Tobacco flake loosening and moisture regaining effect detection and analysis method and system based on machine vision detection
Through machine vision detection technology, the agglomeration and condensation profile of cigarette pieces are identified, the looseness rate is calculated, and the knowledge graph is constructed, which solves the problem of time-consuming and low reliability of the loose reflux evaluation of cigarette pieces, and achieves efficient and accurate cigarette pieces The detection and analysis of loose tide resurrection effects improves production efficiency and quality.
Patent Information
- Application Number
- CN202411310013.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-19
- Publication Date
- 2024-11-15
AI Technical Summary
In the prior art, the evaluation of the loose moisture resurrection effect of tobacco sheets takes a long time, and the efficiency of manual selection of agglomerated cigarette sheets is low, resulting in low evaluation reliability and the inability to comprehensively analyze factors such as moisture content and temperature, which affects the normal operation of the production process. .
Using machine vision detection and analysis method for detecting and analyzing smoke sheet loose reflux effect, by setting the sampling frequency, conveying rate and the photo frequency of industrial vision cameras, identifying the agglomeration and condensation profile in the image, calculating the looseness rate, and constructing a knowledge graph To analyze the status and causes of cigarettes, improve the reliability and efficiency of the evaluation.
It realizes fast and accurate detection of loose and moisture reflux effect of smoke sheets, avoids the time-consuming and missed selection problems of manual selection, improves the reliability and production efficiency of evaluation, and can promptly discover and adjust process parameters to improve production quality.
Smart Images

Figure CN118961702A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of tobacco loosening and rehydration, and in particular to a method and system for detecting and analyzing the loosening and rehydration effect of tobacco sheets based on machine vision detection. Background Art
[0002] The purpose of loosening and moistening tobacco is to increase the moisture content and temperature of tobacco sheets, improve the processing resistance of tobacco sheets, loosen the tobacco sheets after cutting, and improve the sensory quality of tobacco sheets. The indicators of the loosening and moistening effect of tobacco sheets mainly include the moisture content, temperature and looseness of tobacco sheets after loosening and moistening.
[0003] At present, the evaluation of agglomeration rate mainly relies on people picking out agglomerated tobacco sheets from the loosening and rehydration discharge port, and then weighing and calculating the loosening rate. This method of manually picking out agglomerated tobacco sheets requires spreading out the discharge after loosening and rehydration before picking out agglomerated tobacco sheets, which is not only time-consuming, but also due to the requirements of the loosening and rehydration process, after the loosening and rehydration is completed, there is a large amount of loosening and rehydration discharge at the discharge port within the specified time. The manual screening of agglomerated tobacco sheets lags behind the loosening and rehydration discharge, which directly affects the normal operation of the discharge and other subsequent processes; and, since the rate of manual selection of agglomerated tobacco sheets is much lower than the rate of loosening and rehydration discharge, agglomerated tobacco sheets are often missed or under-selected, and there is a problem of reduced reliability in the overall loosening effect evaluation; on the other hand, manual weighing and calculating the loosening rate can only obtain a result of the loosening rate, and cannot combine condensation, moisture content and temperature to comprehensively evaluate the loosening and rehydration effect, nor can it analyze the reasons for the poor loosening and rehydration effect, and there is a problem of low reliability in the detection and analysis of the loosening effect. Summary of the invention
[0004] The purpose of the present invention is to provide a method and system for detecting and analyzing the loosening and moisture regaining effect of tobacco sheets based on machine vision detection in order to address the problems in the prior art that the selection of agglomerated tobacco sheets takes a long time, affecting the subsequent production process of the tobacco sheets, and the reliability of the loosening effect evaluation is low.
