A device and method for detecting the shape of tobacco based on fluidized dispersion of tobacco and high-speed imaging technology

Abstract

The application provides a tobacco shred form detection device and method based on tobacco shred fluidization dispersion and high-speed imaging technology, which comprises a high-speed imaging system, a tobacco shred fluidization module, an illumination light source and an image processing system; the high-speed imaging module is used for collecting tobacco shred form information from different angles; the image processing system is connected with the high-speed imaging module, and the tobacco shred form features collected from multiple angles are processed to generate a three-dimensional reconstruction image of the tobacco shred, and computer algorithm is used to extract feature information such as length, width, curling degree and thickness of the tobacco shred, thereby providing technical support for tobacco shred form detection analysis and quality control.

Description

Technical Field

[0001] This invention relates to the field of tobacco processing, and more specifically, to a device and method for detecting the morphology of tobacco shreds based on fluidized dispersion and high-speed imaging technology. Background Technology

[0002] In the cigarette manufacturing industry, the morphology of tobacco shreds is a key factor affecting the quality of cigarettes. Its characteristics, such as length, width, curl, thickness and distribution, directly affect the rolling quality of cigarettes (such as weight, hardness, draw resistance, ventilation, filling state, etc.), combustion performance, sensory quality, appearance quality, as well as material consumption and production efficiency.

[0003] Traditional methods for detecting the morphology of tobacco shreds mainly rely on manual labor or simple instruments for two-dimensional projection measurements, which cannot obtain thickness and true three-dimensional curl. With the development of fluidized bed technology, existing high-speed imaging technologies based on fluidized beds (such as CN112255233B) utilize fluidized bed reactors and gas-solid two-phase flow operations, easily achieving the discretization of tobacco particles required for image acquisition and analysis. This avoids the cumbersome operation of single-layer material particle spreading in conventional image detection methods, enabling online detection on the production line. However, significant drawbacks still exist:

[0004] Setting up only a single camera or multiple cameras on the same plane cannot capture the shape of tobacco shreds simultaneously from multiple angles, resulting in a lack of three-dimensional information;

[0005] The fluidized bed dispersion structure failed to be effectively integrated with the multi-angle imaging system, and the tobacco shreds were prone to overlapping and stacking during the fluidization process, which affected the accuracy of three-dimensional reconstruction.

[0006] Therefore, there is an urgent need to develop a new device that can accurately detect the three-dimensional morphology of tobacco shreds.

[0007] In order to solve the above problems, people have been seeking an ideal technological solution. Summary of the Invention

[0008] Therefore, it is necessary to provide a tobacco morphology detection device and method based on tobacco fluidization dispersion and high-speed imaging technology to address the above-mentioned technical problems.

[0009] To achieve the above objectives, the first aspect of the present invention provides a tobacco morphology detection device based on tobacco fluidization dispersion and high-speed imaging technology, comprising a high-speed imaging system, a tobacco fluidization module, an illumination source, and an image processing system;

[0010] The tobacco fluidization module includes a feeding hopper, a compressed air generator, and a fluidized bed. The fluidized bed includes a fluidization chamber and a transparent transfer pipe. The feeding hopper is connected to the fluidization chamber through a material leveling and rolling device. The compressed air generator is connected to the fluidization chamber. A tobacco recycling device is provided at the outlet of the transparent transfer pipe.

[0011] The high-speed imaging module includes multiple high-speed cameras circumferentially distributed along a horizontal cross-section of a transparent transmission pipe, used to acquire tobacco morphology information from different angles;

[0012] The illumination source is disposed around the high-speed imaging system to provide light for the high-speed imaging system;

[0013] The image processing system is connected to the high-speed imaging module.

[0014] Understandably, in order to solve the problem of the impact of overlapping and stacking of tobacco shreds in fluidized bed on the accuracy of 3D reconstruction, this solution adopts a material leveling and rolling device combined with multiple high-speed cameras. The mechanical material leveling actively creates the best observation conditions, and then the multi-view high-speed vision accurately captures details, thus transforming online 3D morphology analysis into a stable, accurate and feasible industrial solution.

