Device and method for measuring material passing time of drum type equipment based on dye tracing and machine vision

Abstract

The application discloses a kind of based on dyeing trace and machine vision's drum type equipment material passing time measuring device and method, the device includes dyeing trace module, visual detection module, data processing module and power supply module;The dyeing trace module uses food-grade blue dye to dye ordinary cut tobacco to obtain trace tobacco;The visual detection module includes the feed camera being set at the inlet of drum and the outlet industrial camera being set at the discharge port, for recording the entering and leaving time of trace tobacco;The data processing module is according to the preset inlet time compensation and outlet time compensation, combined with the visual identification of image and the time mean calculation of outlet trace tobacco, according to specific formula, the actual passing time of material in drum is obtained.The application can simulate the real motion trajectory of cut tobacco in harsh industrial environment, through optimizing feeding amount, high adaptability industrial camera and specific image color threshold processing algorithm, solve the problem of difficult detection and large result error in high temperature, high humidity and dusty environment, with high measurement accuracy and without the need to recover tracer.

Description

Technical Field

[0001] This invention relates to the field of process performance inspection technology for tobacco processing equipment, specifically to a device and method for measuring the material transit time of roller-type equipment based on dyeing tracing and machine vision, which is applicable to the precise inspection and optimization of the process performance of various roller-type equipment in cigarette manufacturing workshops. Background Technology

[0002] Rotary drum equipment is the core equipment in tobacco processing, mainly including flavoring drums, feeding drums, drying drums, and rehumidification drums. The transit time (or residence time) of tobacco shreds or leaves in the drum is a key parameter that determines indicators such as the uniformity of the liquid or moisture content, which directly determines the consistency of the flavor quality and product quality of the tobacco shreds.

[0003] In actual production, the process conditions of different roller equipment vary greatly, and the internal environment is extremely complex: First, for flavoring rollers, a cold flavoring process is often used, where the processing is not heated and the material moves inside the roller at room temperature. However, slight fluctuations in the airflow inside the roller can still cause uncertainty in the movement trajectory of the tobacco shreds, and there is no significant thermal effect tracer characteristic during the flavoring application process. Second, for feeding rollers, steam is usually used to spray the liquid material. Although there is no active heating system inside the roller, the internal environment presents a strong turbulent gas-solid two-phase state due to the instantaneous high temperature of the steam and the effect of the ejected airflow. The local temperature can reach a high level, and the high temperature and high humidity environment can easily form a visual barrier. Third, for rehydration rollers, strong steam ejection also accompanies the process of adding water to the tobacco shreds or blades. The inside of the roller is filled with saturated water vapor and atomized water mist, resulting in extremely low visibility.

[0004] Existing technologies for comprehensive monitoring of cigarette production drums have several limitations: First, the device includes coded tapes and complex photoelectric detection gate modules, requiring the installation of coded tapes on the outer ring of the drum, which is difficult and costly to modify. Second, it lacks robustness in visual recognition under high humidity environments such as feeding and rehumidification, and is easily affected by saturated water vapor and atomized water mist, leading to misidentification. Third, it fails to finely control and calibrate the amount of tracer input, resulting in excessive input easily interfering with the normal material mixing ratio, while insufficient input affects the recognition rate at the back end. Fourth, the equipment layout fails to provide physical protection and algorithm optimization for the special high-temperature, high-humidity, and dusty conditions of the cigarette-making workshop, resulting in insufficient long-term operational stability.

[0005] Therefore, developing an intelligent material passage time inspection device for roller-type equipment that is accurate, efficient, real-time, and has a more streamlined structure and is more adaptable to harsh production environments is of great significance for ensuring the stability of tobacco product quality and improving process control capabilities. Summary of the Invention

[0006] To address the shortcomings of existing detection solutions, the present invention aims to provide a material transit time measurement device and method for roller-type equipment based on dyeing tracing and machine vision, which aims to solve a series of problems in the prior art, such as complex device structure, high difficulty in modification, low visual recognition accuracy in high humidity and dusty environments, and crude control of tracer feeding amount.