[0005] In order to achieve the above object, the technical solution adopted by the present invention is:
[0006] A method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection comprises the following steps:
[0007] Parameter setting: setting the sampling frequency, the conveying rate of the conveying mechanism and the imaging frequency of the industrial vision camera; matching the sampling frequency, the conveying rate and the imaging frequency;
[0008] Sampling and transmission: according to the sampling frequency, the test material is obtained from the loose moisture return outlet through the sampling plate, and the test material is spread on the conveying mechanism; according to the conveying rate, the test material is conveyed through the industrial vision camera, and then the image of the test material is obtained through the industrial vision camera; the test material passing through the industrial vision camera is synchronously conveyed back to the loose moisture return outlet;
[0009] Detection and analysis: transmitting the image to the detection and analysis platform via an electrical signal; identifying the agglomeration contour of the image, and calculating the first area S1 of the agglomeration contour; calculating the total area S of the detected material according to the conveying rate V, the imaging frequency f and the conveying mechanism width L; calculating the looseness rate of the detected material according to the first area and the total area;
[0010] The calculation formula of the loose rate is as follows:
[0011]
[0012] The technical solution of the present invention sets the sampling frequency so that the sampling plate can take samples at a fixed frequency from the loose and rehumidified material outlet according to the sampling frequency, and then cooperates with the conveying rate set by the conveying mechanism and the photographing frequency of the industrial camera to enable sampling and detection to be carried out in an orderly manner. It can not only avoid the time-consuming manual selection of agglomerated tobacco leaves or the detection of materials, but also improve the detection efficiency and production benefits; and according to historical experience values or loose rate evaluation results, when the set sampling frequency, conveying rate and photographing frequency settings match, the loose and rehumidified tobacco leaves can be sampled evenly as evenly as possible, which can avoid the omission or under-selection of agglomerated tobacco leaves in the manual selection of agglomerated tobacco leaves, and improve the reliability of the loose effect evaluation as a whole.
[0013] A further improvement of the present invention is that the method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection further includes the following steps:
[0014] Identify the condensation contour in the image, calculate the second area of the condensation contour; and analyze the condensation state of the cigarette sheet based on the second area.
[0015] The further improved technical solution of the present invention can calculate the second area of the condensation contour by identifying the condensation contour in the image, and can analyze the condensation state of the tobacco sheet by referring to the condensation size based on historical experience. Compared with the traditional loosening effect analysis, the analysis items of the condensation state can be increased, and the further analysis of the loosening and rehumidification effect of the tobacco sheet can be deepened, thereby improving the analysis performance and further promoting the quality improvement of the loosening and rehumidification process.
[0016] A further improvement of the present invention is that the method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection further comprises the following steps:
[0017] Obtain historical data on loosening and regaining effects and store loosening rate data; mark the first and second moments when agglomerated tobacco sheets and condensed tobacco sheets are obtained, respectively, and build and display a loosening and regaining knowledge graph in conjunction with online temperature detection results and online moisture detection results;
[0018] The knowledge graph includes: first-level status items and second-level cause items arranged in a mutual hierarchy; the first-level status items include: agglomerated tobacco leaves, condensation, low temperature, high temperature, low moisture content, and high moisture content; the second-level cause items are the causes of occurrence of each of the first-level status items; the causes of occurrence are historical cause conclusions recorded by operators.
[0019] A further improved technical solution of the present invention, on the basis of being able to obtain information on agglomeration and condensation of tobacco sheets, can determine the presence of agglomerated tobacco sheets and condensed tobacco sheets at the first moment and the second moment respectively by marking the first moment of agglomerated tobacco sheets and condensed tobacco sheets, thereby displaying the primary status items of agglomerated tobacco leaves and condensation to users in the knowledge graph; in combination with the online temperature detection results and the online moisture detection results, it can be used to display status items such as high temperature, low temperature, high moisture content and low moisture content, and the user can obtain the status of agglomerated tobacco sheets, condensation, temperature and moisture content from the display after the loosening and moisture recovery is completed; and by obtaining historical data on loosening and moisture recovery, a loosening and moisture recovery process knowledge graph is constructed according to historical data and historical cause conclusions recorded by operators, so that the user can visualize the cause reference of each primary status item after detecting the loosening and moisture recovery effect of the tobacco sheets, can promptly discover problems in the loosening and moisture recovery, and then can promptly adjust the parameters or settings of the loosening and moisture recovery, thereby further improving the production quality of the loosening and moisture recovery.
[0020] A further improvement of the present invention is that the method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection constructs the knowledge graph, and specifically also includes the following steps: establishing a solution corresponding to each of the secondary cause items based on the historical data and the historical solutions recorded by the operator, and creating a link in the secondary cause item so that when the user clicks on the secondary cause item, the corresponding solution can be opened.
