A method for rapidly detecting the stem content in cut tobacco

Abstract

The application provides a method for rapidly detecting the content of stem in tobacco, and belongs to the technical field of tobacco processing. The method comprises the following steps: placing tobacco and stem in a constant-temperature and constant-humidity environment for 48 hours of balanced placement for standby; blending the tobacco and stem to obtain a plurality of mixed blending samples with different contents of stem as mixtures; using a hydraulic instrument to compress the obtained mixtures until a preset pressure value is reached, stopping the compression operation and keeping for 200 seconds, recording the pressure value change in the 200 seconds as a stress curve; repeating the stress determination step three times; establishing a standard equation according to a fractional order mathematical model; data fitting, obtaining a relaxation curve and substituting it into the standard equation to obtain a stress-blending ratio equation; and performing the above operations on the tobacco and stem mixture with an unknown blending ratio to obtain the blending ratio.

Description

Technical Field

[0001] This invention belongs to the field of tobacco processing technology, and more specifically, relates to a rapid method for detecting the stem content in tobacco. Background Technology

[0002] Traditional cigarettes are generally composed of shredded tobacco leaves, stems, expanded tobacco, and reconstituted tobacco. The stability of the content of each component has a significant impact on the chemical, physical, and sensory properties of the finished cigarette. With the advancement of tobacco shredding technology, the utilization rate of stems is increasing. Stems play a role in reducing the harm of cigarettes, improving the filling value, enhancing the smoking quality, and reducing cigarette production costs. Their blending ratio and uniformity in cigarette formulation are also highly valued by many industrial enterprises. The uniformity and stability of the distribution of different tobacco leaves in cigarette blending are crucial for controlling the quality and sensory stability of the finished cigarette.

[0003] Currently, the main method for determining the proportion of stems in blended tobacco is to measure the stem content by analyzing the morphological characteristics, color differences, physical density, tobacco chemical characteristics, and tobacco pyrolysis characteristics of the tobacco and stems. The morphological characteristics of stems and tobacco are primarily determined by manual observation or selection and weighing using a visual magnification system. While this method is simple, it suffers from drawbacks such as being time-consuming, susceptible to human error, and unsuitable for large-scale testing. With the continuous development of stem processing technology, the morphological differences between processed stems and leaves are becoming increasingly smaller, making manual selection more difficult. Furthermore, due to the maturation of processing technologies for leaves, stems, and reconstituted tobacco, their apparent densities are becoming increasingly similar, resulting in poor repeatability and ineffectiveness when using density differences for content analysis. Secondly, using near-infrared spectroscopy to analyze the blending ratio of tobacco and stems based on the different spectral characteristics of stems and tobacco requires complex calibration and sample preparation processes, consuming considerable time, involving complex equipment, increasing experimental costs, and requiring professional personnel to operate. Currently, some researchers are utilizing the difference in chemical composition and properties between stems and tobacco shreds, which leads to different pyrolysis behaviors, to obtain mass change curves of different samples during programmed heating using thermogravimetric analyzers. By analyzing the differences in the pyrolysis curves, the content of stems in tobacco shreds can be obtained. This method requires sophisticated and complex equipment as well as supporting analytical software, and also requires highly skilled personnel. It is not suitable for rapid measurement by process quality inspectors on the front line of processing enterprises. Summary of the Invention

[0004] In summary, a more accurate, effective, low-cost method suitable for frontline personnel to quickly determine the content and blending uniformity of stems in tobacco still needs to be established. This invention provides a path and steps to solve the above problems.

[0005] This invention is implemented as follows:

[0006] This invention provides a rapid method for detecting the stem content in tobacco shreds, comprising the following steps:

[0007] S10: Sample pretreatment: Place the tobacco shreds and stems in a constant temperature and humidity environment for 48 hours for equilibration, with a temperature requirement of 22℃±1℃ and a relative humidity requirement of 60%±2%.

[0008] S20: Mixture preparation: Leaf shreds and stem shreds are mixed according to a predetermined blending ratio to obtain multiple groups of mixed samples of tobacco shreds and stem shreds with different stem shred contents as the mixture;

[0009] S30: Stress Measurement: The obtained mixture is compressed using an adjustable digital electro-hydraulic proportional hydraulic instrument until a preset pressure value is reached. The compression operation of the hydraulic instrument is stopped and held for 200 seconds. The pressure value change during these 200 seconds is recorded as the stress curve. The pressure value is the pressure data collected by the pressure sensor on the pressure head of the hydraulic instrument.

