Visual inspection test device for tobacco production line

Abstract

The present application belongs to the technical field of tobacco detection, and particularly relates to a visual detection test device for a tobacco production line, which comprises a conveyor belt, a vibrating device, a first camera and a second camera. The rear end of the vibrating device is located above the front end of the conveyor belt, the vibrating device is arranged on a vibrating device support, and the conveyor belt is arranged on a conveyor belt support. A first camera support is arranged on the vibrating device, and the first camera is arranged on the first camera support above the vibrating device. A second camera support is arranged on the conveyor belt, and the second camera is arranged on the second camera support above the conveyor belt. The first camera and the second camera are both connected to an industrial computer electric signal. The technical scheme integrates various conveying parameter adjustment, tobacco thickness control, image analysis processing and automation in one device, can accurately control various parameters to realize detection of tobacco mixing uniformity, and can simulate different sections of an actual production line.

Description

Technical Field

[0001] This invention belongs to the field of tobacco shred inspection technology, specifically relating to a visual inspection testing device for a tobacco shred production line. Background Technology

[0002] Tobacco blending uniformity testing is a crucial step in ensuring stable product quality during tobacco production. The uniformity of tobacco blending affects the taste, combustion characteristics, and chemical stability of cigarettes. Therefore, accurate testing of its uniformity is essential for cigarette production processes, especially those involving tobacco processing, grouping, leaf threshing and destemming, blending and flavoring, vacuum rehydration, and re-drying. The core objective of tobacco blending uniformity testing during tobacco processing is to ensure that different types and grades of tobacco achieve a uniform distribution of components after mixing, thereby guaranteeing the stability of the cigarette product's flavor, combustion characteristics, and chemical composition. As the tobacco industry's requirements for product quality consistency increase, testing methods have gradually shifted from traditional manual experience and random sampling to assessments based on advanced technologies such as physics, chemistry, spectral analysis, computer vision, and artificial intelligence. Currently, the development of computer vision and artificial intelligence technologies has significantly improved the efficiency and accuracy of tobacco blending uniformity testing. High-resolution industrial cameras combined with image processing algorithms can analyze the distribution of each component after tobacco blending in real time. Then, deep learning models are used for image recognition, improving the automated assessment capability of blending uniformity. Simultaneously, by combining big data analysis, long-term measured data on tobacco uniformity can be modeled and optimized to improve the blending process and enhance production stability. While some testing equipment for tobacco production lines exists on the market, most only monitor a single parameter (such as thickness or speed), lacking a comprehensive testing system. Existing uniformity testing systems are mostly applicable only to fixed production parameters, making it difficult to automatically adjust device speed and tobacco thickness as needed, and unable to simulate the impact of different production line configurations on tobacco uniformity. Secondly, the testing methods are limited. Traditional testing equipment mostly uses mechanical contact measurement, which struggles to provide stable real-time monitoring data in high-speed production environments. Some systems only provide post-processing analysis data, resulting in delayed data feedback and hindering real-time adjustments and optimizations during production. Furthermore, if the testing equipment lacks vibration-proof design, the camera system will be affected by production line vibrations, leading to decreased detection accuracy and difficulty in obtaining high-quality image data. Summary of the Invention

[0003] The purpose of this invention is to provide a visual inspection and testing device for tobacco production lines. Equipped with an integrated encoder, adjustable speed control system, high-precision camera detection system, and vibration-damping bracket, the entire device integrates various conveying parameter adjustments, tobacco thickness control, image analysis and processing, and automation. It can accurately control various parameters to detect the uniformity of tobacco blending and can simulate different sections of the actual production line. This has important practical significance for determining the actual impact of different parameters on the uniformity of tobacco.

[0004] To achieve the above objectives, this application employs the following technical solution:

[0005] A visual inspection testing device for a tobacco production line, comprising a conveyor belt, a vibration device, a first camera, and a second camera;

[0006] The rear end of the vibration device is located above the front end of the conveyor belt, the vibration device is mounted on the vibration device support, and the conveyor belt is mounted on the conveyor belt support.