[0021] A further improvement scheme of the present invention establishes a solution corresponding to each of the secondary cause items and creates a connection, which can visually display the secondary cause items for users. While analyzing the causes, the solution can correspond to the secondary cause items through historical data and historical solutions, and quickly obtain solutions to defects in loose moisture regain, which is convenient for operators and maintenance personnel to quickly make adjustments and feedback, so as to reduce loose moisture regain defects and further improve the production quality and production efficiency of loose moisture regain.
[0022] A further improvement of the present invention is that the method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection identifies the agglomeration contour of the image, specifically comprising the following steps:
[0023] Read the image imread through the Python program, convert the image imread into a grayscale image cvtColor; apply Gaussian blur to remove noise GaussianBlur; set a threshold, process the grayscale image according to the threshold, so that the grayscale image highlights the dark area threshold; open operation to remove small objects, close operation to connect adjacent objects, and then find and select the agglomeration contour;
[0024] The further improved technical solution of the present invention, based on the ability to obtain the agglomeration contour by traditional means, specifically acquires the image through a Python program, and then converts the image into a grayscale image. According to image processing steps such as thresholding and processing the grayscale image, the accuracy of image processing can be improved, and the identified agglomeration contour can be made more accurate, and the accuracy of the first area calculation of the agglomeration contour can be further improved. Compared with traditional manual selection or identification of agglomerated tobacco sheets, the reliability of the loosening effect evaluation can be further improved.
[0025] A further improvement of the present invention is that the method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection calculates the first area of the agglomeration contour, specifically comprising the following steps:
[0026] Draw the edge of the agglomerate outline and calculate the first area of the agglomerate outline; if the first area is ≥ 3 cm 2 , it is determined that the tobacco sheet corresponding to the image is agglomerated; if the first area is less than 3cm 2 ; It is determined that the tobacco sheet corresponding to the image is not agglomerated.
[0027] A further improvement scheme of the present invention, on the basis of being able to calculate the loose rate of tobacco sheets, narrows the scope of identifying and calculating tobacco sheet agglomerations according to the specific first area range of agglomerated tobacco sheets, can make qualitative judgments on agglomerated tobacco sheets in combination with empirical values, reduce the screening and judgment of data that does not conform to the agglomerated tobacco sheet, improves the efficiency of detection and data analysis, reduces energy consumption and analysis time, and further improves detection efficiency and production benefits.
[0028] A further improvement of the present invention is that the method for detecting and analyzing the loosening and rehydration effect of tobacco sheets based on machine vision detection, sampling specifically further includes the following steps: firstly, taking material from below the loosening and rehydration discharge port, and then scraping the sampled detection material to a transmission mechanism;
[0029] The transmission mechanism transmits the inspection material in sections and continuously, specifically including the following steps: first, the inspection material is transmitted upward at an angle, and then the inspection material is transmitted horizontally; during the horizontal transmission process, the conveying direction of the inspection material is turned 180°; the inspection material after the turn passes through the industrial vision camera, and finally the inspection material is transmitted back to the subsequent process below the loose rehumidification discharge port.
[0030] The further improved technical solution of the present invention, on the basis of setting the sampling frequency to match and thus avoiding the detection material from being stacked again on the conveying mechanism, makes the sampling and transmission run more orderly, and in the process of conveying the detection material, by first transmitting the detection material upward at an angle and then transmitting the detection material horizontally, the detection material can be fully spread out in the horizontal stage under the action of its own gravity, thereby avoiding the stacking of cigarette pieces again, further improving the authenticity of the image acquired by the industrial vision camera, further avoiding the stacking of cigarette pieces, improving the detection accuracy, and further improving the detection reliability.
[0031] A tobacco loosening and rehumidification effect detection and analysis system comprises: a rotary sampling mechanism, a conveying mechanism, an industrial visual camera and a detection and analysis platform; the rotary sampling mechanism can rotate horizontally to obtain detection materials from a loosening and rehumidification discharge port, and can place the detection materials on the conveying mechanism by rotation; the conveying mechanism is used to spread and convey the detection materials, so that the detection materials pass through the industrial camera, and then the image of the detection materials can be obtained; the conveying mechanism is also used to return the detection materials that have passed the industrial visual camera to the discharge vibration trough.
