A mechanism cigar gluing board cleanliness on-line detection and control system and method

Abstract

The application provides a mechanism cigar coating plate cleanliness online detection and control system and method, which comprises a cigar coating transfer manipulator, a coating plate, a coating plate surface cleanliness detection subsystem and a coating plate cleanliness control subsystem; the coating plate surface cleanliness detection subsystem comprises a phase acquisition unit and a cleanliness judgment unit, the phase acquisition unit is used for acquiring real-time images of the coating plate surface, and the cleanliness judgment unit is used for judging and grading the cleanliness of the coating plate surface; and the coating plate cleanliness control subsystem is used for performing different levels of cleaning treatment on the coating plate surface according to the judgment result of the cleanliness judgment unit. The application adopts the image comparison method to perform real-time cleanliness detection and treatment on the coating plate in the actual working process, effectively improves the qualified rate of the cigar coating, and solves the problems of raw material waste, low qualified rate of the cigar coating, downtime and idleness caused by the dirt of the coating plate.

Description

Technical Field

[0001] This invention relates to the field of machine-made cigar production and processing technology, specifically to an online detection and control system and method for the cleanliness of the coating plate of machine-made cigars. Background Technology

[0002] Traditional high-end cigars are hand-rolled, resulting in higher costs. Machine-made cigars, on the other hand, occupy an important position in the global cigar market due to their lower prices, diverse flavors, and ease of purchase. In recent years, as the machine-made cigar market has shown a positive trend, the quality control requirements for its production process have gradually increased, especially the quality inspection of the wrapper process, which has become one of the key factors affecting the product qualification rate.

[0003] In the production of machine-made cigars, the wrapping process is mainly completed by automated equipment. The specific process is as follows: a robotic arm picks up the inner tobacco and places it onto the joining rollers; the robotic arm uses negative pressure suction to draw the wrapper from the wrapper conveyor belt and applies glue to it via a coating plate; the glued wrapper is then joined with the inner tobacco on the joining rollers to complete the joining process; the robotic arm transfers the shaped cigar to the finished product conveyor belt, completing the wrapping process for a machine-made cigar. The coating plate is a crucial component in ensuring the level of glue contamination during the wrapping process; however, this process is currently still in a semi-automated production mode, and its processing quality and efficiency are significantly affected by human factors.

[0004] The cleanliness of the coating plate is a key factor affecting the bonding effect of the wrapper. If the coating plate is contaminated (e.g., with residual glue, damaged wrappers), it will lead to uneven adhesion of the wrapper, or even some wrapper residue on the coating plate, affecting the quality of subsequent cigar splicing and causing continuous defective products. Currently, it mainly relies on manual inspection or timed cleaning, which has problems such as delayed response and untimely cleaning, which can easily lead to batch quality defects, waste of wrappers and inner tobacco materials, and affect production efficiency. It also requires high levels of skill and responsibility from operators and poses significant quality risks.

[0005] The patent "Metal Surface Cleanliness Detection Device" (CN 222704534 U) uses a photoelectric detector to determine whether the surface is dirty by observing changes in the intensity of reflected laser light. However, it is limited by the optical properties of tobacco materials, misjudgment of sticky contaminant types, environmental dust interference, and high calibration frequency requirements, making it unsuitable for detecting the cleanliness of machine-made cigar coating plates. Meanwhile, the patents "An Automatic Cleaning Device for Spray Nozzles" (CN 110180723 B) and "An Online Automatic Cleaning Device for Spray Nozzles" (CN222710005 U) both use adhesive cleaning paper to clean the surface of the spray nozzles, without providing a method for checking cleanliness. They rely purely on manual experience or programmed cleaning, lacking objective evidence and failing to guarantee the stability of the coating plate's condition.

[0006] For the reasons mentioned above, how to detect and control the cleanliness of the coating plate of machine-made cigars is a technical problem that urgently needs to be solved by those skilled in the art.