[0010] S40: Establishing the Standard Equation: Repeat the stress measurement steps three times, and using the recorded stress curves, establish the standard equation based on the fractional mathematical model.

[0011]

[0012] In the formula, σ(t) is the stress change with time, K is the viscoelastic characteristic parameter of the material, Γ is the gamma function equation, and α is the fractional order.

[0013] S50: Data Fitting: Statistical analysis software was used to perform data fitting to obtain relaxation curves for mixtures with different blending ratios. These curves were then substituted into the standard equation to obtain the stress-blending ratio equation.

[0014] Y = AX + B;

[0015] In the formula, X represents the horizontal axis, Y represents the vertical axis, A is the linear coefficient, and B is a constant;

[0016] S60: Detection of materials with unknown proportions: Stress is measured on the mixture of tobacco shreds and stems with unknown proportions to obtain the stress curve. The stress-mixing ratio equation is then substituted to obtain the mixing ratio of the mixture of tobacco shreds and stems with unknown proportions.

[0017] The stress curve calculation uses the finite element method (FEM) in numerical simulation. The specific method includes: geometrically modeling the tested mixture; discretizing the geometric model into a finite element mesh; analyzing the finite element mesh, including defining boundary conditions, applying loads, and material properties; establishing structural elements within each finite element, including selecting appropriate strain-displacement relationships and material constitutive equations to describe the object's behavior. Depending on the material properties, a linear elastic model or a nonlinear model can be chosen to represent the material's behavior; the mechanical behavior of the structural elements is represented by algebraic equations and combined into a large system of linear equations. By solving this system of equations, the displacement and stress values ​​at each node can be obtained; based on the displacement values ​​and strain-displacement relationships, the strain of each finite element is calculated. Then, according to the material constitutive equations, the strain values ​​are converted into stress values; and stress curves are plotted based on the stress values ​​at the nodes.

[0018] Based on the above technical solution, the rapid detection method for stem content in tobacco shreds of the present invention can be further improved as follows:

[0019] In the process of compressing the obtained mixture using a hydraulic instrument with adjustable digital electro-hydraulic ratio, the pressing speed of the pressure head is set to 1 mm / s.

[0020] Furthermore, during the process of using an adjustable digital electro-hydraulic ratio hydraulic instrument to compress the obtained mixture until a preset pressure value is reached, the pressure value is set to 3 kg.

[0021] The step of compressing the obtained mixture using an adjustable digital electro-hydraulic proportional hydraulic instrument until a preset pressure value is reached, specifically includes the step of obtaining the preset pressure value:

[0022] Tobacco shreds and stem shreds with different stem contents were obtained as test samples, and images of the test samples were taken to obtain the first image;

[0023] The test sample is compressed to a preset sequence pressure value using an adjustable digital electro-hydraulic ratio hydraulic instrument. The preset sequence pressure value is 20 to 50 sets, and the value is between 0.5 kg and 5 kg.

[0024] After the compressed test sample was left to stand for 15 minutes, an image was taken to obtain the second image;

[0025] Calculate the similarity between the first image and the second image, and sort the sequence pressure values ​​corresponding to the similarity in descending order of similarity to obtain a similarity sequence;

[0026] The average of the top ten stress values ​​in the similarity sequence is taken as the preset stress value.

[0027] Through the above steps, different pressure values ​​can be selected according to different ratios of tobacco and stem mixture samples, without the need for operators to make prior changes, which improves efficiency while ensuring the quality of cigarettes.

[0028] Furthermore, calculating the similarity between the first image and the second image includes: representing the first image as a first vector and the second image as a second vector; normalizing the first vector and the second vector to eliminate the influence of vector length; and calculating the cosine value according to the cosine similarity formula to represent the similarity between the first image and the second image.

[0029] The similarity between the first image and the second image ranges from -1 to 1. The closer the value is to 1, the more similar the two images are; the closer the value is to -1, the less similar the two images are; and a value of 0 indicates that the two images have no similarity.

[0030] Furthermore, the step of representing the first image as a first vector and the second image as a second vector specifically includes:

[0031] The first image and the second image are denoised and smoothed using a Gaussian filter;

[0032] The first image after noise reduction and smoothing and the second image after noise reduction and smoothing are respectively subjected to histogram equalization to adjust the gray level distribution, thus obtaining the first gray level image and the second gray level image.