[0032] The detection and analysis system described in the present invention can realize the detection and analysis method described in the present invention. Through the mutual cooperation of the rotating sampling mechanism, the conveying mechanism and the industrial vision camera, the operation of the detection and analysis system will not affect the normal operation of loose regain, so that the detection process can be more orderly, the efficiency of detection and analysis is higher, and the loose regain can be coordinated to shorten the time of detection and analysis, reduce the adverse effects on production, and thus improve production efficiency; and through the detection and analysis platform, the loose regain effect analysis of the detection material can be performed intelligently, making the loose regain effect analysis more convenient and efficient.
[0033] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:
[0034] 1. The detection and analysis method avoids the defects of traditional manual selection of agglomerated tobacco sheets or time-consuming material detection, thereby improving detection efficiency and production benefits;
[0035] 2. The detection and analysis method, based on historical experience values or loosening rate assessment results, can avoid the omission or under-selection of agglomerated tobacco sheets in manual selection, and improve the reliability of loosening effect assessment as a whole, when the sampling frequency, conveying rate and photographing frequency settings are matched.
[0036] 3. The detection and analysis system can realize the detection and analysis method of the present invention. Through the cooperation of the rotating sampling mechanism, the conveying mechanism and the industrial vision camera, the operation of the detection and analysis system will not affect the normal operation of the loose regain, so that the detection process can be more orderly, the detection and analysis efficiency is higher, and the loose regain can be coordinated to shorten the detection and analysis time, reduce the adverse effects on production, and thus improve production efficiency; and the detection and analysis platform can intelligently analyze the loose regain effect of the detection material, making the loose regain effect analysis more convenient and efficient. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 It is a schematic diagram of the process of the detection and analysis method of the present invention;
[0038] Figure 2 It is a schematic diagram showing the knowledge graph of the present invention;
[0039] Figure 3 It is a schematic diagram of the module structure of the detection and analysis system of the present invention. DETAILED DESCRIPTION
[0040] The present invention will be described in detail below in conjunction with the accompanying drawings.
[0041] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0042] Example 1
[0043] A method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection, referring to Figure 1 As shown, the specific steps include:
[0044] S1: Set the sampling frequency, the conveying rate of the conveying mechanism and the imaging frequency of the industrial vision camera; make the sampling frequency, the conveying rate and the imaging frequency match;
[0045] S2: According to the sampling frequency, the test material is obtained from the loose moisture return outlet through the sampling plate, and the test material is spread on the conveying mechanism;
[0046] S3: conveying the inspection material through the industrial vision camera according to the conveying rate;
[0047] S4: Acquire an image of the inspected material through an industrial vision camera; synchronously convey the inspected material that has passed through the industrial vision camera back to the loose moisture regain outlet;
[0048] S5: transmitting the image to the detection and analysis platform via an electrical signal; identifying the agglomeration contour of the image, and calculating the first area S1 of the agglomeration contour;
[0049] S6: Calculate the total area S of the detected material according to the conveying rate V, the imaging frequency f and the conveying mechanism width L; calculate the looseness rate of the detected material according to the first area and the total area;
[0050] The calculation formula of loose rate is as follows:
[0051]
[0052] It should be noted that the sampling frequency is understood as the number of times samples are taken from the discharge port through the sampling plate per unit time. This can be achieved by controlling the rotation speed of the sampling plate so that the rotation speed of the sampling plate per unit time, the conveying rate of the conveying mechanism and the imaging frequency match each other; "matching" is understood as adjusting the conveying rate of the conveying mechanism based on experience so that the sampling frequency is not too fast and the conveying rate is low, so as to avoid stacking of the test material when it is scraped onto the conveying mechanism, and to avoid the sampling frequency being too slow or unbalanced, resulting in the omission of agglomerated tobacco flakes in the test material or less sampling; synchronously adjust the conveying rate of the conveying mechanism to match the imaging frequency based on empirical values, so as to avoid overlapping parts of the two images of the test material obtained, so as to avoid errors in the calculation of the loose rate; it should also be noted that the imaging frequency can be understood as the frequency of acquiring images. For example, an industrial vision camera acquires images by taking pictures, and the imaging frequency is understood as the frequency of taking pictures.
[0053] In this embodiment, in order to control the sampling frequency and to make the sampling, transmission and detection more coordinated, specifically, the sampling includes the following steps: first, the material is taken from below the loose rehumidification outlet, and then the sampled detection material is scraped to the transmission mechanism; the transmission mechanism transmits the detection material in sections and continuously, specifically including the following steps: first, the detection material is transmitted upward at an angle, and then the detection material is transmitted horizontally; during the horizontal transmission process, the conveying direction of the detection material is turned 180°; the turned detection material passes through an industrial vision camera, and finally the detection material is transmitted back to the subsequent process below the loose rehumidification outlet for transportation, so that the detection of the loose rehumidification effect will not affect the normal operation of the next loose rehumidification process.