[0007] Therefore, the present invention provides the following technical solution:

[0008] Firstly, a material passage time measurement device for roller-type equipment based on dyeing tracing and machine vision mainly includes a dyeing tracing module, a vision detection module, a data processing module, and a power supply module. The dyeing tracing module uses a food-grade blue dye to dye ordinary tobacco shreds to obtain tracer tobacco shreds, which are then used to track the movement of the tracer tobacco shreds. The vision detection module captures image frames of the tracer tobacco shreds after they exit the roller using an industrial camera located downstream of the roller's discharge port (preferably, an industrial camera can also be located at the front end of the roller's inlet). Regarding time calculation, the time of material insertion or the time of entry into the roller captured by the inlet industrial camera can be recorded, and the first compensation time for the material to reach the roller inlet from the insertion point can be calculated by considering the speed and distance of the front-end conveyor belt. Similarly, the second compensation time for the material to reach the detection point of the outlet industrial camera from the roller outlet can be calculated by considering the speed and distance of the outlet conveyor belt. Furthermore, in terms of time calculation, the exit industrial camera is used to identify tracer tobacco and record the exit time series data of tracer tobacco effectively detected downstream of the discharge port; the data processing module in this device is connected to the industrial camera via communication. This module has a built-in image recognition algorithm based on color threshold segmentation and morphological processing, which is used to extract color features and clean up misidentified targets from the acquired images, and calculate the time average; the data processing module pre-stores the above-mentioned compensation amount, and, in combination with the aforementioned acquired parameters, calculates the actual average passage time of the material in the drum according to a predetermined algorithm.

[0009] Preferably, in this invention, the tracer tobacco shreds, after being dried and dyed, maintain the same physical morphology and motion characteristics as the tobacco shreds in their original state. Therefore, it can accurately simulate the actual tumbling, mixing, and conveying motion trajectory of ordinary tobacco shreds inside the drum. Furthermore, since food-grade dyes are used, the tracer tobacco shreds do not need to be separated and recycled from the material after the measurement is completed, and can directly enter the subsequent production process without introducing any food safety risks.

[0010] In particular, the amount of input per batch can be obtained through multiple sets of comparative experiments. In order to ensure that the export industrial camera can stably extract a sufficient number of color feature points in harsh environments, and at the same time to avoid the excessive amount from adversely affecting the mixing state of normal production materials and the uniformity of flavoring / addition, in some implementation methods, after multiple sets of comparative experiments, the amount of input per batch m was finally selected as 5g.

[0011] In addition, the shooting equipment selected is an industrial camera that can adapt to harsh industrial environments. Its installation height is 1.5m to ensure that the field of view covers the entire width of the production line; the resolution is 1920×1080, the capture frame rate is 30 frames / second, and the camera is equipped with a dustproof protective cover and a high temperature-resistant heat insulation structure to meet the requirements of high temperature and high humidity working environment of 60~140℃, ensuring that the equipment will not be damaged or the image acquisition quality will be greatly reduced due to high temperature or dust accumulation during long-term service in the cigarette manufacturing workshop environment.

[0012] Furthermore, the image processing algorithm of the data processing module adopts a color threshold segmentation-based recognition method to extract the target of blue tracer tobacco. The algorithm includes a multi-level verification step: first, the blue area is extracted by color space conversion; then, the target area and contour features are combined to remove minor background noise; finally, the recognition result is verified by the continuous frame changes (such as the consistency of the target's motion trajectory) to eliminate misidentification caused by environmental reflection, water mist refraction, background noise and local occlusion, and finally obtain a high signal-to-noise ratio tracer tobacco recognition time series.

[0013] As mentioned earlier, the time compensation amount can be obtained through calibration of the corresponding conveying distance and conveying speed. The compensation time is calculated in advance by physical measurement and stored in the built-in memory of the data processing module. Compared with the prior art that only performs simple approximate comparison, the compensation amount of the present invention can be calculated using segmented parameters, making the measurement results more accurate and rigorous.

[0014] In some other possible implementation schemes, the data processing module also includes a data management module. The data management module is used to write the material passage time and corresponding detection parameters (such as the effective identification quantity, average inlet and outlet time, compensation amount, etc. mentioned above) obtained from each test into the database, and to use it for subsequent traceability analysis and trend prediction of process data, so as to provide basic data support for intelligent manufacturing based on big data.

