray-level co-occurrence matrices.

[0030] In the embodiment of the present application, the gray-level co-occurrence matrix is a statistical method for calculating texture features. Its disadvantage is that the calculation amount is relatively large. If the gray-level co-occurrence matrices at each interval are calculated and then the differences between these matrices are calculated, the calculation amount will be even larger. Therefore, in this embodiment, a fixed number of intervals are set according to the size of the image. The specific interval size is generated according to random numbers. The formula for setting the number of times according to the image size is as follows: Among them, represents the number of times of calculating the gray-level co-occurrence matrix, is a constant, representing the relationship between the image size and the number of times of calculating the gray-level co-occurrence matrix, represents the width of the image, represents the height of the image, represents taking the maximum value.

[0031] Specifically, the calculation method of the above interval is: randomly generate a number between 0 and the image size as the interval, and the total number of randomly generated numbers is the number of times of calculating the gray-level co-occurrence matrix. After obtaining the gray-level co-occurrence matrices at different intervals, calculate the differences of the gray-level co-occurrence matrices at different intervals. When the difference is less than the set threshold, the evenness of the conductive powder mixing is qualified; otherwise, the evenness of the conductive powder mixing is unqualified.

[0032] As an example, determine the number of times of calculating the gray-level co-occurrence matrix according to the size (width and height) of the image. The formula is as follows: the number of times = α × max(image width, image height). Among them, α is used to adjust the scale of the number of times. For example, if the width of the image is 1024 pixels, the height is 768 pixels, and α = 0.001, then the number of times is: the number of times = 0.001 × max(1024, 768) = 0.001 × 1024 = 1.024 ≈ 1.

[0033] Among them, according to the above number of times, randomly generate interval values between 0 and the image size (usually take the width or height of the image). For example, if the image width is 1024 pixels and the number of times is 1, then randomly generate an interval value, such as 50.

[0034] For each randomly generated interval value, calculate the gray-level co-occurrence matrix. The calculation method of the gray-level co-occurrence matrix is as follows: Count the number of times that the pixel values of two pixel points with an interval of i are (a, b) in the target direction (such as 0°, 45°, 90°, 135°). Fill these statistical values into the gray-level co-occurrence matrix to form the element value of the a-th row and b-th column of the matrix. Take the average of the gray-level co-occurrence matrices in different directions to obtain the comprehensive gray-level co-occurrence matrix at this interval. Calculate the difference between the gray-level co-occurrence matrices at different intervals. Compare the calculated difference value with a preset threshold. If the difference value is less than the set threshold, it is considered that the mixing uniformity of the conductive powder is qualified; otherwise, it is considered that the mixing is uneven.

[0035] For example, there is a gray-scale image of conductive powder with a width of 1024 pixels and a height of 768 pixels. There are three types of powders before mixing, and the gray levels are 85, 170, and 255 respectively. Select α = 0.001 to calculate the number of times of calculating the gray-level co-occurrence matrix: Calculate the number of times of calculating the gray-level co-occurrence matrix: The number of calculations = 0.001 × max(1024, 768) = 0.001 × 1024 = 1.024 ≈ 1; Randomly generate an interval: Randomly generate an interval value between 0 and 1024. Suppose the generated interval value is 50; Calculate the gray-level co-occurrence matrix with an interval of 50: Count the number of occurrences of pixel pairs (a, b) with an interval of 50 in the target directions of 0°, 45°, 90°, and 135°, construct the gray-level co-occurrence matrix in each direction, and take the average to obtain the comprehensive gray-level co-occurrence matrix M1; Calculate the gray-level co-occurrence matrix of another interval (interval of 30): Similarly, count the number of occurrences of pixel pairs (a, b) with an interval of 30 in the directions of 0°, 45°, 90°, and 135°, construct the gray-level co-occurrence matrix in each direction, and take the average to obtain the comprehensive gray-level co-occurrence matrix M2; Calculate the difference value: Calculate the difference value between M1 and M2 using the above difference formula. Suppose the calculated difference value is 0.05; Judge the mixing uniformity: The preset threshold is 0.1. Since the calculated difference value of 0.05 is less than the threshold of 0.1, it is judged that the mixing uniformity of the conductive powder is qualified.

[0036] In summary, the embodiment of the present invention provides a method for detecting the mixing uniformity of conductive powder. According to the gray values of the mixed powders, the gray levels of the image of the conductive powder after mixing are set. After converting the gray levels, the gray-level co-occurrence matrices of the conductive powder under different interval parameters are calculated, and the differences between the gray-level co-occurrence matrices at different intervals are calculated. According to the differences, it can be accurately judged whether the conductive powder is mixed evenly.

[0037] The above is only the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art of this technology, without departing from the counting principle of the present invention, several improvements and substitutions can be made, and these improvements and substitutions should also be regarded as the protection scope of the present invention.

Claims

1. A method for detecting the mixing uniformity of conductive powder, characterized in that, It includes the following steps: Obtain the image of the conductive powder and convert the image of the conductive powder into a grayscale image; Set the gray level according to the gray values of each powder before mixing, and perform gray level conversion on the grayscale image; Calculate the gray level co-occurrence matrix of the converted grayscale image under different interval parameters and the difference between the gray level co-occurrence matrices under different intervals; Judge whether the conductive powder is evenly mixed according to the difference between the gray level co-occurrence matrices; Among them, calculating the gray level co-occurrence matrix includes: counting the number of times that the pixel values of two pixel points with an interval of i in the target direction are (a, b), and taking it as the element value of the a-th row and b-th column in the gray level co-occurrence matrix in the target direction when the interval is k, calculating the gray level co-occurrence matrices in different target directions and taking the average value as the gray level co-occurrence matrix of the conductive powder when the interval is i.

2. The method for detecting the mixing uniformity of the conductive powder according to claim 1, wherein When setting the gray level, sort the gray values of each powder before mixing from small to large, and set the gray level according to the size of the front and back gray values.

3. The method for detecting the mixing uniformity of the conductive powder according to claim 1, characterized in that When calculating the difference between the gray level co-occurrence matrices under different intervals, calculate the texture feature vector of the conductive powder through the gray level co-occurrence matrix, and calculate the difference between the texture feature vectors of the conductive powder under different intervals.

4. The method for detecting the mixing uniformity of the conductive powder according to claim 3, wherein The texture feature vector of the conductive powder includes: the entropy, energy value and angular second moment of the conductive powder.

5. The method for detecting the mixing uniformity of the conductive powder according to claim 3, characterized in that, The value range of the difference is from 0 to positive infinity, and the larger the difference value, the greater the difference between the gray level co-occurrence matrices.

6. The method for detecting the mixing uniformity of the conductive powder according to claim 1, wherein Set the interval of the gray level co-occurrence matrix according to the image size.

7. The method for detecting the mixing uniformity of the conductive powder according to claim 6, wherein, The calculation method of the interval is: randomly generate a number between 0 and the image size as the interval; the total number of randomly generated numbers is the calculation times of the gray level co-occurrence matrix.

8. The method for detecting the mixing uniformity of the conductive powder according to claim 1, wherein When the difference between the gray level co-occurrence matrices is less than the set threshold, the mixing uniformity of the conductive powder is qualified, otherwise, the mixing uniformity of the conductive powder is unqualified.

Images