[0054] In this embodiment, the agglomeration contour of the identified image is understood as the color of the agglomeration contour is darker than that of the non-agglomerated portion because the tobacco sheets overlap and stick together in the agglomerated portion of the image. Therefore, the parameter value of the agglomeration contour can be obtained through the image processing program, and the first area can be calculated based on the parameter value. Specifically, the agglomeration contour of the identified image includes the following steps: reading the image imread through the Python program, converting the image imread into a grayscale image cvtColor; applying Gaussian blur to remove noise GaussianBlur; setting a threshold, processing the grayscale image according to the threshold value, so that the grayscale image highlights the dark area threshold; using an open operation to remove small objects, and using a closed operation to connect adjacent objects, and then finding and selecting the agglomeration contour, thereby obtaining the agglomeration contour.
[0055] Specifically, the detection and analysis of the present invention also includes the following steps: identifying the condensation contour in the image, calculating the second area of the condensation contour; and analyzing the condensation state of the cigarette according to the second area. The implementation method of the condensation contour identification method and the second area calculation method refers to the implementation method of the agglomeration contour and the first area. More specifically, the identification of agglomeration, condensation and other cigarette pieces is implemented by Python program, and the steps include: reading the image imread, converting it into a grayscale image cvtColor, applying Gaussian blur to remove noise GaussianBlur, using threshold processing to highlight the dark part indicating agglomeration and the bright part indicating condensation area threshold, opening operation to remove small objects, closing operation to connect adjacent objects, finding contours, filtering out larger contours, drawing contours, and calculating and displaying results.
[0056] According to the empirical value of the size of the agglomerated tobacco sheets after loosening and rehydration, the agglomeration contour can be determined by reference and comparison. In this embodiment, specifically, the first area of the agglomeration contour is calculated, which specifically includes the following steps: drawing the edge line of the agglomeration contour, calculating the first area of the agglomeration contour; if the first area is ≥3cm 2 , it is determined that the tobacco leaf corresponding to the image has agglomerated tobacco leaves; if the first area is less than 3cm 2. It is determined that the cigarette sheet corresponding to the image is not agglomerated.
[0057] It should be noted that the implementation method of calculating the total area S of the test material is that, through the method of the present invention, the test material can be spread on the conveying mechanism, for example, on the top of the conveyor belt, and the product of the conveying rate V and the width L of the conveying mechanism is equal to the photographic area of the test material within the shooting range of the industrial vision camera per unit time; and then the photographic area is multiplied by the imaging frequency, that is, it is understood that the area of each image is multiplied by the number of shots, then the total area of the test material photographed per unit time can be obtained; since the first area S1 of the agglomeration contour is also obtained from the image, the time for obtaining the image of the agglomeration contour is equal to the time for obtaining all the images of the test material, which does not affect the size of the loose rate value in the loose rate formula, and therefore the percentage of the loose regain rate of the tobacco sheet can be calculated based on the area of the tobacco sheet after loose regain and the agglomeration area.
[0058] In this embodiment, specifically, the detection and analysis also includes the following steps: obtaining historical data on loosening and regaining effects, and storing loosening rate data; marking the first moment and the second moment when agglomerated tobacco sheets and condensed tobacco sheets are obtained respectively, and in conjunction with online temperature detection results and online moisture detection results, constructing and displaying a loosening and regaining knowledge graph.
[0059] It should be noted that the knowledge graph is constructed by operators inputting historical data and historical documents, and then by organizing historical documents and executing hyperlink operations on files, the data obtained by the detection and analysis platform can be associated with the existing historical data, thereby realizing the construction of the knowledge graph; specifically, through the operating system of the detection and analysis platform, such as the PC side, it is possible to read historical data and historical documents, store and control the association of execution data, and then visualize it as a knowledge graph; refer to Figure 2 As shown, the knowledge graph includes: first-level status items and second-level cause items arranged in a mutual hierarchy; the first-level status items include: agglomerated tobacco leaves, condensation, low temperature, high temperature, low moisture content, and high moisture content; the second-level cause item is the cause of occurrence of each first-level status item; the cause of occurrence is the historical cause conclusion recorded by the operator.