[0015] Secondly, based on the above-mentioned device, the present invention also provides a method for measuring the material passage time in a drum based on dyeing tracer and machine vision, mainly including the following steps: First, ordinary tobacco shreds are dyed with food-grade blue dye. After dyeing and appropriate air drying or baking treatment, tracer tobacco shreds are obtained. Then, the optimal single input amount of 5g is weighed and fed into the drum feed end. Second, preferably, an inlet industrial camera set upstream of the drum feed inlet and an outlet industrial camera set at the discharge outlet can be started simultaneously to record the actual import time and outlet detection time sequence of the tracer tobacco shreds. Third, after color threshold recognition, false recognition rejection and data cleaning, the time data of the effective tracer tobacco shreds are collected, and the average values ​​of the import detection time and the outlet detection time are calculated respectively. Fourth, combined with the pre-calibrated compensation parameters, the actual passage time of the material in the drum is calculated, and finally the passage time is output and stored, and the inspection result is synchronously uploaded to the factory MES system for analysis.

[0016] Compared to existing technologies, the main concept of this invention lies in eliminating the complex photoelectric encoding and detection modules, significantly simplifying the equipment structure. While maintaining high-precision material throughput measurement capabilities, it reduces manufacturing costs, shortens equipment modification cycles, and eliminates the need for downtime disassembly and modification of the roller spindle, thus increasing its versatility. Furthermore, through experimentation, the optimal amount of tracer tobacco added per cycle has been determined. This optimal amount ensures a stable supply of the required number of tobacco shreds for the downstream industrial camera without affecting the mixing ratio of production materials, avoiding detection failures or material ratio deviations caused by traditional, rough material feeding methods. In particular, it incorporates protective and heat-insulating features to address the harsh industrial environment of the tobacco processing workshop. The structured industrial camera, combined with an image recognition algorithm based on color threshold segmentation and multi-level verification, effectively resists the interference of strong airflow, steam, and high-humidity water mist on visual recognition, avoiding the misidentification and missed identification that are prone to occur in complex environments by traditional machine vision algorithms. Through systematic physical distance compensation calculation, the running time of materials on the conveyor belt is accurately deducted. Combined with the calculation of the average time of discrete tobacco exit, compared with the existing technology's coarse statistical method, it is more in line with the true physical law that the movement of tobacco inside the drum is discrete and normally distributed. This significantly improves the measurement accuracy of the actual passage time of materials in the drum, providing reliable data support for the consistent control of tobacco quality. Attached Figure Description

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described below with reference to the accompanying drawings, wherein:

[0018] Figure 1 This is an overall structural diagram of the measuring device provided in an embodiment of the present invention;

[0019] Figure 2This is a discrete distribution diagram of tracer tobacco at the drum outlet over time, provided in an embodiment of the present invention.

[0020] Figure 3 This is a schematic diagram of the layout of the measuring device provided in an embodiment of the present invention;

[0021] Figure 4 This is a schematic diagram illustrating the principle of material transit time calculation provided in an embodiment of the present invention.

[0022] Explanation of reference numerals in the attached diagram: 1-Roller equipment; 2-Inlet conveyor belt; 3-Feed inlet; 4-Discharge outlet; 5-Outlet conveyor belt; 6-Outlet industrial camera. Detailed Implementation

[0023] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0024] Based on the actual working environment of this project, the material passing through the rollers for time-point inspection will employ a dyeing method for tracing. A food-grade dye will be added to the surface of the tobacco being tested, and a calibrated industrial camera will track color feature points in the image to trace and calculate the material's movement time. The device and method designed in this invention will be implemented in the roller equipment of a cigarette factory's tobacco processing workshop (taking flavoring and adding rollers as an example, but not limited to this type of roller; this device is also applicable to other types of tobacco processing rollers such as drying rollers and rehydration rollers). During the detection process, the tracer tobacco can completely simulate the passing state of normal tobacco material in terms of physical morphology and motion characteristics. Furthermore, since the dye is food-grade, there is no need for special recycling of the tracer tobacco after the measurement; it can directly enter subsequent production stages with the tobacco, greatly simplifying the detection process.

[0025] like Figure 1 and Figure 3 As shown, the point detection and measurement device of this invention mainly includes a dyeing tracer module (not shown separately in the figure, but composed of food-grade blue dye and tobacco raw materials), a visual inspection module, a data processing module, and a power supply module. These modules work closely together to complete functions such as tracer tobacco preparation, image acquisition, color recognition, time recording, throughput time calculation, and result output, thereby achieving intelligent point detection of material throughput time in roller-type equipment. The inspection device has a compact structure, is easy to install, and can adapt to the online rapid detection needs of continuous production environments in cigarette manufacturing workshops.