[0007] A first camera bracket is mounted across the vibration device, and the first camera is mounted on the first camera bracket above the vibration device.

[0008] A second camera bracket is mounted across the conveyor belt, and the second camera is mounted on the second camera bracket above the conveyor belt.

[0009] Both the first camera and the second camera are connected to industrial control computer electrical signals;

[0010] The vibration device is driven by a first drive motor, which is equipped with a first incremental rotary encoder. The conveyor belt is driven by a second drive motor, which is equipped with a second incremental rotary encoder. Both the first and second incremental rotary encoders are connected to the PLC controller and industrial control electromechanical signals.

[0011] Sensors are installed on both sides of the vibration device and both sides of the conveyor belt. The sensors on both sides of the vibration device and both sides of the conveyor belt are connected to the electrical signals of the industrial control computer.

[0012] Furthermore, casters are installed below both the vibrating device support and the conveyor belt support, and height adjustment devices are installed on both the legs of the vibrating device support and the legs of the conveyor belt support.

[0013] Furthermore, baffles are installed on both sides of the vibrating device and the conveyor belt.

[0014] Furthermore, grooves are provided on the surface of both the vibrating device and the conveyor belt.

[0015] Furthermore, the vibration device includes a crank-rocker mechanism and a universal joint rod, both of which are driven by a first drive motor.

[0016] The testing method for any of the above-mentioned visual inspection testing devices for tobacco production lines shall adopt the following steps:

[0017] S1. Determine the initial parameters of the device;

[0018] S2. Add the manually proportioned tobacco to the front end of the vibration device;

[0019] S3. Start the device. The incremental rotary encoder transmits the detected data to the industrial control computer. The industrial control computer judges the speed difference between the vibration device and the conveyor belt and analyzes whether it meets the expected requirements. If so, proceed to step S4.

[0020] The sensors on both sides of the vibration device detect the movement and screening effect of the tobacco shreds and transmit the data to the industrial control computer. The industrial control computer determines whether the screening effect of the tobacco shreds meets the expected requirements. If yes, it proceeds to step S7; otherwise, it proceeds to step S6.

[0021] S4. By sampling and testing the uniformity of the tobacco shreds, determine whether the test results meet the expected requirements. If so, proceed directly to step S5.

[0022] S5. The first camera captures an image of the tobacco shreds and transmits it to the industrial control computer. The industrial control computer determines whether the image effect meets the expected requirements. If not, proceed to step S6; if yes, proceed to step S8.

[0023] S6. After changing the speed mode, return to step S5;

[0024] S7. The second camera captures an image and transmits it to the industrial control computer, then proceeds to step S8;

[0025] S8. Establish a standard for tobacco uniformity after conducting tobacco uniformity testing.

[0026] Furthermore, in step S1, the initial parameters of the device are determined, including the device position, height, speed, and angle parameters.

[0027] Furthermore, step S3 also includes, if not, shutting down the device, adjusting the various parameters of the device, and then restarting the device.

[0028] Furthermore, in step S8, the uniformity of the tobacco shreds is detected, specifically as follows:

[0029] Based on machine vision and deep learning technologies, an artificial neural network model was built, and network structures such as convolutional layers, pooling layers, fully connected layers, and activation functions were designed to identify tobacco samples with different blending ratios. By improving and adjusting the model parameters of existing artificial neural networks, a large number of experiments and model training were conducted on a self-built tobacco dataset. The model can determine the content ratio of each component in tobacco and use object detection algorithms to outline and identify specific tobacco components.