[0060] It should be noted that the "first level" and "second level" are understood to mean that the sub-items of the status item are the cause items, and the second level is a classification and sorting item relative to the first level, and no specific number is explained; the specific second level cause items are understood as follows: Figure 2 The cause analysis of the semantics of the text shown, for example, one of the secondary causes corresponding to the agglomerated tobacco leaves is that the slicer slices the tobacco leaves too thickly. It is understood that during the loose rehumidification, according to the structure of the incoming tobacco leaves, they are generally cut into three knives and four slices to four knives and five slices. The extreme difference in the thickness of the tobacco blocks after cutting is no more than 15mm. If the extreme difference is greater than 15mm, then according to the standard requirement value, the cause is analyzed to be agglomeration caused by excessive slicing.
[0061] In this embodiment, refer to Figure 2 The constructed knowledge graph example also includes the following steps in the construction process: establishing a corresponding solution for each secondary cause item based on historical data and historical solutions recorded by operators, and creating a link in the secondary cause item so that the user can open the corresponding solution when clicking on the secondary cause item.
[0062] Specifically, for example, the existing loosening and regaining requirements are that the moisture content of the export material is 17.0%±1.5, and the temperature of the export material is 60.0℃±3.0. The traditional process design standards do not make any requirements for the loosening rate. Through the present invention, based on the evaluation of the loosening and regaining effect and the quantitative calculation of the loosening and regaining rate, combined with experience and product quality analysis of the loosening and regaining, it is possible to establish a loosening rate requirement index for the loosening and regaining, for example, a requirement of loosening rate ≥90%, so as to improve the quality of the loosening and regaining of tobacco leaves; correspondingly, based on the online temperature detection and online moisture detection on the production line, the quality of the tobacco leaves after loosening and moisturizing is judged. If the online detection parameters do not meet the standards, the knowledge graph will be used to find specific problem points and solutions, so that the problems in the loosening and regaining can be solved as soon as possible, thereby improving production efficiency.
[0063] Example 2
[0064] Used to implement the side analysis method in Example 1, this embodiment provides a tobacco loosening and rehumidification effect detection and analysis system, including: a rotating sampling mechanism, a conveying mechanism, an industrial vision camera and a detection and analysis platform; the rotating sampling mechanism can rotate horizontally, thereby obtaining the detection material from the loosening and rehumidification discharge port, and can place the detection material on the conveying mechanism by rotation; the conveying mechanism is used to spread and convey the detection material, so that the detection material passes through the industrial camera, and then the image of the detection material can be obtained; the conveying mechanism is also used to return the detection material that has passed the industrial vision camera to the discharge vibration trough.
[0065] Specifically, the implementation method of the detection and analysis system is as follows: the rotating sampling mechanism is understood to be, for example, a turntable that can be driven to rotate horizontally by a reduction motor, and can receive the loose and rehumidified tobacco sheet material discharged from the loose and rehumidification discharge port, and through horizontal rotation, rotate it out of the loose and rehumidification discharge port, and transfer it to the top of the conveying mechanism, and through a mechanical arm scraper or a blocking plate arranged on the upper surface of the horizontal rotating disk, the taken detection material can be scraped to the conveying mechanism during the rotation of the horizontal rotating disk, and then the detection material is transported through the industrial vision camera by the conveying mechanism. The industrial vision camera does not affect or stop the continued transportation of the detection material during the process of acquiring the image of the detection material, and can thus transport the detection material back to the discharge vibration trough in this process; it should be noted that the discharge vibration trough is understood to be a structure that is arranged below the loose and rehumidification discharge port and can transport the loose and rehumidified material to the subsequent process.