[0015] Specifically, the uniform material spreading and rolling device actively forces the tobacco shreds to disperse, thus solving the problems of overlap and obstruction, allowing the tobacco shreds to enter the observation area in a near-single-layer, spread-out state. This not only creates a stable and repeatable imaging environment, greatly reducing the complexity of subsequent image processing, but also enables the high-speed camera to capture the clear and independent instantaneous posture of each tobacco shred, fundamentally ensuring the feasibility of subsequent analysis.

[0016] Multiple high-speed cameras arranged around the tobacco ensure complete capture of its 3D contours from different directions. Their high speed, combined with precise synchronous triggering, instantly freezes the tobacco's movement, completely eliminating motion blur. The redundant information provided by multiple perspectives is particularly valuable; even with extremely slight contact, the shape of the occluded portion can be reconstructed through complementary perspectives, achieving high-fidelity reproduction of the tobacco's 3D morphology—especially the curl and true shape crucial to its processing quality. The uniform thin layer formed by the evenly rolled material allows the multi-view vision algorithm to reach its maximum efficiency, achieving precise matching of feature points and high-completeness 3D point cloud reconstruction.

[0017] Furthermore, the outlet of the fluidization chamber is a shrink-fit funnel outlet, and the transparent transfer pipe is connected to the shrink-fit funnel outlet.

[0018] Understandably, according to Bernoulli's principle, a smaller cross-section leads to an increase in airflow velocity. This ensures that the tobacco is reliably entrained and carried away by the high-speed airflow as it leaves the fluidizing chamber, preventing blockage at the outlet. Furthermore, the tapered structure of the funnel reduces flow eddies and pressure loss, allowing the kinetic energy of the fluidizing gas to be more effectively converted into transport kinetic energy.

[0019] To achieve the above objectives, a second aspect of the present invention provides a method for detecting tobacco morphology using the tobacco morphology detection device based on tobacco fluidization dispersion and high-speed imaging technology described in the first aspect, comprising the following steps:

[0020] The moisture content of the tobacco sample is balanced, a certain mass of the sample to be tested is weighed and placed into the feed hopper;

[0021] Turn on the lighting source, and set the image acquisition frequency, image resolution, and single acquisition duration of the high-speed camera in the image processing system;

[0022] The tobacco shreds to be tested are fed quantitatively through the feeding hopper and the uniform feeding and rolling device.

[0023] The quantitatively fed tobacco shreds are evenly fed into the fluidized bed. The compressed air provided by the compressed air generator fluidizes, disperses, and separates the tobacco shreds in the fluidized bed, forming dispersed and independent tobacco shred particles in the transparent pipe of the fluidized bed.

[0024] When the fluidized tobacco shreds pass through multiple high-speed cameras circumferentially distributed on the same horizontal plane in the transparent fluidized bed pipe, the high-speed cameras acquire tobacco shred image data from multiple different angles and transmit the data to the image processing system.

[0025] After acquiring images of tobacco shreds inside a fluidized bed transparent pipe, compressed air transports the tobacco shreds to a tobacco shreds recycling device to achieve the recycling and reuse of the tobacco shreds;

[0026] After receiving tobacco shred image data from multiple high-speed cameras at different angles, the image processing system reconstructs the three-dimensional morphology of the tobacco shreds to form a three-dimensional image.

[0027] For all tobacco shreds in all images acquired at each detection time point, morphological analysis is used to obtain the length, width, curl, and thickness of each tobacco shred in the image, and the characteristic length, characteristic width, characteristic curl, and characteristic thickness of the tobacco shred at that detection time point are calculated.