[0026] In detail, the dyeing and tracing module uses a food-grade blue dye to dye ordinary tobacco shreds. During implementation, after dyeing, the tracer tobacco shreds are properly air-dried or dried at low temperature to avoid excessive residual moisture from the surface dye affecting their dispersibility and movement within the drum. This ensures that the dyed tracer tobacco shreds are as close as possible to ordinary tobacco shreds in terms of shape, quality, softness, and movement characteristics, thus guaranteeing that the trajectory of the tracer tobacco shreds within the drum is essentially consistent with that of normal tobacco shreds. Regarding the amount of tracer tobacco shreds fed each time, to avoid the problems caused by the coarse feeding method of the feeding hopper in the prior art, this invention, through at least 10 sets of comparative experiments, finally determined that the single feeding amount m=5g is the optimal value. This input amount ensures that the exit camera can reliably identify at least 8 valid tracer tobacco shreds even in harsh high-humidity environments, without interfering with the material distribution and liquid absorption state under normal production conditions inside the drum. If the input amount is too large (e.g., more than 10g), a large number of tobacco shred clumps will form at the tobacco shred exit, thus affecting the dispersion of tobacco shreds on the conveyor belt and causing misidentification by the exit camera. If the input amount is too small (e.g., less than 2g), it will be easily blocked by water mist or covered by other normal tobacco shreds in high-humidity environments, resulting in insufficient number of valid samples identified by the exit camera, which in turn affects the reliability of the statistical mean results.

[0027] In some preferred embodiments, the vision inspection module may include an inlet industrial camera (not shown separately in the figure, located at...) Figure 1 The feed inlet 3 (above the inlet conveyor belt 2) and the outlet industrial camera 6. Combined Figure 3 As shown, the export industrial camera 6 is mounted at the discharge port 4 of the roller device 1 via a camera bracket, with its lens facing the material transport path on the export conveyor belt 5. In practical applications, it can... Figure 1The export industrial camera 6 is raised to a height of 1.5m above the surface of the export conveyor belt 5. This installation height ensures that the camera can clearly capture images of the tracer tobacco across the entire width of the conveyor belt without causing low recognition rates for small targets due to excessive height, or blind spots due to insufficient height. In this embodiment, the selection of the export industrial camera 6 fully considers the harsh environment of the tobacco processing workshop. The camera resolution is set to 1920×1080, the capture frame rate is 30 frames / second, and the housing and lens front end are designed with high-temperature resistant heat insulation structures and equipped with protective covers to prevent dust and moisture from entering. The camera itself has a stable operating temperature range that meets the requirements of high-temperature working environments of 60-140℃. Actual testing has proven that this camera configuration can effectively resist the steam mist and high-temperature radiation generated by the feeding roller and the rehydration roller, ensuring long-term stable acquisition of high-quality image data. To further improve the image recognition rate, the detection area of ​​the export industrial camera 6 is also equipped with an auxiliary white LED light source with constant brightness. The light source is angled at 30 degrees to the optical axis of the camera to illuminate the material surface, thereby enhancing the contrast between the blue tracer tobacco and the background ordinary tobacco and conveyor belt, and improving the accuracy and stability of image recognition.

[0028] Continuing from the previous section, the data processing module, as the core of the entire inspection device, is typically implemented using an embedded industrial computer or a high-performance industrial computer. It communicates with the inlet and outlet industrial cameras via an industrial Ethernet or USB 3.0 high-speed data cable. In some preferred embodiments, the data processing module incorporates a high-precision image processing algorithm and a time calculation program. This image processing algorithm is not a simple color histogram matching but is specifically optimized for the unique interferences encountered during material feeding and rehydration environments. Specifically, the data processing module first converts the RGB color image captured by the camera to the HSV color space. It then uses a two-layer filtering condition, setting a blue feature threshold based on the H (hue) and S (saturation) components, to lock down suspected target areas. Next, morphological opening and closing operations are used to remove isolated bright spots and noise interference smaller than a preset threshold (e.g., less than 50 pixels) caused by fog refraction and ambient light reflection. Finally, the algorithm verifies the continuity of target motion, extracting tobacco shreds that have been continuously tracked for more than three frames, identifying only these as genuine, valid tracer tobacco shreds. Then, the system uses a data cleaning mechanism to remove abnormal time points and duplicate counts that are lost midway due to obstruction, extracts and records the accurate timestamp of each valid tobacco shred, and finally generates a purified detection time series.