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

[0008] The purpose of this invention is to address the shortcomings of existing technologies by providing an online detection and control system for the cleanliness of coating plates in machine-made cigars. This system uses an image comparison method to detect and process the cleanliness of the coating plate in real time during actual operation, effectively improving the pass rate of the wrapper on machine-made cigars and solving problems such as raw material waste, low pass rate of wrapper application, and downtime caused by dirty coating plates.

[0009] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A mechanism for online detection and control of cigar wrapper cleanliness includes a wrapper transfer robot, a wrapper, a wrapper surface cleanliness detection subsystem, and a wrapper cleanliness control subsystem. The eggplant wrapper transfer robot is used to transfer the eggplant wrapper to the glue coating plate for glue application, and then transfer the glued eggplant wrapper to the next process. The coating plate is used to apply adhesive to the surface of the eggplant skin, and the surface of the coating plate is the object to be inspected and cleaned; The surface cleanliness detection subsystem of the coated plate includes an image acquisition unit and a cleanliness judgment unit. The image acquisition unit is used to acquire real-time images of the coated plate surface, and the cleanliness judgment unit is used to compare the analysis results of the real-time images with built-in standard parameters in order to classify the cleanliness of the coated plate surface. The judgment process is as follows: Considering the material characteristics of the coating plate for machine-made cigars (such as stainless steel, with a smooth surface and specific reflective properties), the characteristics of the adhesive residue (the special adhesive for cigars is colorless and transparent, but forms a transparent film after residue, which has a significant difference in reflectivity from the coating plate itself), and the characteristics of the residual wrapper and its structure (mostly curved surfaces or irregularly shaped structures with drainage channels), the cleanliness judgment unit adopts the core method of "image preprocessing - feature extraction - threshold comparison - grading judgment". The specific steps are as follows: (1) Image preprocessing of the coated plate surface Considering the presence of tobacco dust, light fluctuations (a combination of natural and artificial light in the production workshop) and image distortion caused by the curved surface structure of the coated sheet in the working environment, the real-time images acquired by the image acquisition unit are first preprocessed to eliminate interference factors and restore the true surface condition. Specifically: (a) Distortion correction: Based on the three-dimensional structural parameters of the coating plate (preset in the parameter library of the cleanliness judgment unit), the perspective transformation algorithm is used to correct the geometric distortion caused by the curved surface imaging to ensure that the size and shape of the coating plate area in the image are consistent with the standard image. (b) Noise filtering: An adaptive median filtering algorithm is used to accurately filter the point noise generated by tobacco dust and the irregular noise at the edges of adhesive residue, while preserving the edge features of adhesive residue and impurities, and avoiding over-filtering that could lead to feature loss. (c) Lighting balance: The Retinex enhancement algorithm is used to eliminate the problem of uneven image brightness caused by workshop light fluctuations, highlight the difference in reflection between the coated plate body and the transparent film residue, and make the transparent film residue clearly distinguishable by enhancing the reflection contrast. (2) Feature extraction of coated plate Based on the material of the adhesive coating plate and the characteristics of the transparent adhesive residue, two types of core feature parameters are extracted from the pre-processed real-time image as the basis for cleanliness judgment, specifically including: (a) Illumination equalization: The Retinex enhancement algorithm is used to eliminate the uneven brightness of the image caused by the fluctuation of light in the workshop, highlighting the grayscale difference between the coating plate body and the residual impurities. In particular, for the transparent film residue, the reflective difference characteristics between it and the coating plate surface are enhanced by illumination equalization.