[0028] The characteristic length calculation formula is as follows: The RR particle size distribution equation is used to describe the length distribution characteristics of tobacco shreds samples:

[0029] (1)

[0030] (2)

[0031] In the formula, x 0.5 F(x) represents the tobacco length when the proportion of tobacco particle length is 50%; x is the tobacco length in millimeters; F(x) is the proportion of tobacco length equal to or greater than x. Calculations show that The sum of the lengths of all tobacco shreds in the sample that are equal to or greater than x, where n is the number of tobacco shreds that are equal to or greater than x. The sum of the lengths of the tobacco shreds in the sample is denoted as N, and the total number of tobacco shreds in the sample is denoted as N. A and B are the equation parameters, which are obtained by fitting equation (2).

[0032] The formula for calculating the feature width is:

[0033] (3)

[0034] In the formula, W i is the width of the i-th tobacco shred, in millimeters; m is the number of tobacco shreds, in shreds; The average width of the tobacco sample is measured in millimeters. At least five local widths were collected from a single tobacco shred, and the average value was taken as the tobacco shred width.

[0035] The formula for calculating characteristic curl is:

[0036] (4)

[0037] In the formula, S i denoted as the curl of the i-th tobacco shred; m represents the number of tobacco shreds, in units of shreds. The average curl of the tobacco sample is defined as the ratio of the actual length of a single tobacco shred to the length of the minimum bounding rectangle.

[0038] The formula for calculating the feature thickness is:

[0039] (5)

[0040] In the formula, H i Let be the thickness of the i-th tobacco shred, in millimeters; where is the average of at least 5 local thickness measurements taken on a single tobacco shred as the thickness measurement value of the single tobacco shred; m is the number of tobacco shreds, in units of shreds; The average thickness of the tobacco sample is in millimeters.

[0041] The beneficial effects of this invention are as follows:

[0042] This invention provides a tobacco morphology detection technology and method based on tobacco fluidization dispersion and high-speed imaging technology. It acquires the morphological features of fluidized and dispersed tobacco from different angles through multiple high-speed cameras circumferentially distributed along a horizontal cross section of a transparent transmission pipe, and realizes three-dimensional reconstruction of tobacco morphology. It uses a pre-set algorithm to extract feature information such as length, width, curl, and thickness of tobacco, providing technical support for improving tobacco morphology control. Attached Figure Description

[0043] Figure 1 This is a simplified structural diagram of a tobacco morphology detection device based on tobacco fluidization dispersion and high-speed imaging technology.

[0044] Figure 2 This is a schematic diagram showing the distribution of high-speed cameras;

[0045] Figure 3 This is a flowchart of a tobacco morphology detection method based on fluidized dispersion and high-speed imaging technology.

[0046] The components include: 1. High-speed camera; 2. Transparent pipe; 3. Light source; 4. Tobacco shreds; 5. Image processing system; 6. Compressed air generator; 7. Material leveling and rolling device; 8. Feed hopper; 9. Tobacco shreds recycling device. Detailed Implementation

[0047] The technical solution of the present invention will be further described in detail below through specific embodiments.

[0048] To facilitate understanding, the interactive parties and / or terms and / or custom terms involved in this invention will first be explained in conjunction with the technical solution of this invention:

[0049] Example 1

[0050] This embodiment provides a tobacco morphology detection device based on fluidized bed dispersion and high-speed imaging technology, such as... Figures 1-3 As shown, it includes a high-speed imaging system, a tobacco fluidization module, an illumination source 3, and an image processing system;

[0051] The tobacco fluidization module includes a feed hopper 8, a compressed air generator 6, and a fluidized bed. The fluidized bed includes a fluidization chamber and a transparent transfer pipe 2. The feed hopper 8 is connected to the fluidization chamber through a uniform material rolling device 7. The compressed air generator 8 is connected to the fluidization chamber. A tobacco recycling device 9 is provided at the outlet of the transparent transfer pipe 2.

[0052] The high-speed imaging module includes multiple high-speed cameras 1 circumferentially distributed along a horizontal cross-section of the transparent transmission pipe 2, used to acquire tobacco morphology information from different angles; in a preferred embodiment, the multiple high-speed cameras 1 are circumferentially and uniformly distributed around a horizontal cross-section of the transparent transmission pipe.