[0033] The first grayscale image and the second grayscale image are filtered by enhancement filters respectively to obtain a first filtered image and a second filtered image;

[0034] Edge detection is performed on the first filtered image and the second filtered image respectively to obtain a first vector of the first image and a second vector of the second image. Further, the step of applying enhancement filters to the first grayscale image and the second grayscale image respectively uses mean filtering, a linear filtering method that averages the pixel values ​​within the filtering window to remove noise. In mean filtering, each pixel in the filter has the same weight, so the filtered pixel value is a simple average of its neighboring pixels. It is suitable for removing slight Gaussian noise, but performs poorly with salt-and-pepper noise; because salt-and-pepper noise can make pixel values ​​extremely high or low, mean filtering is particularly sensitive to such outliers, leading to blurred filtering results.

[0035] Furthermore, in the step of performing edge detection on the first filtered image and the second filtered image respectively, the edge detection methods include: Sobel edge detection method and Canyon edge detection method.

[0036] Furthermore, the first vector and the second vector are normalized using L2 norm normalization.

[0037] Furthermore, the process involves using statistical analysis software to fit the data, obtaining relaxation curves for mixtures with different blending ratios, and then substituting these curves into the standard equation to obtain the stress-blending ratio equation. This process specifically includes:

[0038] Obtain experimental data on mixtures with different blending ratios, including stress and the corresponding blending ratios;

[0039] Perform data conversion and processing, including data type conversion and null value handling;

[0040] The stress-blending ratio relationship was fitted using a linear regression method to obtain the stress-blending ratio equation.

[0041] Compared with the prior art, the beneficial effect of the rapid detection method for stem content in tobacco provided by the present invention is that by first conducting stress tests on a mixture of tobacco and stem with a known blending ratio, establishing a stress curve, and then establishing a stress-blending ratio equation based on the stress curve, the content of stem in tobacco and the blending uniformity can be rapidly and accurately determined. Attached Figure Description

[0042] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart of a rapid detection method for stem content in tobacco shreds according to an embodiment of the present invention;

[0044] Figure 2 This is a curve showing the stress relaxation of tobacco samples with different stem contents over time in an embodiment of the present invention.

[0045] Figure 3 The figures show the stress-time curves for different values ​​of fractional order α in embodiments of the present invention.

[0046] Figure 4 This is the fitting result of the filament content and fractional order α value in the first embodiment of the present invention. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0048] like Figure 1 The diagram shown is a flowchart of a rapid detection method for stem content in tobacco provided by the present invention. This method includes the following steps:

[0049] S10: Sample pretreatment: Place the tobacco shreds and stems in a constant temperature and humidity environment for 48 hours for equilibration, with a temperature requirement of 22℃±1℃ and a relative humidity requirement of 60%±2%.

[0050] S20: Blending Preparation: Leaf shreds and stem shreds were blended according to a predetermined blending ratio to obtain multiple blended samples with different stem shred contents as blends. The composition results of the test samples are shown in Table 1:

[0051] Table 1 Composition results of the test samples

[0052]

[0053]

[0054] S30: Stress Measurement: The obtained mixture is compressed using an adjustable digital electro-hydraulic proportional hydraulic instrument until the preset pressure value is reached. The compression operation of the hydraulic instrument is stopped and held for 200 seconds. The pressure value change during these 200 seconds is recorded as the stress curve. The pressure value is the pressure data collected by the pressure sensor on the pressure head of the hydraulic instrument.

[0055] For example, in the stress relaxation test, the mass of the mixture and the volume reached during pressure holding (mass of the mixture ≥ 10.0g) are determined according to the pre-designed compression pressure. The displacement distance of the indenter moving downward is calculated. A microcomputer-controlled adjustable digital electro-hydraulic proportional hydraulic instrument is used to perform the compression test on the sample to be tested. Each time, 15g of the above-mentioned blended tobacco sample is slowly loaded into the compression sleeve, maintaining its natural stacking state. During the compression process, the downward speed of the indenter is set to 1mm / s, and the maximum compression force is 3kg. After reaching the maximum compression force, the position of the indenter remains unchanged and is maintained for 200s. The experiments are conducted on the samples in Table 1 in sequence, and the stress relaxation curves of the above samples with different blending ratios over time are obtained. The stress relaxation curves of samples numbered 1, 2, 3, 4, 5, 6, and 7 are shown below. Figure 2 As shown. Keeping the displacement constant, the pressure holding time is 200s, and the stress change is recorded using a sensor on the pressure head.