[0010] (b) Noise filtering: An adaptive median filtering algorithm is used to accurately filter the point noise generated by tobacco dust and the irregular noise at the edges of adhesive residue, while preserving the edge features of adhesive residue and impurities, and avoiding over-filtering that could lead to feature loss. (c) Reflective characteristics: The coated plate body (stainless steel material) has stable mirror reflective properties, with a reflective intensity value stable between 200-230 (standard image calibration value). However, the colorless transparent adhesive film residue will change the surface reflective path, resulting in a decrease in local reflective intensity (reflective intensity value 120-180). Through the image reflective intensity detection algorithm, the percentage of pixels whose reflective intensity deviates from the body range in the real-time image is counted (denoted as P1). This parameter directly reflects the overall coverage of the transparent adhesive film residue. (d) Regional distribution characteristics: Based on the structural characteristics of the coating plate (the guide groove area is the key area for coating, and the probability of transparent film residue is higher), the pre-processed image is divided into key areas (the guide groove and the surrounding 5mm range) and non-key areas (other surface areas). The proportion of abnormal reflective pixels in the two areas is counted separately (denoted as P2 and P3). The cleanliness status of the key areas is focused on to avoid misjudging the cleanliness level due to a small amount of residue in the non-key areas. (3) Threshold comparison The weights of the two types of core feature parameters extracted are calculated to obtain a comprehensive parameter for comparison as the analysis result. Then, the cleanliness of the coated plate surface is graded by comparing this comprehensive parameter with the standard parameter used for comparison. The adhesive coating plate cleanliness control subsystem includes a multi-level cleaning unit, which is used to perform different levels of cleaning treatment on the adhesive coating plate surface according to the judgment result of the cleanliness judgment unit.

[0011] This invention utilizes the time interval of the coating process in machine-made cigars. By employing image analysis, the coating plate is first photographed, and then compared with standard images and parameters to determine whether it is dirty. The degree of dirtiness is then graded, and different levels of cleaning are applied to the coating plate according to different grading levels, thereby achieving the purpose of controlling the degree of dirtiness on the coating plate surface.

[0012] Based on the above, the phase acquisition unit takes a phase after each coat coating is completed, and the cleanliness judgment unit makes a judgment on each phase acquisition accordingly.

[0013] The main purpose is to achieve real-time control and avoid affecting the effect of coating the eggplant skin.

[0014] Based on the above, the image acquisition unit is a high-definition camera.

[0015] Based on the above, the cleanliness judgment unit includes a built-in standard dirt image comparison expert library and corresponding parameter information for judgment, to determine the degree of dirt on the coating plate.

[0016] This method uses images and corresponding parameters from a standard image library as a reference for comparison, which is more efficient and direct. The standard images themselves are used for human observation, to assist in calibration, and as a reference.

[0017] Based on the above, the multi-level cleaning unit of the coated plate surface cleanliness control subsystem includes a cleaning agent spraying module, a clean water cleaning module, and an air compressor drying module. Depending on the degree of soiling, different combinations of these modules are used for cleaning. This is primarily to address varying cleaning needs by setting different levels of cleaning operations, thereby improving efficiency and avoiding damage to the coated plate caused by over-cleaning.

[0018] Based on the above, the device used to call different combinations of modules in the detergent spraying module, the clean water cleaning module, and the air compressor drying module is a PLC.

[0019] Based on the above, it also includes an inner tobacco conveyor belt, a wrapper conveyor belt, a finished tobacco conveyor belt, a splicing roller mechanism, an inner tobacco transfer robot, and a finished tobacco transfer robot; The inner tobacco transfer robot is used to transfer the inner tobacco from the inner tobacco conveyor belt to the twisting roller mechanism. The wrapper transfer robot is used to transfer the wrapper from the wrapper conveyor belt to the coating plate and the twisting roller mechanism in sequence. The finished tobacco transfer robot is used to transfer the twisted finished product from the twisting roller mechanism to the finished tobacco conveyor belt.

[0020] It mainly outlines the entire process of making the inner-coated cigarette wrapper into a finished product, illustrating the working conditions under which the testing and control steps of the coating plate are performed.