[0053] The illumination source 3 is disposed around the high-speed imaging system to provide light for the high-speed imaging system; in one embodiment, the illumination source 3 is an LED lamp.

[0054] The image processing system 5 is connected to the high-speed imaging module.

[0055] The aforementioned device uses a uniform material rolling device and adjustable compressed air to fully fluidize and disperse the tobacco shreds, allowing them to form independent individual tobacco shreds within the fluidized bed transparent pipe.

[0056] In practical implementation, the material leveling and rolling device includes a roller whose surface is designed with different shapes of teeth, nails, needles, or blades according to the material characteristics. These teeth are specially designed to effectively grasp and break up clumps of material while minimizing shear damage to the tobacco. The roller is usually made of stainless steel, and its surface may be polished or specially coated to reduce adhesion.

[0057] Specifically, after the tobacco shreds enter through the feed hopper 8, the uniform feeding and rolling device first uses its mechanical teeth, nails, needles, or blades to forcibly comb and disperse the loose, easily clumped incoming tobacco shreds, breaking down their internal entanglement and agglomeration structure. This ensures that the tobacco shreds entering the fluidization chamber are pre-dispersed, creating optimal starting conditions for the fluidization process. The uniform, single-layered feed can be penetrated and enveloped by the fluidizing airflow more quickly and thoroughly, greatly improving heat and mass transfer efficiency and making the drying, expansion, or flavoring processes more uniform and consistent. Then, by controlling the rotation speed or linkage of the rolling device, this uniformized tobacco flow is continuously and stably injected into the fluidization chamber in a quantitative manner, thereby significantly improving the process stability and product consistency of the entire system.

[0058] Furthermore, multiple high-speed cameras arranged around the tobacco ensure that the three-dimensional contours of the tobacco shreds can be captured completely from different directions. Their high-speed characteristics, combined with precise synchronous triggering, can instantly freeze the dynamics of the tobacco shreds, completely eliminating motion blur. Even in the presence of extremely slight contact, the shape of the obscured part can be reconstructed through complementary perspectives, thereby achieving a high-fidelity reproduction of the three-dimensional morphology of the tobacco shreds—especially the curl and true shape that are crucial to its processing quality.

[0059] The feed hopper 8 is disposed on the outer side wall of the fluidization chamber. A shut-off valve and a feed pipe are sequentially disposed at the lower part of the feed hopper 8. The feed pipe is provided with the material leveling and rolling device 7, and one end of the feed pipe is connected to the side wall of the fluidization chamber.

[0060] The outlet of the fluidizing chamber is a shrinkage funnel outlet, and the transparent transmission pipe 2 is connected to the shrinkage funnel outlet.

[0061] The compressed air generator 6 is located below the fluidizing chamber. The air outlet of the compressed air generator 6 is also equipped with a flow regulating valve and an air inlet pipe, and the air inlet pipe is connected to the bottom of the fluidizing chamber.

[0062] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0063] Example 2

[0064] Based on the same inventive concept, this application also provides a method for detecting tobacco morphology using a tobacco morphology detection device based on tobacco fluidization dispersion and high-speed imaging technology as described above.

[0065] The method for detecting tobacco morphology based on fluidized bed dispersion and high-speed imaging technology includes:

[0066] S1. Quantitative feeding of tobacco shreds, specifically as follows:

[0067] Turn on the illumination source, and set the image acquisition frequency, image resolution, and single acquisition duration of the high-speed camera in the image processing system. In one embodiment, the image acquisition frequency is 60-120 frames / s; the image resolution is 4096*3072; and the total sampling time is positively correlated with the amount of material. For example, a complete sampling of a 200g sample takes approximately 2 to 4 minutes. This process involves continuously taking multiple photos, with the exposure time of each photo set between 1 / 60 and 1 / 120 of a second.

[0068] The moisture content of the tobacco sample is balanced, a certain mass of the sample to be tested is weighed and placed into the feed hopper 8.