[0030] The beneficial effects of this invention are:

[0031] This invention proposes a testing device for a tobacco production line, integrating an encoder, an adjustable speed control system, a high-precision camera detection system, and a vibration-damping bracket. It achieves the following innovations: 1. Simulating speed variations across different production lines: Through the encoder and adjustable speed control module, the speed of the later stage can be set to be greater than, equal to, or less than the speed of the earlier stage, accurately simulating the uniformity changes of different types of tobacco production lines or each section of the production line. Furthermore, the thickness and distribution area of ​​the tobacco shreds can be controlled by speed variations, analyzing the impact of different speeds on tobacco quality to optimize production parameters. 2. Real-time monitoring of tobacco shred thickness and uniformity: A high-precision camera system is used to acquire tobacco shred images in real time and analyze the trend of tobacco thickness changes. By adjusting the camera at multiple angles and heights, comprehensive coverage of the production line area is ensured, avoiding blind spots and ensuring the uniformity of the blending of the same batch of tobacco shreds or different types of tobacco shreds. The device can also automatically adjust various parameters to change the uniformity of the tobacco shreds by analyzing the quality of the tobacco shred uniformity images acquired by the camera. 3. Improving the detection capability of blending uniformity: Image processing algorithms are used to analyze the color and texture distribution of the tobacco shreds to determine the blending uniformity, achieving intelligent evaluation of tobacco shred quality. 4. Reduce vibration interference and improve detection accuracy: The camera is mounted on an independent anti-vibration bracket, reducing the impact of production line vibration on image acquisition and improving detection accuracy. 5. Data-driven intelligent adjustment: By combining real-time detection data, the system can automatically optimize speed parameters, reducing human intervention and improving production efficiency and stability. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the entire device of the present invention;

[0033] Figure 2 This is a side view of the complete device of the present invention;

[0034] Figure 3 This is a schematic diagram of the overall conveyor belt of the present invention;

[0035] Figure 4 This is a schematic diagram of the overall vibration device of the present invention;

[0036] Figure 5 This is a flowchart illustrating the overall framework of the present invention.

[0037] Explanation of reference numerals in the attached figures:

[0038] 1. Vibration device; 2. Conveyor belt; 3. Vibration device bracket; 4. Casters; 5. Crank rocker mechanism; 6. Universal fisheye rod; 7. First camera bracket; 8. First camera; 9. Conveyor belt bracket; 10. Second camera bracket; 11. Second camera; 12. First drive motor; 13. Second drive motor; 14. Baffle; 15. Lever. Detailed Implementation

[0039] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings. The following embodiments are merely exemplary and can only be used to explain and illustrate the technical solution of the present invention, and should not be construed as limiting the technical solution of the present invention.

[0040] like Figures 1 to 4 As shown, this invention provides a visual inspection testing device for a tobacco production line, used for real-time detection and simulation of speed changes, thickness changes, uniformity, and blending conditions during the tobacco production process. This device mainly achieves intelligent monitoring and simulation testing through an adjustable speed conveyor belt and vibrating tobacco separating device, a dynamic speed control system, a high-precision visual inspection system, and a data analysis and feedback control system, and optimizes the tobacco production process. By analyzing the influence of various conveying parameters on tobacco uniformity and adjusting these parameters to obtain different images of tobacco blending uniformity differences, this device can not only simulate the influence of different production processes on tobacco thickness and uniformity, and determine the different effects of various conveying parameters on uniformity images, thereby determining the optimal parameters for visual inspection, but also be used for quality inspection, process optimization, and experimental research.

[0041] A visual inspection testing device for a tobacco production line includes a conveyor belt 2, a vibration device 1, a first camera 8, and a second camera 11. The rear end of the vibration device 1 is located above the front end of the conveyor belt 2. The vibration device 1 is mounted on a vibration device support 3, and the conveyor belt 2 is mounted on a conveyor belt support 9. A first camera support 7 spans across the vibration device 1, and the first camera 8 is mounted on the first camera support 7 above the vibration device. A second camera support 10 spans across the conveyor belt 2, and the second camera 11 is mounted on the second camera support 10 above the conveyor belt. Both the first camera 8 and the second camera 11 are connected to industrial control electromechanical signals. The vibration device 1 is driven by a first drive motor 12, which is equipped with a first incremental rotary encoder. The conveyor belt 2 is driven by a second drive motor 13, which is equipped with a second incremental rotary encoder. Both the first and second incremental rotary encoders are connected to a PLC controller and industrial control electromechanical signals. Sensors are installed on both sides of the vibration device and both sides of the conveyor belt, and these sensors are all connected to industrial control electromechanical signals.