[0029] Combined Figure 4 As shown, the measurement method of the present invention includes the following key steps in its specific implementation:

[0030] Step S1: Use food-grade blue dye to dye ordinary tobacco shreds, and after preparation, measure the dyed tracer tobacco shreds; according to the predetermined optimal single input amount, in this embodiment, 5g of blue tracer tobacco shreds are accurately weighed for each test as a sample for measuring the material transit time of a single test.

[0031] Step S2: 5g of blue tracer tobacco is evenly spread onto the inlet conveyor belt 2 through the feed inlet 3. At this time, the inlet industrial camera and the outlet industrial camera 6 can be started synchronously and continuously capture images. Since the belt speed v1 of the inlet conveyor belt 2 can be set, and the physical distance l1 between the inlet detection and discharge position and the roller feed inlet 3 can be set as a known quantity, the first time compensation amount T1 from the camera detection point to the roller can be pre-calculated according to the formula T1=l1 / v1. Similarly, the distance l2 between the outlet 4 of the roller device 1 and the center of the field of view of the outlet industrial camera 6 is also fixed, and the speed v2 of the outlet conveyor belt 5 can also be considered as a known quantity. Therefore, the second time compensation amount T2=l2 / v2 from the outlet to the detection point of the outlet industrial camera 6 can be pre-calculated. Preferably, both sets of time compensation amounts can be stored in the storage medium of the data processing module.

[0032] Step S3: After undergoing a complex process of tumbling, scattering, and axial conveying inside the drum device 1, the tracer tobacco is discharged from the outlet 4 onto the outlet conveyor belt 5. Because the movement of the tobacco inside the drum is highly random and discrete, not all the tracer tobacco will arrive at the outlet conveyor belt 5 at the same time; instead, they will appear as follows: Figure 2 The distribution shown indicates that the detection time of the tracer tobacco at the outlet exhibits a normal distribution (e.g., Figure 2 The horizontal axis represents time t, and the vertical axis represents the number of detected tobacco strands m. Based on this specific phenomenon, the export industrial camera 6 of the data processing module will continuously capture blue tracer tobacco strands in the image within a certain time period. After the image recognition algorithm extracts and cleans the data, the export timestamp of each valid tracer tobacco strand being detected will be recorded in chronological order, denoted as t3i, where i=1,2,…,n, and n is the total number of valid tracer tobacco strands successfully identified by the export camera in this experiment.

[0033] Step S4: After completing image acquisition, the data processing terminal first performs statistical analysis on the time data of all valid tracer tobacco shreds. In some embodiments, the data terminal acquires time data of 12 valid tracer tobacco shreds. After data cleaning, the average exit detection time is calculated. For example, if the starting time of feeding in front of the drum is t0=10:00:00 and T1=2s, then the actual inlet time t1=t0+T1=10:00:02; and the average timestamp of the valid tobacco shreds identified by the exit camera is, for example, =10:05:30 (totaling 330s relative to the starting time), subtracting the compensation time T2=5s, the actual time at the roller exit is 10:05:25. Substituting the above parameters into the calculation formula, the actual throughput time is obtained. T = (330s − 5s) − (0s + 2s) = 325s − 2s = 323s. That is, the actual average material transit time inside the drum in this embodiment is 323 seconds. Compared with existing methods, this 323s eliminates the material's transport time on the conveyor belt and, in particular, takes into account the randomness and discreteness introduced by the drum equipment, thus sufficiently reflecting the movement / residence time of the tobacco shreds inside the drum.

[0034] It should be added that the above process is combined Figure 4 It can also include the following derivation: where t1 can be obtained by adding T1 to the known time t0, and T1 is the time required for the material to enter the drum from the conveyor belt; Where m is the amount of tobacco shreds; T2 is the time required for the material to travel from the drum to the outlet camera. Figure 4 middle This refers to the average time of all marked tobacco at the drum outlet position: = -T2. That is, in the conception of this invention, the derivation process of the formula mentioned above is... t1= .