[0069] The tobacco shreds to be tested are fed quantitatively from the feed hopper 8 through the uniform feeding and rolling device 7. In one embodiment, the feeding rate is 50-100 g / min.

[0070] S2, tobacco fluidization, specifically:

[0071] The tobacco shreds are uniformly fed into the fluidized bed using a metered feeder. Compressed air supplied by the compressed air generator 6 fluidizes, disperses, and separates the tobacco shreds within the fluidized bed, forming dispersed and independent tobacco particles within the transparent pipe 2 of the fluidized bed. In one embodiment, the compressed air velocity of the compressed air generator is 2-4 m / s.

[0072] S3. A high-speed camera captures images of tobacco shreds, specifically:

[0073] When the fluidized tobacco shreds 4 pass through multiple high-speed cameras 1 that are evenly distributed circumferentially on the same horizontal plane in the fluidized bed transparent pipe 2, the high-speed cameras acquire tobacco shred image data from multiple different angles and transmit the data to the image processing system 5.

[0074] S4. Tobacco shred recycling, specifically:

[0075] After acquiring images of tobacco shreds inside the fluidized bed transparent pipe 2, compressed air transports the tobacco shreds to the tobacco shreds recycling device 9, realizing the recycling and reuse of the tobacco shreds.

[0076] S5. The image processing system reconstructs the acquired image data, specifically as follows:

[0077] After receiving tobacco image data from multiple high-speed cameras 1 at different angles, the image processing system 5 reconstructs the three-dimensional morphology of the tobacco.

[0078] S6. Extract feature values ​​based on the reconstructed three-dimensional tobacco morphology, specifically as follows:

[0079] For all tobacco shreds in all images acquired at each detection time point, morphological analysis is used to obtain the length, width, curl, and thickness of each tobacco shred in the image, and the characteristic length, characteristic width, characteristic curl, and characteristic thickness of the tobacco shred at that detection time point are calculated.

[0080] Furthermore, a method for characterizing the characteristic length of batches of tobacco shreds:

[0081] The following Rosin-Rammler-Bennet (RR) particle size distribution equation is used to describe the length distribution characteristics of tobacco samples:

[0082] (2)

[0083] In the formula, x represents the length of the tobacco shreds, in millimeters (mm).

[0084] F(x) – The percentage of tobacco shreds with a length equal to or greater than x, i.e., the percentage of the cumulative length of tobacco shreds with a length equal to or greater than x to the total cumulative length of the tobacco sample, %

[0085] A and B are equation parameters.

[0086] Furthermore, F(x) can be determined by the following formula:

[0087]

[0088] In the formula, —The sum of the lengths of all tobacco shreds in the sample that are equal to or greater than x, where n is the number of tobacco shreds that are equal to or greater than x;

[0089] —The total length of the tobacco shreds in the sample, where N is the total number of tobacco shreds in the sample:

[0090] By fitting the relationship between F(x) and x, the constants and B can be determined.

[0091] The feature length value of the sample to be tested is calculated using the following formula:

[0092] (1)

[0093] In the above formula, the characteristic length x 0.5 The length of tobacco shreds is the length of tobacco shreds when the proportion of tobacco shreds in the total length (x) is 50%.

[0094] Furthermore, a method for characterizing the characteristic width of batches of tobacco shreds:

[0095] The characteristic width of tobacco shreds is calculated using the variable diameter circle method. The maximum diameter of the variable diameter circle is used as the local width of the tobacco shreds. The average value of the average value of the local width of a single tobacco shreds is taken by collecting the local width at multiple locations (≥5 locations) on a single tobacco shred.

[0096] The arithmetic mean of the widths of all tobacco particles is used as the width value of the tested tobacco sample. The calculation formula is as follows:

[0097] (3)

[0098] In the formula, W i — Width of the i-th tobacco strand, in millimeters (mm);

[0099] m—Number of tobacco shreds, in units of shreds;

[0100] —Average width of tobacco sample, in millimeters (mm).