[0042] The experimental setup mainly includes a conveyor belt, a vibration device, and two camera supports. The conveyor belt and vibration device are supported on both sides by baffles 14 to prevent the tobacco from detaching during movement, and casters are installed at the bottom for adjustment of the device's fixation and movement. During operation, the device can be fixed in one position for stability, and can be moved when a change of location is needed. Aluminum profiles are additionally installed on each leg of the device and connected with sliders, nuts, and bolts, allowing for manual adjustment of the device's height and angle. The device uses high-quality bearing bases and high-strength gears, ensuring safe, stable, and reliable operation without vibration. Thickened angle brackets are used to independently connect the aluminum profiles, providing a sturdy and robust structure with strong support for weight and preventing wobbling. All parts of the machine body are connected by screws, making installation and disassembly convenient. The vibration device uses a crank-rocker mechanism 5 and a universal fisheye rod 6 to better separate the tobacco during vibration. The camera supports are floor-mounted, avoiding the vibration and instability associated with installation on the conveyor belt, and allowing the camera supports to be moved to any location for experiments on an actual production line. The conveyor belt and vibrating device are textured with herringbone patterns, primarily to increase friction, prevent the tobacco from slipping irregularly, and guide the tobacco towards the conveying direction. When the entire system starts, a manually fed sample of pre-mixed or complexly blended tobacco is placed on the vibrating device away from the conveyor belt. The motor drives the crank-rocker mechanism, causing the entire machine to vibrate and screen the tobacco. Simultaneously, the first camera begins capturing images and performing analysis. After screening, the mixed tobacco falls onto the conveyor belt and is transported to the other end. There, a second camera captures images of the tobacco and uploads them to a computer for uniformity testing. The entire process is automated and can simulate different tobacco production lines by adjusting the conveyor belt angle and the height difference between the conveyor belt and the vibrating device, providing a more complete image of the tobacco blending process.

[0043] like Figure 5 As shown, this application provides a detection method for the visual inspection testing device for the above-mentioned tobacco production line, which adopts the following steps:

[0044] S1. Determine the initial parameters of the device, including the device position, height, speed, and angle parameters.

[0045] S2. Add the manually proportioned tobacco to the front end of the vibration device.

[0046] S3. Start the device. The incremental rotary encoder transmits the detected data to the industrial control computer. The industrial control computer judges the speed difference between the vibration device and the conveyor belt and analyzes whether it meets the expected requirements. If so, proceed to step S4.

[0047] The sensors on both sides of the vibration device will detect the movement and screening effect of the tobacco shreds and transmit the data to the industrial control computer. The industrial control computer will determine whether the screening effect of the tobacco shreds meets the expected requirements. If so, proceed to step S7; otherwise, proceed to step S6.

[0048] S4. By sampling and testing the uniformity of the tobacco shreds, determine whether the test results meet the expected requirements. If so, proceed directly to step S5.

[0049] S5. The first camera captures an image of the tobacco shreds and transmits it to the industrial control computer. The industrial control computer determines whether the image effect meets the expected requirements. If not, proceed to step S6; if yes, proceed to step S8.

[0050] S6. After changing the speed mode, return to step S5.

[0051] S7. The second camera captures an image and transmits it to the industrial control computer, then proceeds to step S8.

[0052] S8. Establish a standard for tobacco uniformity after conducting tobacco uniformity testing.