[0066] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection, It is characterized in that The following steps are involved: Parameter setting: setting the sampling frequency, the conveying rate of the conveying mechanism and the imaging frequency of the industrial vision camera; matching the sampling frequency, the conveying rate and the imaging frequency; Sampling and transmission: according to the sampling frequency, the test material is obtained from the loose moisture return outlet through the sampling plate, and the test material is spread on the conveying mechanism; according to the conveying rate, the test material is conveyed through the industrial vision camera, and then the image of the test material is obtained through the industrial vision camera; the test material passing through the industrial vision camera is synchronously conveyed back to the loose moisture return outlet; Detection and analysis: The image is transmitted to the detection and analysis platform via an electrical signal; the agglomeration contour of the image is identified, and the first area S of the agglomeration contour is calculated. 1 ; According to the conveying rate V, the imaging frequency f and the conveying mechanism width L, calculate the total area S of the detected material; according to the first area S 1 and the total area S, and calculate the looseness rate of the test material; The calculation formula of the loose rate is as follows:
2. According to the method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection in claim 1, It is characterized in that The detection analysis specifically also includes the following steps: Identify the condensation contour in the image, calculate the second area of the condensation contour; and analyze the condensation state of the cigarette sheet based on the second area.
3. The method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection according to claim 2, It is characterized in that The detection analysis also includes the following steps: Obtain historical data on loosening and regaining effects and store loosening rate data; mark the first and second moments when agglomerated tobacco sheets and condensed tobacco sheets are obtained, respectively, and build and display a knowledge graph of loosening and regaining in conjunction with online temperature detection results and online moisture detection results; The knowledge graph includes: first-level status items and second-level cause items arranged in a mutual hierarchy; the first-level status items include: agglomerated tobacco leaves, condensation, low temperature, high temperature, low moisture content, and high moisture content; the second-level cause items are the causes of occurrence of each of the first-level status items; the causes of occurrence are historical cause conclusions recorded by operators.
4. The method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection according to claim 3, It is characterized in that Constructing the knowledge graph specifically also includes the following steps: establishing a solution corresponding to each of the secondary cause items based on the historical data and the historical solutions recorded by the operator, and creating a link in the secondary cause item so that the user can open the corresponding solution when clicking on the secondary cause item.
5. The method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection according to any one of claims 1 to 4, It is characterized in that Identifying the agglomeration contour of the image specifically comprises the following steps: The image imread is read through the Python program, and the image imread is converted into a grayscale image cvtColor; Gaussian blur is applied to remove noise GaussianBlur; a threshold is set, and the grayscale image is processed according to the threshold so that the grayscale image highlights the dark area threshold; an opening operation is used to remove small objects, and a closing operation is used to connect adjacent objects, thereby finding and selecting the agglomeration contour.
6. The method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection according to any one of claims 1 to 4, It is characterized in that Calculating the first area of the agglomeration profile specifically comprises the following steps: Draw the edge of the agglomerate outline and calculate the first area of the agglomerate outline; if the first area is ≥ 3 cm 2 , it is determined that the tobacco sheet corresponding to the image has agglomerated tobacco leaves; if the first area is less than 3cm 2 ; It is determined that the tobacco sheet corresponding to the image is not agglomerated.
7. The method for detecting and analyzing the loosening and rehumidification effect of tobacco sheets based on machine vision detection according to any one of claims 1 to 4, It is characterized in that The sampling specifically includes the following steps: firstly, taking the material from below the loose moisture return discharge port, and then scraping the sampled test material to the transmission mechanism; The transmission mechanism transmits the inspection material in sections and continuously, specifically including the following steps: first, the inspection material is transmitted upward at an angle, and then the inspection material is transmitted horizontally; during the horizontal transmission process, the conveying direction of the inspection material is turned 180°; the inspection material after the turn passes through the industrial vision camera, and finally the inspection material is transmitted back to the subsequent process below the loose rehumidification discharge port.
8. A tobacco sheet loosening and moisture regaining effect detection and analysis system, It is characterized in that Used to implement the detection and analysis method described in any one of claims 1 to 7, comprising: a rotating sampling mechanism, a conveying mechanism, an industrial vision camera and a detection and analysis platform; the rotating sampling mechanism can rotate horizontally to obtain the detection material from the loose rehumidification discharge port, and can place the detection material on the conveying mechanism by rotation; the conveying mechanism is used to spread and convey the detection material, so that the detection material passes through the industrial camera, and then the image of the detection material can be obtained; the conveying mechanism is also used to return the detection material that has passed the industrial vision camera to the discharge vibration trough.
Citation Information
Cited By
Device for detecting stretching degree of tobacco leaves after loosening and moisture regaining processing
CN119845908A
A device for detecting the degree of relaxation of tobacco leaves after bulk conditioning
CN119845908B