[0101] Furthermore, a method for characterizing the curvature of batches of tobacco shreds:

[0102] The minimum bounding rectangle method uses the ratio of the actual length of a single tobacco shred to the length of the minimum bounding rectangle as the curvature of the tobacco shred.

[0103] The arithmetic mean of the curliness of all tobacco shreds was used as the curliness of the test sample, calculated as follows:

[0104] (4)

[0105] In the formula, S i — Curl of the i-th tobacco strand;

[0106] m – Number of tobacco shreds, in units of shreds;

[0107] —Average curl of tobacco sample.

[0108] Furthermore, a method for characterizing the characteristic thickness of batches of tobacco shreds:

[0109] The thickness of a single tobacco shred is measured by collecting thickness data at multiple local locations (≥5 locations) along the shredded tobacco. The average value is then used as the thickness of the sample. The arithmetic mean of the thicknesses of all tobacco particles is used as the final thickness value of the tested tobacco sample. The calculation formula is as follows:

[0110] (5)

[0111] In the formula, H i —Thickness of the i-th tobacco strand, in millimeters (mm);

[0112] m—Number of tobacco shreds, in units of shreds;

[0113] —Average thickness of tobacco sample, in millimeters (mm).

[0114] This scheme uses multiple high-speed cameras circumferentially distributed along a horizontal cross-section of a transparent transmission pipe to collect the morphological features of fluidized and dispersed tobacco shreds from different angles, and realizes three-dimensional reconstruction of the tobacco shred morphology. It also uses a pre-defined algorithm to extract feature information such as the length, width, curl, and thickness of the tobacco shreds, providing technical support for improving the control of tobacco shred morphology.

[0115] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them; although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can still be made to the specific implementation of the present invention or equivalent substitutions can be made to some technical features without departing from the spirit of the technical solutions of the present invention, and all such modifications and substitutions should be covered within the scope of the technical solutions claimed in the present invention.

Claims

1. A tobacco morphology detection device based on fluidized bed dispersion and high-speed imaging technology, characterized in that: Includes a high-speed imaging system, a tobacco fluidization module, an illumination source, and an image processing system; The tobacco fluidization module includes a feeding hopper, a compressed air generator, and a fluidized bed. The fluidized bed includes a fluidization chamber and a transparent transfer pipe. The feeding hopper is connected to the fluidization chamber through a material leveling and rolling device. The compressed air generator is connected to the fluidization chamber. A tobacco recycling device is provided at the outlet of the transparent transfer pipe. The high-speed imaging module includes multiple high-speed cameras circumferentially distributed along a horizontal cross-section of a transparent transmission pipe, used to acquire tobacco morphology information from different angles; The illumination source is disposed around the high-speed imaging system to provide light for the high-speed imaging system; The image processing system is connected to the high-speed imaging module.

2. The tobacco morphology detection device based on fluidized bed dispersion and high-speed imaging technology according to claim 1, characterized in that: The feed hopper is located on the outer side wall of the fluidizing chamber. A shut-off valve and a feed pipe are sequentially arranged at the lower part of the feed hopper. The feed pipe is equipped with the material leveling and rolling device, and one end of the feed pipe is connected to the side wall of the fluidizing chamber.

3. The tobacco morphology detection device based on fluidized bed dispersion and high-speed imaging technology according to claim 1, characterized in that: The outlet of the fluidizing chamber is a shrink-fit funnel outlet, and the transparent transfer pipe is connected to the shrink-fit funnel outlet.

4. The tobacco morphology detection device based on fluidized bed dispersion and high-speed imaging technology according to claim 1, 2, or 3, characterized in that: The compressed air generator is located below the fluidizing chamber. The air outlet of the compressed air generator is also equipped with a flow regulating valve and an air inlet pipe, and the air inlet pipe is connected to the bottom of the fluidizing chamber.