[0021] A method for online detection and control of the cleanliness of machine-made cigar coating plates, based on the aforementioned online detection and control system for the cleanliness of machine-made cigar coating plates, is implemented through the following steps: S1. During the interval between transferring the wrapper to the coating plate and completing the coating process, the wrapper is transferred to the rubbing roller mechanism. The online detection and control program for the cleanliness of the cigar coating is executed. S2. The image acquisition unit acquires real-time image information of the coating plate and sends it to the cleanliness judgment unit for judgment. The standard parameters corresponding to the built-in standard image are compared with the parameter information obtained from the real-time image analysis to judge and classify the degree of dirtiness of the coating plate in the real-time image. S3. Based on the judgment and grading results of the degree of dirtiness of the real-time image of the coating plate, determine whether cleaning is required. If cleaning is required, the coating plate cleanliness control subsystem selects the appropriate level of cleaning method to execute the cleaning procedure. S4. Perform secondary image acquisition on the cleaned coating plate and execute step S2. Re-inspect the cleaned coating plate. If dirt is still present, continue to execute step S3. Repeat this process of re-inspection and cleaning until the coating plate no longer needs cleaning. If the cycle exceeds the set time or set number of times, trigger an alarm and suspend the coating operation of the eggplant coating and the detection of the cleanliness of the coating plate.

[0022] Based on the above, in step S2, the classification of the degree of dirt on the coating plate includes: Class A is severely dirty; Class B is dirty; and Class C is undirty. In step S3, the corresponding coating plate cleanliness control subsystem performs the following cleaning procedure for Class A: first, the coating plate is cleaned by the cleaning agent spraying module, then by the clean water cleaning module, and finally by the air compressor drying module. For scenario B, the cleaning procedure includes: first, the adhesive-coated plate is cleaned by the water cleaning module, and then it is dried by the air compressor drying module. For Category C cases, do not clean; continue with the eggplant skin coating application.

[0023] This invention has significant substantive features and remarkable progress compared to existing technologies. Specifically, it improves upon the entire process of applying cigar wrappers by providing a gap between the coating and transfer to the swivel roller mechanism and the removal of the next wrapper. This gap allows for the detection and control of the cleanliness of the coating plate. Specifically, the coating plate is photographed to obtain real-time images, which are then compared with stored standard images to determine whether cleaning is necessary. If cleaning is required, the degree of dirtiness is graded, and different cleaning procedures are performed accordingly, improving targeting and flexibility while saving time overall. After cleaning, continued testing enables efficient cleaning with dynamic feedback, ultimately improving the pass rate of the applied cigar wrappers. It solves problems such as raw material waste caused by dirty coating plates, low wrapper qualification rate, and machine downtime caused by idleness. At the same time, it reduces the labor intensity of operators, gets rid of the problem of over-reliance on the responsibility of operators, can also reduce the number of auxiliary production personnel, improve work efficiency, reduce raw material consumption, and reduce the quality defects such as cracked mouth, loose joint, and misaligned teeth in machine-made cigars. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the online detection and control system for the cleanliness of the coating plate of a cigar in this invention.

[0025] Figure 2 This is a flowchart of the online detection and control method for the cleanliness of the coating plate of a cigar in this invention.

[0026] In the diagram: 1. Inner tobacco conveyor belt; 2. Inner tobacco; 3. Inner tobacco transfer robot; 4. Jointing roller mechanism; 5. Wrapper transfer robot; 6. Wrapper conveyor belt; 7. Wrapper; 8. Coating plate; 9. Finished tobacco; 10. Finished tobacco transfer robot; 11. Finished tobacco conveyor belt; 12. Phase acquisition unit; 13. Cleanliness judgment unit; 14. Cleaning agent spraying module; 15. Clean water cleaning module; 16. Air compressor drying module. Detailed Implementation

[0027] The technical solution of the present invention will be further described in detail below through specific embodiments.

[0028] like Figure 1 and Figure 2 As shown, the process of coating the wrapper and rolling the inner cigarette is explained. The hardware required for the entire process includes an inner cigarette conveyor belt 1, a wrapper conveyor belt 6, a finished cigarette conveyor belt 11, a twisting roller mechanism 4, a wrapper transfer robot 5, an inner cigarette transfer robot 3, and a finished cigarette transfer robot 10.

[0029] The inner tobacco transfer robot 3 is used to transfer the inner tobacco 2 from the inner tobacco conveyor belt 1 to the twisting roller mechanism 4. The peel transfer robot 5 is used to transfer the peel of the 7 from the peel conveyor belt 6 to the coating plate 8 and the twisting roller mechanism 4 in sequence. The finished tobacco transfer robot 10 is used to transfer the twisted finished tobacco 9 from the twisting roller mechanism 4 to the finished tobacco conveyor belt 11.