[0056] Alternatively, an adjustable digital electro-hydraulic proportional hydraulic instrument can be used to compress the obtained mixture until a preset pressure value is reached. The step of obtaining the preset pressure value specifically includes:

[0057] Tobacco shreds and stem shreds with different stem contents were obtained as test samples, and images of the test samples were taken to obtain the first image;

[0058] The test sample was compressed to a preset sequence pressure value using an adjustable digital electro-hydraulic ratio hydraulic instrument. The preset sequence pressure value was 20 to 50 sets, and the value was between 0.5 kg and 5 kg.

[0059] After the compressed test sample was left to stand for 15 minutes, an image was taken to obtain the second image;

[0060] Calculate the similarity between the first image and the second image, and sort the sequence pressure values ​​corresponding to the similarity in descending order of similarity to obtain the similarity sequence;

[0061] The average of the top ten stress values ​​in the similarity sequence is taken as the preset stress value.

[0062] Furthermore, in the above technical solution, calculating the similarity between the first image and the second image includes: representing the first image as a first vector and the second image as a second vector; normalizing the first vector and the second vector to eliminate the influence of vector length; and calculating the cosine value according to the cosine similarity formula to represent the similarity between the first image and the second image.

[0063] Furthermore, the step of representing the first image as a first vector and the second image as a second vector specifically includes:

[0064] The first image and the second image are denoised and smoothed using a Gaussian filter;

[0065] The first image after noise reduction and smoothing and the second image after noise reduction and smoothing are respectively subjected to histogram equalization to adjust the gray level distribution, thus obtaining the first gray level image and the second gray level image.

[0066] The first grayscale image and the second grayscale image are respectively subjected to mean filtering through a mean filter to obtain a first filtered image and a second filtered image;

[0067] Edge detection is performed on the first filtered image and the second filtered image using either the Sobel edge detection method or the Canny edge detection method, respectively, to obtain the first vector of the first image and the second vector of the second image.

[0068] Furthermore, the first vector and the second vector are normalized using L2 norm normalization.

[0069] The formula for the L2 norm is:

[0070]

[0071] In the formula, z i Let n represent the vectors of the first and second vectors, n represents the total number of vectors of the first and second vectors, n is 30, and i represents the i-th vector, i varies from 1 to n.

[0072] To normalize z to the unit L2 norm, i.e., to establish a rule from z to z... ′ The mapping of z makes z ′ If the L2 norm is 1, then:

[0073]

[0074] The formula for cosine similarity is:

[0075]

[0076] Where, x i These are the components of the first vector, y i These are the components of the second vector.

[0077] S40: Establishing the Standard Equation: Repeat the stress measurement steps three times, and using the recorded stress curves, establish the standard equation based on the fractional mathematical model.

[0078]

[0079] In the formula, σ(t) is the stress change over time, K is the viscoelastic characteristic parameter of the material, Γ is the gamma function equation, and α is the fractional order, reflecting the difference in the relaxation time variation of the mixture. If α is 0, then K is the elastic modulus; if α is 1, then K is scaled to the viscosity coefficient. The stress-time variation curves for different values ​​of fractional order α are shown below. Figure 3 As shown.

[0080] The elastic modulus is the stress required for the mixture to produce a unit elastic deformation under the pressure. The elastic modulus is an indicator of how easily a material undergoes elastic deformation; a higher value indicates a greater stress required for elastic deformation, and conversely, a smaller elastic deformation under a given stress.

[0081] The viscosity coefficient is the tangential force required to maintain a unit velocity difference between two liquid layers of a unit area in two liquid layers that are a unit distance apart.

[0082] For example, based on the obtained stress relaxation curves, a fractional-order equation was used for fitting analysis to calculate the α value in different samples, i.e., the numerical value representing the difference in the relaxation time variation of tobacco shreds with different blending ratios. The main results are as follows:

[0083]

[0084]

[0085] S50: Data Fitting: Statistical analysis software was used to perform data fitting to obtain relaxation curves for mixtures with different blending ratios. These curves were then substituted into the standard equation to obtain the stress-blending ratio equation.