[0053] Dynamic speed control is the core part of the detection and test device for the cut tobacco production line of the present invention. Through real-time speed measurement by an encoder, motor drive, and intelligent control algorithms, it realizes the simulation and optimized control of different technological processes. This system can set different conveyor speed modes, dynamically adjust the cut tobacco thickness and uniformity, and adapt to the operating conditions of different production lines. Specifically, incremental rotary encoders are respectively installed on the stepping motors of the conveyor belt and the vibration device to monitor the speed of the conveyor belt and the vibration speed of the vibration device in real time. The encoder converts the rotational movement of the shaft into an electrical pulse signal by means of photoelectric detection or magnetic detection, and outputs orthogonal pulses of phase A and phase B to transmit data to the PLC and the computer for real-time rotational speed analysis. The motor encoder and the PLC are interconnected using a motor driver. The intelligent control algorithm issues instructions to the motor drive system to ensure that the equipment operates according to the set speed mode and dynamically adjusts the cut tobacco uniformity. 0. By programming on the computer, different speed modes of the two devices can be set and adjusted according to different requirements, mainly including the constant speed mode V1 = V2, the acceleration mode V1 < V2, the deceleration mode V1 > V2, and the dynamic adjustment mode, to control the accumulation speed of the cut tobacco on the device. When the accumulation of a certain section of cut tobacco is too thick or too thin, the system will automatically adjust the speed of the corresponding part. When the speed of one device is lower than, higher than, or equal to the speed of another device, the encoder can immediately transmit the speed data to the computer and then automatically control the speed of the other device, where the operator can set the conveying speed through the interface. In addition, the uniformity analysis detection results can be obtained by analyzing the cut tobacco images captured by the camera, and the operating speed of the device can be automatically adjusted by analyzing different screening thicknesses of the cut tobacco to control the thickness and distribution of the cut tobacco. Sensors are installed on both sides of the whole device, and the sensors will detect the operation and screening effect of the cut tobacco and feedback it to the camera to control the start, stop, and shooting efficiency of the camera. The whole equipment is installed with a PLC system, and the operator can remotely monitor and control through a wireless network, monitor the equipment status and adjust the set parameters, and can also predict equipment anomalies in advance to reduce downtime and improve the automation level.

[0054] A high-precision visual inspection system utilizes a high-speed industrial camera combined with a high-precision optical system to capture real-time images of the uniform distribution of tobacco shreds on conveyor belts and vibrating devices. Based on image feedback, the system detects the uniformity of each component of the tobacco shreds and can also control the operating speed of the device in real time. The camera is mounted on a floor-standing, adjustable, and stable support to prevent external vibrations from affecting the inspection results. By adjusting the height of the support and the camera's shooting angle and installation position, the system can accurately determine the blending uniformity of the same batch of tobacco shreds. The camera captures a large number of images of the tobacco shreds at high speed, acquiring their surface contour information. Based on the obtained original images, preprocessing is performed, including normalization, noise reduction, image enhancement, and image segmentation. The effectiveness and accuracy of different image preprocessing algorithms in extracting target features are compared. The system extracts and identifies color, texture, shape, local and deep features from a vast dataset of various tobacco shred ratios and compositional structures. A ring light source is installed below the images to create contrast between the tobacco shreds and the background, enhancing the edge features of the tobacco shreds. Then, by combining known calibration ratios, the system determines the blending uniformity of the tobacco shreds, or by using extensive experimental data to determine more different ratio standards that can be applied to actual production.

[0055] Blending uniformity detection refers to the process of building an artificial neural network model based on machine vision and deep learning technologies. This model incorporates convolutional layers, pooling layers, fully connected layers, and activation functions to identify tobacco samples with different blending ratios. By improving and adjusting the model parameters of existing artificial neural networks, and conducting extensive experiments and model training on a self-built tobacco dataset, the system can determine the content proportion of each component in the tobacco shreds and use object detection algorithms to outline and identify specific tobacco components. The system compares pre-classified tobacco types from the test production line with manually selected tobacco components to establish standards for tobacco uniformity.

[0056] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention. Those skilled in the art can make various modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the scope defined in the claims.