[0030] After the peeler transfer robot 5 transfers the peeler 7 from the coating plate 8 to the joining roller mechanism 4, and then moves to the peeler conveyor belt 6 to pick up the next peeler, there is a period of downtime on the coating plate 8, during which the corresponding cleanliness detection and control procedures can be executed. Furthermore, an image is taken after each peeler coating is completed, and each image is evaluated.

[0031] Specifically, its core components include a surface cleanliness detection subsystem for coated plates and a surface cleanliness control subsystem for coated plates.

[0032] The surface of the coating plate 8 is the object to be inspected and cleaned. It is a relatively flat planar structure. After the coating plate 7 is coated, glue will remain on the surface. If the coating plate is damaged, some coating plate will easily remain, which will cause uneven coating or local glue stains in the next coating process. Therefore, it is necessary to inspect and clean it.

[0033] The surface cleanliness detection subsystem of the coated plate includes an image acquisition unit 12 and a cleanliness judgment unit 13. The image acquisition unit 12 is used to acquire real-time images of the coated plate surface, and the cleanliness judgment unit 13 is used to compare the real-time images with built-in standard images and parameters in order to classify the cleanliness of the coated plate surface.

[0034] Specifically, the image acquisition unit 12 in this embodiment uses a high-definition camera to capture images of the surface of the coated plate.

[0035] Explanation of the comparison principle: The eggplant wrapper transfer robot is used to transfer the eggplant wrapper to the glue coating plate for glue application, and then transfer the glued eggplant wrapper to the next process. The coating plate is used to apply adhesive to the surface of the eggplant skin, and the surface of the coating plate is the object to be inspected and cleaned; The surface cleanliness detection subsystem of the coated plate includes an image acquisition unit and a cleanliness judgment unit. The image acquisition unit is used to acquire real-time images of the coated plate surface, and the cleanliness judgment unit is used to compare the analysis results of the real-time images with built-in standard parameters in order to classify the cleanliness of the coated plate surface. The judgment process is as follows: Considering the material characteristics of the coating plate for machine-made cigars (such as stainless steel, with a smooth surface and specific reflective properties), the characteristics of the adhesive residue (the special adhesive for cigars is colorless and transparent, but forms a transparent film after residue, which has a significant difference in reflectivity from the coating plate itself), and the characteristics of the residual wrapper and its structure (mostly curved surfaces or irregularly shaped structures with drainage channels), the cleanliness judgment unit adopts the core method of "image preprocessing - feature extraction - threshold comparison - grading judgment". The specific steps are as follows: (1) Image preprocessing of the coated plate surface Considering the presence of tobacco dust, light fluctuations (a combination of natural and artificial light in the production workshop) and image distortion caused by the curved surface structure of the coated sheet in the working environment, the real-time images acquired by the image acquisition unit are first preprocessed to eliminate interference factors and restore the true surface condition. Specifically: (a) Distortion correction: Based on the three-dimensional structural parameters of the coating plate (preset in the parameter library of the cleanliness judgment unit), the perspective transformation algorithm is used to correct the geometric distortion caused by the curved surface imaging to ensure that the size and shape of the coating plate area in the image are consistent with the standard image. (b) Noise filtering: An adaptive median filtering algorithm is used to accurately filter the point noise generated by tobacco dust and the irregular noise at the edges of adhesive residue, while preserving the edge features of adhesive residue and impurities, and avoiding over-filtering that could lead to feature loss. (c) Lighting balance: The Retinex enhancement algorithm is used to eliminate the problem of uneven image brightness caused by workshop light fluctuations, highlight the difference in reflection between the coated plate body and the transparent film residue, and make the transparent film residue clearly distinguishable by enhancing the reflection contrast. (2) Feature extraction of coated plate Based on the material of the adhesive coating plate and the characteristics of the transparent adhesive residue, two types of core feature parameters are extracted from the pre-processed real-time image as the basis for cleanliness judgment, specifically including: (a) Illumination equalization: The Retinex enhancement algorithm is used to eliminate the uneven brightness of the image caused by the fluctuation of light in the workshop, highlighting the grayscale difference between the coating plate body and the residual impurities. In particular, for the transparent film residue, the reflective difference characteristics between it and the coating plate surface are enhanced by illumination equalization.