[0035] Step S5: Finally, the data processing terminal outputs the material throughput time and stores this data in the local database. After this inspection is completed, the exit industrial camera 6 can be turned off, while retaining the storage and data upload functions of the data processing module for subsequent process analysis. Preferably, the data processing terminal also synchronously uploads the inspection time, tracer tobacco identification quantity, mean calculation result, standard deviation, and final throughput time of this inspection to the workshop MES management system or process quality analysis platform to form historical data records for subsequent quality trend analysis, equipment status assessment, and process optimization. In addition, since this device has a data recording function, when multiple consecutive inspection results show abnormal fluctuations (for example, a test result exceeding ±10% of the average of the last 10 tests), the system will automatically determine the current roller has abnormal material discharge or internal material accumulation risk based on historical data in the database, and then provide early warning prompts for the roller's operating status, assisting operators in the tobacco processing workshop to promptly detect abnormal conditions and adjust process parameters such as roller tilt angle or rotation speed.

[0036] In summary, compared with the prior art, the present invention has the following core improvements:

[0037] First, in terms of measurement function and device structure, the inspection device of the present invention eliminates the coding tape module and photoelectric detection module for measuring the drum speed, thereby eliminating the modification process of wrapping the coding tape around the drum, reducing the hardware complexity and installation cost of the equipment, shortening the downtime of equipment modification, and avoiding the influence of high humidity dust interference of the photoelectric coding module on the synchronous speed measurement of the drum in the feeding and humid environment, so that the present invention can focus on the precise measurement and evaluation of material throughput time;

[0038] Secondly, in terms of the refined control of the feeding method, the amount of material fed in the past was determined by the volume of the feeding bin, and the optimal amount was not verified for the recognition capability of the camera under different working conditions. However, the present invention clearly proposes the optimal single feeding amount through multiple sets of comparative experiments, such as 5g mentioned in the previous example. This optimal feeding amount based on experimental calibration can reduce the risk of fluctuation in the tobacco processing parameters caused by abnormal material mixing while ensuring that the exit camera can stably recognize and trace the tobacco.

[0039] Third, in response to the high temperature and humidity steam environment in the silk-making workshop, this invention explicitly requires that the exported industrial camera be an industrial-grade high-definition camera, such as one with a temperature resistance range of 60~140℃, and also requires an external dustproof and anti-fog protective cover and an auxiliary light source, so as to ensure that the camera can maintain high-definition image acquisition for a long time in the steam-filled and dust-filled silk-making production line.

[0040] Fourth, regarding the complexity of image processing algorithms and data cleaning, this invention specifically introduces a threshold segmentation algorithm based on color space conversion (HSV color space) into the data processing module. It also combines target area, contour features, and continuous frame changes to perform multi-level verification and anomaly cleaning of the recognition results, effectively eliminating erroneous timestamps caused by ambient light reflection or partial occlusion. This multi-level image processing and data quality review mechanism, compared to existing coarse image processing methods, can significantly improve the purity and accuracy of effectively traced tobacco timestamp extraction under low visibility conditions.

[0041] Fifth, regarding the accuracy of time compensation, this invention precisely breaks down time compensation into the initial compensation from the material inlet monitoring point to the drum inlet and the subsequent compensation from the material outlet to the outlet camera detection point. These two independent physical calibration values ​​are then substituted as independent parameters into the transit time algorithm model, instead of roughly estimating by dividing the sum of the distances between the two monitoring points by the constant belt speed. This segmented calibration and compensation method of this invention can further eliminate the synchronization delay error of the mechanical movements of the material entering and leaving the drum, making the measurement results closer to the actual physical transit time of the material.

[0042] Sixth, this invention uses tracer tobacco as the detection carrier, which can realistically simulate the movement of normal tobacco in the flavoring and feeding drum. Compared with traditional methods, it has the advantages of intuitive detection process, accurate results, timely response, and good repeatability. Furthermore, this invention uses food-grade dye for labeling, eliminating the need for special recovery of the tracer tobacco after detection, making it convenient to operate and suitable for widespread application in cigarette manufacturing production lines. In particular, the camera installation position, capture frame rate, dye type, and image recognition parameters in this invention can be adaptively adjusted according to actual needs for different specifications of flavoring and feeding drums, different conveyor belt speeds, and different ambient lighting conditions.

[0043] In this invention, when directional terms are mentioned, they are relative concepts based on the embodiments. Furthermore, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects have an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0044] The above description of the structure, features, and effects of the present invention is based on the embodiments shown in the figures. However, the above are only preferred embodiments of the present invention. It should be noted that the technical features involved in the above embodiments and their preferred methods can be reasonably combined and matched by those skilled in the art to form a variety of equivalent solutions without departing from or changing the design concept and technical effects of the present invention. Therefore, the present invention is not limited to the scope of implementation shown in the figures. Any changes made in accordance with the concept of the present invention, or modifications to equivalent embodiments, that do not exceed the spirit covered by the specification and figures, should be within the protection scope of the present invention.