5. The tobacco morphology detection device based on fluidized bed dispersion and high-speed imaging technology according to claim 4, characterized in that: Multiple high-speed cameras are evenly distributed around a horizontal cross-section of the transparent transmission pipe.

6. The tobacco morphology detection device based on fluidized bed dispersion and high-speed imaging technology according to claim 4, characterized in that: The lighting source is an LED lamp.

7. The tobacco morphology detection device based on fluidized bed dispersion and high-speed imaging technology according to claim 4, characterized in that: The image processing system processes tobacco morphology information collected from different angles to generate a three-dimensional reconstructed image of the tobacco, and calculates the length, width, curl, and thickness of the tobacco, providing technical support for tobacco morphology detection, analysis, and quality control.

8. A method for detecting tobacco morphology using the tobacco morphology detection device based on tobacco fluidization dispersion and high-speed imaging technology as described in claims 1 to 7, characterized in that, Includes the following steps: The moisture content of the tobacco sample is balanced, a certain mass of the sample to be tested is weighed and placed into the feed hopper; Turn on the lighting source, and set the image acquisition frequency, image resolution, and single acquisition duration of the high-speed camera in the image processing system; The tobacco shreds to be tested are fed quantitatively through the feeding hopper and the uniform feeding and rolling device. The quantitatively fed tobacco shreds are evenly fed into the fluidized bed. The compressed air provided by the compressed air generator fluidizes, disperses, and separates the tobacco shreds in the fluidized bed, forming dispersed and independent tobacco shred particles in the transparent pipe of the fluidized bed. When the fluidized tobacco shreds pass through multiple high-speed cameras circumferentially distributed on the same horizontal plane in the transparent fluidized bed pipe, the high-speed cameras acquire tobacco shred image data from multiple different angles and transmit the data to the image processing system. After acquiring images of tobacco shreds inside a fluidized bed transparent pipe, compressed air transports the tobacco shreds to a tobacco shreds recycling device to achieve the recycling and reuse of the tobacco shreds; After receiving tobacco shred image data from multiple high-speed cameras at different angles, the image processing system reconstructs the three-dimensional morphology of the tobacco shreds to form a three-dimensional image. For all tobacco shreds in all images acquired at each detection time point, morphological analysis is used to obtain the length, width, curl, and thickness of each tobacco shred in the image, and the characteristic length, characteristic width, characteristic curl, and characteristic thickness of the tobacco shred at that detection time point are calculated. The characteristic length calculation formula is as follows: The RR particle size distribution equation is used to describe the length distribution characteristics of tobacco shreds samples: （1） （2） In the formula, x 0.5 F(x) represents the tobacco length when the proportion of tobacco particle length is 50%; x is the tobacco length in millimeters; F(x) is the proportion of tobacco length equal to or greater than x. Calculations show that The sum of the lengths of all tobacco shreds in the sample that are equal to or greater than x, where n is the number of tobacco shreds that are equal to or greater than x. The sum of the lengths of the tobacco shreds in the sample is denoted as N, and the total number of tobacco shreds in the sample is denoted as N. A and B are the equation parameters, which are obtained by fitting equation (2). The formula for calculating the feature width is: （3） In the formula, W i is the width of the i-th tobacco shred, in millimeters; m is the number of tobacco shreds, in shreds; The average width of the tobacco sample is measured in millimeters. At least five local widths were collected from a single tobacco shred, and the average value was taken as the tobacco shred width. The formula for calculating characteristic curl is: （4） In the formula, S i denoted as the curl of the i-th tobacco shred; m represents the number of tobacco shreds, in units of shreds. The average curl of the tobacco sample is defined as the ratio of the actual length of a single tobacco shred to the length of the minimum bounding rectangle. The formula for calculating the feature thickness is: （5） In the formula, H i Let be the thickness of the i-th tobacco shred, in millimeters; where is the average of at least 5 local thickness measurements taken on a single tobacco shred as the thickness measurement value of the single tobacco shred; m is the number of tobacco shreds, in units of shreds; The average thickness of the tobacco sample is in millimeters.

Images