[0036] (b) Noise filtering: An adaptive median filtering algorithm is used to accurately filter the point noise generated by tobacco dust and the irregular noise at the edges of adhesive residue, while preserving the edge features of adhesive residue and impurities, and avoiding over-filtering that could lead to feature loss. (c) Reflective characteristics: The coated plate body (stainless steel material) has stable mirror reflective properties, with a reflective intensity value stable between 200-230 (standard image calibration value). However, the colorless transparent adhesive film residue will change the surface reflective path, resulting in a decrease in local reflective intensity (reflective intensity value 120-180). Through the image reflective intensity detection algorithm, the percentage of pixels whose reflective intensity deviates from the body range in the real-time image is counted (denoted as P1). This parameter directly reflects the overall coverage of the transparent adhesive film residue. (d) Regional distribution characteristics: Based on the structural characteristics of the coating plate (the guide groove area is the key area for coating, and the probability of transparent film residue is higher), the pre-processed image is divided into key areas (the guide groove and the surrounding 5mm range) and non-key areas (other surface areas). The proportion of abnormal reflective pixels in the two areas is counted separately (denoted as P2 and P3). The cleanliness status of the key areas is focused on to avoid misjudging the cleanliness level due to a small amount of residue in the non-key areas. (3) Threshold comparison The weights of the two types of core feature parameters extracted are calculated to obtain a comprehensive parameter for comparison as the analysis result. Then, the cleanliness of the coated plate surface is graded by comparing this comprehensive parameter with the standard parameter used for comparison. The adhesive coating plate cleanliness control subsystem includes a multi-level cleaning unit, which is used to perform different levels of cleaning treatment on the adhesive coating plate surface according to the judgment result of the cleanliness judgment unit.

[0037] In a preferred embodiment, the built-in comparison standard object in each system is the coating plate body used in the current device. In this embodiment, the coating plate is compared with its own dirt state, eliminating the influence caused by the differences in shape, structure and surface parameters of different coating plates. Moreover, the standard images of different dirt levels are all made under the same working environment as the image acquisition unit 12 with the same viewing angle, the same focal length and the same light source, which can further improve the accuracy and adaptability of the judgment.

[0038] Specifically, in this embodiment, the classification of the degree of dirt on the coating board includes: Class A is severely dirty; Class B is dirty; and Class C is undirty.

[0039] The adhesive coating plate cleanliness control subsystem includes a multi-level cleaning unit, which is used to perform different levels of cleaning treatment on the adhesive coating plate surface according to the judgment result of the cleanliness judgment unit.

[0040] Specifically, in this embodiment, the multi-level cleaning unit of the coating plate surface cleanliness control subsystem includes a cleaning agent spraying module 14, a clean water cleaning module 15, and an air compressor drying module 16. Depending on the degree of dirtiness, different combinations of these modules are invoked for cleaning. This is primarily to address varying cleaning needs by setting different levels of cleaning operations, thereby improving efficiency and avoiding damage to the coating plate itself caused by over-cleaning. In this embodiment, the device used to invoke different combinations of these modules is a PLC.