Claims

1. A visual inspection testing device for a tobacco production line, characterized in that, Includes a conveyor belt, a vibrating device, a first camera, and a second camera; The rear end of the vibration device is located above the front end of the conveyor belt, the vibration device is mounted on the vibration device support, and the conveyor belt is mounted on the conveyor belt support. A first camera bracket is mounted across the vibration device, and the first camera is mounted on the first camera bracket above the vibration device. A second camera bracket is mounted across the conveyor belt, and the second camera is mounted on the second camera bracket above the conveyor belt. Both the first camera and the second camera are connected to industrial control computer electrical signals; The vibration device is driven by a first drive motor, which is equipped with a first incremental rotary encoder. The conveyor belt is driven by a second drive motor, which is equipped with a second incremental rotary encoder. Both the first and second incremental rotary encoders are connected to the PLC controller and industrial control electromechanical signals. Sensors are installed on both sides of the vibration device and both sides of the conveyor belt. The sensors on both sides of the vibration device and both sides of the conveyor belt are connected to the electrical signals of the industrial control computer.

2. The visual inspection testing device for tobacco production line according to claim 1, characterized in that, Both the vibrating device support and the conveyor belt support are equipped with casters, and both the legs of the vibrating device support and the legs of the conveyor belt support are equipped with height adjustment devices.

3. The visual inspection testing device for tobacco production line according to claim 1, characterized in that, Baffles are installed on both sides of the vibrating device and the conveyor belt.

4. The visual inspection testing device for tobacco production line according to claim 1, characterized in that, Textured grooves are provided on the surface of both the vibrating device and the conveyor belt.

5. The visual inspection testing device for tobacco production line according to claim 1, characterized in that, The vibration device includes a crank-rocker mechanism and a universal joint rod, both of which are driven by a first drive motor.

6. The detection method of the visual inspection testing device for tobacco production line according to any one of claims 1 to 5, characterized in that, The following steps are adopted: S1. Determine the initial parameters of the device; S2. Add the manually proportioned tobacco to the front end of the vibration device; S3. Start the device. The incremental rotary encoder transmits the detected data to the industrial control computer. The industrial control computer judges the speed difference between the vibration device and the conveyor belt and analyzes whether it meets the expected requirements. If so, proceed to step S4. The sensors on both sides of the vibration device detect the movement and screening effect of the tobacco shreds and transmit the data to the industrial control computer. The industrial control computer determines whether the screening effect of the tobacco shreds meets the expected requirements. If yes, it proceeds to step S7; otherwise, it proceeds to step S6. S4. By sampling and testing the uniformity of the tobacco shreds, determine whether the test results meet the expected requirements. If so, proceed directly to step S5. S5. The first camera captures an image of the tobacco shreds and transmits it to the industrial control computer. The industrial control computer determines whether the image effect meets the expected requirements. If not, proceed to step S6; if yes, proceed to step S8. S6. After changing the speed mode, return to step S5; S7. The second camera captures an image and transmits it to the industrial control computer, then proceeds to step S8; S8. Establish a standard for tobacco uniformity after conducting tobacco uniformity testing.

7. The detection method of the visual inspection testing device for tobacco production line according to claim 6, characterized in that, In step S1, the initial parameters of the device are determined, including the device position, height, speed, and angle parameters.

8. The detection method of the visual inspection testing device for tobacco production line according to claim 6, characterized in that, Step S3 also includes, if not, shutting down the device, adjusting the parameters of the device, and then restarting the device.

9. The detection method of the visual inspection testing device for tobacco production line according to claim 6, characterized in that, Step S8 involves detecting the uniformity of the tobacco shreds, specifically as follows: Based on machine vision and deep learning technologies, an artificial neural network model was built, and network structures such as convolutional layers, pooling layers, fully connected layers, and activation functions were designed to identify tobacco samples with different blending ratios. By improving and adjusting the model parameters of existing artificial neural networks, a large number of experiments and model training were conducted on a self-built tobacco dataset. The model can determine the content ratio of each component in tobacco and use object detection algorithms to outline and identify specific tobacco components.

Images