Claims

1. A material passage time measuring device for roller-type equipment based on dyeing tracing and machine vision, characterized in that, include: The dyeing and tracing module is used to attach food-grade dye to the surface of ordinary tobacco to obtain tracer tobacco, which is consistent with the original ordinary tobacco in terms of physical morphology and motion characteristics. The vision inspection module includes at least an exit industrial camera located downstream of the discharge port of the drum equipment. The exit industrial camera is used to record the exit detection time t of the tracer tobacco on the conveyor belt after it leaves the drum. 3i ; The data processing module is communicatively connected to the visual detection module. It is used to receive and process images captured by the camera, extract color feature points of the tracer tobacco to generate corresponding recognition time series, and calculate the average value of the exit detection time after retaining the valid data according to the data cleaning rules. The data processing module pre-stores the first time compensation amount T1 of the material from the feeding detection point to the drum inlet and the second time compensation amount T2 of the material from the drum outlet to the outlet camera detection point, and calculates the material's transit time T inside the drum according to the following formula: Where t0 is the time when the tobacco is placed on the conveyor belt. To effectively trace the average exit detection time of the tobacco at the exit camera.

2. The material passage time measuring device for roller-type equipment based on dyeing tracing and machine vision according to claim 1, characterized in that, The first time compensation amount and the second time compensation amount are obtained by calibration using the actually measured conveying distance and the corresponding conveyor belt running speed: T1 = l1 / v1, T2 = l2 / v2, where l1 is the distance from the tobacco placement position to the roller inlet, l2 is the distance from the roller outlet to the industrial camera detection position at the outlet, and v1 and v2 are the running speeds of the inlet and outlet conveyor belts, respectively.

3. The material passage time measuring device for roller-type equipment based on dyeing tracing and machine vision according to claim 1, characterized in that, The data processing module has built-in image processing algorithms and time calculation programs; The image processing algorithm employs a color threshold segmentation-based recognition method, which combines target area, contour features, and continuous frame changes to verify the recognition results. This is used to eliminate misidentifications caused by environmental reflections, background noise, and local occlusion, and to generate a purified tracer tobacco outlet detection time series.

4. The material passage time measuring device for roller-type equipment based on dyeing tracing and machine vision according to claim 1, characterized in that, The amount of tracer tobacco added in each experiment is obtained through experimental calibration and is the minimum amount that ensures the export industrial camera can clearly identify it without interfering with the normal material mixing state.

5. The material passage time measuring device for roller-type equipment based on dyeing tracing and machine vision according to any one of claims 1 to 4, characterized in that, The exported industrial camera is an industrial-grade high-definition camera with an installation height of 1.5m, a resolution of 1920×1080, and a capture frame rate of 30 frames per second. The industrial camera is equipped with a dustproof protective cover and a heat insulation structure to meet the requirements of working environments with high temperature, high humidity, and dust in the range of 60~140℃.

6. A method for measuring the material throughput time of roller-type equipment based on dyeing tracing and machine vision, employing the measuring device as described in any one of claims 1 to 5, characterized in that, include: Step 1: Use food-grade blue dye to dye ordinary tobacco shreds to prepare tracer tobacco shreds, and measure the tracer tobacco shreds according to the calibrated single input amount; Step 2: The measured tracer tobacco is fed into the roller from the feed end, and the time of material feeding is recorded, and the industrial camera exits simultaneously. Step 3: Record the inlet time before the effective tracer tobacco enters the drum and the outlet detection time after it leaves the drum, and calculate the average inlet time and the average outlet detection time based on the number of identified tobacco. Step 4: Using the first time compensation amount from the material entering the detection point to the drum inlet, and the second time compensation amount from the drum outlet to the exit camera detection point, combined with the average inlet time and the average outlet detection time, calculate the material's transit time inside the drum.

7. The method for measuring material throughput time in roller-type equipment based on dyeing tracing and machine vision according to claim 6, characterized in that, In step three, the effective tracer tobacco is set to meet the screening conditions of clear color features, complete outline, reasonable changes in continuous frames, and no overlapping misjudgment. Furthermore, if the number of valid identifications in a single test is lower than a preset threshold, it is determined to be an abnormal test and automatically triggers a refeeding process.

Images