[0041] For scenario A, the cleaning procedure includes: first, the adhesive coating plate is cleaned by the cleaning agent spraying module, then by the clean water cleaning module, and finally by the air compressor drying module. For scenario B, the cleaning procedure includes: first, the adhesive coating plate is cleaned by the clean water cleaning module, and then by the air compressor drying module. For scenario C, no cleaning is performed, and the eggplant coating adhesive application operation continues. The method for cleanliness detection and control using the above system is implemented through the following steps: S1. During the interval between transferring the wrapper to the coating plate and completing the coating process, the wrapper is transferred to the rubbing roller mechanism. The online detection and control program for the cleanliness of the cigar coating is executed. S2. The image acquisition unit acquires real-time image information of the coating plate and sends it to the cleanliness judgment unit for judgment. The built-in standard image is compared with the real-time image and its parameters to judge and classify the degree of dirtiness of the coating plate in the real-time image. The classification of the degree of dirtiness of the coating plate includes: Class A is that the coating plate is severely dirty; Class B is that the coating plate is dirty; Class C is that the coating plate is not dirty. S3. Based on the judgment and grading results of the degree of dirtiness of the real-time image of the coating plate, determine whether cleaning is required. If cleaning is required, the coating plate cleanliness control subsystem selects the appropriate level of cleaning method to execute the cleaning procedure. For scenario A, the cleaning procedure includes: first, the adhesive coating plate is cleaned by the cleaning agent spraying module, then the adhesive coating plate is cleaned by the water cleaning module, and finally the adhesive coating plate is dried by the air compressor drying module. For scenario B, the cleaning procedure includes: first, the adhesive-coated plate is cleaned by the water cleaning module, and then it is dried by the air compressor drying module. For Category C cases, do not clean; continue with the eggplant skin coating application.

[0042] S4. Perform secondary image acquisition on the cleaned coating plate and execute step S2. Re-inspect the cleaned coating plate. If dirt is still present, continue to execute step S3. Repeat this process of re-inspection and cleaning until the coating plate no longer needs cleaning. If the cycle exceeds the set time or set number of times, trigger an alarm and suspend the coating operation of the eggplant coating and the detection of the cleanliness of the coating plate.

[0043] 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. An online detection and control system for the cleanliness of a cigar coating plate, characterized in that: This includes a manipulator for transferring eggplant peel, a coating plate, a coating plate surface cleanliness detection subsystem, and a coating plate cleanliness control subsystem; The eggplant wrapper transfer robot is used to transfer the eggplant wrapper to the glue coating plate for glue application, and then transfer the glued eggplant wrapper to the next process. The coating plate is used to apply adhesive to the surface of the eggplant skin, and the surface of the coating plate is the object to be inspected and cleaned; The surface cleanliness detection subsystem of the coated plate includes an image acquisition unit and a cleanliness judgment unit. The image acquisition unit is used to acquire real-time images of the coated plate surface, and the cleanliness judgment unit is used to compare the analysis results of the real-time images with built-in standard parameters in order to classify the cleanliness of the coated plate surface. The judgment process is as follows: (1) Image preprocessing of the coated plate surface First, the real-time image acquired by the image acquisition unit is preprocessed to eliminate interference factors from ambient light and restore the true surface state. (2) Feature extraction of coated plate Based on the material of the adhesive-coated plate and the characteristics of the transparent adhesive residue, two types of core feature parameters are extracted from the pre-processed real-time image as the basis for cleanliness judgment, specifically including: (a) Reflective difference characteristics: Compared with the stable mirror reflective characteristics of the coated plate body, the area where the local reflective intensity is reduced due to the residue of colorless transparent adhesive film is identified, and the percentage of pixels whose reflective intensity deviates from the body range in the real-time image is counted. This parameter directly reflects the overall coverage of the transparent adhesive film residue. (b) Regional distribution characteristics: Considering the structural characteristics of the coating plate, the guide groove area is the key area for coating and has a higher probability of glue residue. The preprocessed image is divided into a key area centered on the guide groove and other non-key areas. The proportion of impurity pixels in the two areas is counted respectively. (3) Threshold comparison The weights of the two types of core feature parameters extracted are calculated to obtain a comprehensive parameter for comparison as the analysis result. Then, the cleanliness of the coated plate surface is graded by comparing this comprehensive parameter with the standard parameter used for comparison. The adhesive coating plate cleanliness control subsystem includes a multi-level cleaning unit, which is used to perform different levels of cleaning treatment on the adhesive coating plate surface according to the judgment result of the cleanliness judgment unit.