[0086] Y = AX + B;

[0087] In the formula, X represents the horizontal axis, Y represents the vertical axis, A is the linear coefficient, and B is a constant.

[0088] Furthermore, the process involves using statistical analysis software to fit the data, obtaining relaxation curves for mixtures with different blending ratios, and then substituting these curves into the standard equation to obtain the stress-blending ratio equation. This process specifically includes:

[0089] Obtain experimental data on mixtures with different blending ratios, including stress and the corresponding blending ratios;

[0090] Perform data conversion and processing, including data type conversion and null value handling;

[0091] The stress-blending ratio relationship was fitted using a linear regression method to obtain the stress-blending ratio equation.

[0092] Using the stem content of samples with different blending ratios as the ordinate and the α value of samples with different numbers as the abscissa, a linear fit was performed on the obtained data to obtain the standard curve equation for stem content:

[0093] Y = -3753.6 * X + 666.9;

[0094] Correlation coefficient R 2 =97.9%, such as Figure 4 As shown.

[0095] S60: Detection of materials with unknown proportions: Stress is measured on the mixture of tobacco shreds and stems with unknown proportions to obtain the stress curve. The stress-mixing ratio equation is then substituted to obtain the mixing ratio of the mixture of tobacco shreds and stems with unknown proportions.

[0096] For example, a certain amount of the formulated tobacco sample that needs to be measured in the actual process is taken, pretreated according to the conditions in step S10, and the stress relaxation curve of the sample is measured with time using the method in step S30 to calculate the α value; using the standard curve equation of stem content obtained in step S50, the stem content in the actual formulated sample is calculated. The sample is measured 10 times, and the relative error of the 10 experiments is calculated as shown in Table 2.

[0097] Table 2. Relative errors of 10 experiments

[0098]

[0099] The data in the table shows that the relative deviation is within 5.00%. The prediction results are accurate and reliable.

[0100] In the above steps, the boundary conditions are defined, including fixed supports and free ends; the applied loads and material properties include elastic modulus and Poisson's ratio.

[0101] Specifically, the principle of this invention is: to place tobacco shreds and stem shreds in a constant temperature and humidity environment for 48 hours for later use, wherein the temperature requirement is 22℃±1℃ and the relative humidity requirement is 60%±2%.

[0102] The leaf shreds and stem shreds were blended according to the predetermined blending ratio to obtain multiple groups of mixed samples of tobacco shreds and stem shreds with different stem shred contents as the blending material;

[0103] The mixture is compressed using an adjustable digital electro-hydraulic proportional hydraulic press until a preset pressure value is reached. The compression operation of the hydraulic press is stopped and held for 200 seconds. The pressure value change during these 200 seconds is recorded as a stress curve. The pressure value is the pressure data collected by the pressure sensor on the hydraulic press head.

[0104] Repeat the stress measurement steps three times; using the recorded stress curves, establish a standard equation based on a fractional mathematical model:

[0105]

[0106] In the formula, σ(t) is the stress change over time, K is the viscoelastic characteristic parameter of the material, Γ is the gamma function equation, and α is the fractional order.

[0107] Data fitting was performed using statistical analysis software to obtain relaxation curves for mixtures with different blending ratios. These curves were then substituted into the standard equation to obtain the stress-blending ratio equation.

[0108] Y = AX + B;

[0109] In the formula, X represents the horizontal axis, Y represents the vertical axis, A is the linear coefficient, and B is a constant;

[0110] Stress was measured on a mixture of tobacco shreds and stems with an unknown blending ratio to obtain a stress curve. The stress-blending ratio equation was then substituted to obtain the blending ratio of the mixture of tobacco shreds and stems with an unknown blending ratio.