2. The online detection and control system for the cleanliness of the coating plate of machine-made cigars according to claim 1, characterized in that: The phase-taking unit takes a phase after each coat of eggplant is coated with adhesive, and the cleanliness judgment unit makes a judgment on each phase taken.

3. The online detection and control system for the cleanliness of the coating plate of machine-made cigars according to claim 1 or 2, characterized in that: The image acquisition unit is a high-definition camera.

4. The online detection and control system for the cleanliness of the coating plate of machine-made cigars according to claim 1, characterized in that: The cleanliness judgment unit includes a built-in standard dirt image comparison expert library and corresponding parameter information for judgment, to determine the degree of dirt on the coating plate.

5. The online detection and control system for the cleanliness of the coating plate of machine-made cigars according to claim 1 or 4, characterized in that: The multi-level cleaning unit of the coating plate surface cleanliness control subsystem includes a cleaning agent spraying module, a clean water cleaning module, and an air compressor drying module. Depending on the degree of dirt, different combinations of modules from the cleaning agent spraying module, clean water cleaning module, and air compressor drying module are called for cleaning.

6. The online detection and control system for the cleanliness of the coating plate of machine-made cigars according to claim 5, characterized in that: A PLC is a device used to call different combinations of modules in the detergent spraying module, the clean water cleaning module, and the air compressor drying module.

7. The online detection and control system for the cleanliness of the coating plate of machine-made cigars according to claim 1, characterized in that: It also includes inner tobacco conveyor belts, wrapper conveyor belts, finished tobacco conveyor belts, splicing roller mechanisms, inner tobacco transfer robots, and finished tobacco transfer robots; The inner tobacco transfer robot is used to transfer the inner tobacco from the inner tobacco conveyor belt to the twisting roller mechanism. The wrapper transfer robot is used to transfer the wrapper from the wrapper conveyor belt to the coating plate and the twisting roller mechanism in sequence. The finished tobacco transfer robot is used to transfer the twisted finished product from the twisting roller mechanism to the finished tobacco conveyor belt.

8. A method for online detection and control of the cleanliness of a machine-made cigar coating plate, characterized in that: The online detection and control system for the cleanliness of the cigar coating plate according to any one of claims 1-7 is implemented through the following steps: S1. During the interval between transferring the wrapper to the coating plate and completing the coating process, the wrapper is transferred to the rubbing roller mechanism. The online detection and control program for the cleanliness of the cigar coating is executed. S2. The image acquisition unit acquires real-time image information of the coating plate and sends it to the cleanliness judgment unit for judgment. The standard parameters corresponding to the built-in standard image are compared with the parameter information obtained from the real-time image analysis to judge and classify the degree of dirtiness of the coating plate in the real-time image. S3. Based on the judgment and grading results of the degree of dirtiness of the real-time image of the coating plate, determine whether cleaning is required. If cleaning is required, the coating plate cleanliness control subsystem selects the appropriate level of cleaning method to execute the cleaning procedure. S4. Perform secondary image acquisition on the cleaned coating plate and execute step S2. Re-inspect the cleaned coating plate. If dirt is still present, continue to execute step S3. Repeat this process of re-inspection and cleaning until the coating plate no longer needs cleaning. If the cycle exceeds the set time or set number of times, trigger an alarm and suspend the coating operation of the eggplant coating and the detection of the cleanliness of the coating plate.

9. The method for online detection and control of the cleanliness of the coating plate for machine-made cigars according to claim 8, characterized in that: In step S2, the classification of the degree of dirt on the coating plate includes: Class A is severely dirty; Class B is dirty; and Class C is undirty. In step S3, the corresponding coating plate cleanliness control subsystem performs the following cleaning procedure for Class A: first, the coating plate is cleaned by the cleaning agent spraying module, then by the clean water cleaning module, and finally by the air compressor drying module. For scenario B, the cleaning procedure includes: first, the adhesive-coated plate is cleaned by the water cleaning module, and then it is dried by the air compressor drying module. For Category C cases, do not clean; continue with the eggplant skin coating application.

Images