Claims

1. A rapid method for detecting the stem content in tobacco shreds, characterized in that, Includes the following steps: S10: Sample pretreatment: Place the tobacco shreds and stems in a constant temperature and humidity environment for 48 hours for equilibration, with a temperature requirement of 22 ℃±1 ℃ and a relative humidity requirement of 60%±2%. S20: Mixture preparation: Leaf shreds and stem shreds are mixed according to a predetermined blending ratio to obtain multiple groups of mixed samples of tobacco shreds and stem shreds with different stem shred contents as the mixture; S30: Stress Measurement: The obtained mixture is compressed using an adjustable digital electro-hydraulic proportional hydraulic instrument until a preset pressure value is reached. The compression operation of the hydraulic instrument is stopped and held for 200 seconds. The pressure value change during these 200 seconds is recorded as the stress curve. The pressure value is the pressure data collected by the pressure sensor on the pressure head of the hydraulic instrument. S40: Establishing the Standard Equation: Repeat the stress measurement steps three times, and using the recorded stress curves, establish the standard equation based on the fractional mathematical model. ； In the formula, It is the change of stress over time. These are the viscoelastic characteristic parameters of the material. The equation is the gamma function equation. It is the fractional order; S50: Data Fitting: Statistical analysis software was used to perform data fitting to obtain relaxation curves for mixtures with different blending ratios. These curves were then substituted into the standard equation to obtain the stress-blending ratio equation. ； In the formula, The α value represents the sample numbering. The content of stem fibers in samples with different blending ratios. The coefficients are linear. It is a constant; S60: Detection of materials with unknown proportions: Stress is measured on the mixture of tobacco shreds and stems with unknown proportions to obtain the stress curve. The stress-mixing ratio equation is then substituted to obtain the mixing ratio of the mixture of tobacco shreds and stems with unknown proportions. In the step of compressing the obtained mixture using an adjustable digital electro-hydraulic proportional hydraulic instrument until a preset pressure value is reached, the step of obtaining the preset pressure value specifically includes: Tobacco shreds and stem shreds with different stem contents were obtained as test samples, and images of the test samples were taken to obtain the first image; The test sample is compressed to a preset sequence pressure value using an adjustable digital electro-hydraulic ratio hydraulic instrument. The preset sequence pressure value is 20 to 50 sets, and the value is between 0.5 kg and 5 kg. After the compressed test sample was left to stand for 15 minutes, an image was taken to obtain the second image; Calculate the similarity between the first image and the second image, and sort the sequence pressure values ​​corresponding to the similarity in descending order of similarity to obtain a similarity sequence; The average of the top ten stress values ​​in the similarity sequence is taken as the preset stress value. The calculation of the similarity between the first image and the second image includes: representing the first image as a first vector and the second image as a second vector; normalizing the first vector and the second vector to eliminate the influence of vector length; and calculating the cosine value according to the cosine similarity formula to represent the similarity between the first image and the second image. The steps of representing the first image as a first vector and the second image as a second vector specifically include: The first image and the second image are denoised and smoothed using a Gaussian filter; The first and second images, after noise reduction and smoothing, are subjected to histogram equalization to adjust the grayscale distribution, resulting in the first grayscale image and the second grayscale image. The first grayscale image and the second grayscale image are filtered by enhancement filters respectively to obtain a first filtered image and a second filtered image; Edge detection is performed on the first filtered image and the second filtered image respectively to obtain the first vector of the first image and the second vector of the second image.

2. The rapid detection method for stem content in tobacco shreds according to claim 1, characterized in that, During the compression process of the obtained mixture using the adjustable digital electro-hydraulic ratio hydraulic instrument, the pressing speed of the pressure head is set to 1 mm / s.

3. The rapid detection method for stem content in tobacco shreds according to claim 2, characterized in that, The process of using an adjustable digital electro-hydraulic ratio hydraulic instrument to compress the obtained mixture until a preset pressure value is reached is described, where the pressure value is set to 3 kg.

4. The rapid detection method for stem content in tobacco shreds according to claim 3, characterized in that, In the step of enhancing the first grayscale image and the second grayscale image respectively, the image enhancement method is the mean filtering method.

5. The rapid detection method for stem content in tobacco shreds according to claim 4, characterized in that, In the step of performing edge detection on the first filtered image and the second filtered image respectively, the edge detection methods include: Sobel edge detection method and Canny edge detection method.

6. The rapid detection method for stem content in tobacco shreds according to claim 5, characterized in that, The first vector and the second vector are normalized using L2 norm normalization.

7. The rapid detection method for stem content in tobacco shreds according to claim 6, characterized in that, The steps involved in using statistical analysis software to fit data, obtaining relaxation curves for mixtures with different blending ratios, and then substituting these curves into the standard equation to obtain the stress-blending ratio equation include: Obtain experimental data on mixtures with different blending ratios, including stress and the corresponding blending ratios; Perform data conversion and processing, including data type conversion and null value handling; The stress-blending ratio relationship was fitted using a linear regression method to obtain the stress-blending ratio equation.

Images