An automated rice packaging system based on machine vision intelligent monitoring

Abstract

The application discloses an automatic rice packaging system based on intelligent monitoring of machine vision, which comprises a bag feeding device, a conveying device, a feeding device, a computer vision detection array and a main electric control box; through state monitoring and self-adaptive regulation provided by the computer vision detection array, the application realizes the upgrading of the automatic rice packaging process from structure driving to perception driving, so that the bagging stability, quantitative accuracy and action coherence are significantly improved. The bag feeding device ensures that the posture correction of the bag is completed before the bag enters the chain plate through visual identification; the conveying device ensures that the bag opening angle meets the feeding requirements through visual feedback; and the feeding device ensures that the material falling is continuous and smooth through flow state detection, so that an intelligent packaging equipment with reasonable structure, strong self-adjusting capacity and applicable to high-speed operation is finally constructed, and a more efficient and reliable automatic solution is provided for the grain processing industry.

Description

Technical Field

[0001] This invention belongs to the technical field of rice packaging equipment, specifically relating to an automated rice packaging system based on machine vision intelligent monitoring. Background Technology

[0002] Rice packaging equipment is one of the important production equipment for grain processing enterprises. With the development of automation technology, its function has gradually evolved from semi-automatic equipment assisted by manual labor to fully automated equipment throughout the entire process. The fully automatic rice packaging machines that are widely used in the market today typically consist of a bag picking mechanism, a bag conveying mechanism, a bag opening mechanism, a material unloading mechanism, and a sealing and conveying mechanism. Through the coordination of mechanical drives, pneumatic actuators, and PLC control, they achieve continuous automatic operation from bag supply, bag forming, material unloading to final output. The automatic bag feeding mechanism typically employs a suction or mechanical clamping structure to pick up packaging bags from the stacking area and transport them to a designated position. The bag conveying section often uses chains, clamping plates, or multi-segment conveyor belts to transfer and adjust the bag's position, ensuring accurate connection between different workstations. The bag opening process generally relies on suction cups for negative pressure adsorption, clamping plate separation, and top support mechanisms to open the bag opening for subsequent material injection. The material feeding structure usually uses control valves, weighing mechanisms, or multi-stage buffer structures to quantitatively dispense rice. The pre-sealing conveying and bag opening straightening mechanism maintains the stable shape of the bag opening, allowing it to smoothly enter the heat-sealing or sewing process. Overall, current rice packaging equipment can achieve a high degree of automated production and possesses a certain level of speed, stability, and processing adaptability. In actual production, existing equipment is often equipped with bag width adjustment mechanisms, bag bottom forming mechanisms, or auxiliary positioning devices to adapt to different types and materials of packaging bags, thereby improving the equipment's versatility and adaptability. To further improve packaging speed, some equipment has introduced multi-station parallel structures, such as dual-station bag picking and continuous feeding from dual weighing hoppers, which significantly improves the bagging cycle time. Furthermore, with the advancement of intelligent manufacturing concepts, modern rice packaging equipment has gradually added functions such as automatic detection, abnormal alarms, and automatic reset to enhance the overall intelligence of the production line and reduce manual intervention.

[0003] Despite significant advancements in automation, structural coordination, and processing efficiency, rice, being a granular bulk material, exhibits inherent uncertainties in its flowability, stacking properties, and discharge behavior. Furthermore, the significant deformation characteristics of flexible packaging bags during handling and forming complicate the coordination between various stages of equipment operation, including bag conveying, bag opening, discharge control, and bag closing. Particularly under high-speed continuous operation, packaging bags need to enter each station with the appropriate posture, while the rice needs to be bagged stably, continuously, and without spillage during discharge. This places higher demands on the equipment's structural design, motion coordination, and sealing control. However, existing rice packaging equipment still struggles to achieve smooth transitions between bag feeding, opening, and discharge in actual operation, easily leading to problems such as bag jamming and leakage that affect continuous operation. Summary of the Invention

[0004] The main objective of this invention is to overcome the technical problem that existing equipment often experiences packaging bag jamming or rice accumulation and blockage in transfer bins and discharge hoppers during multiple stages such as bag conveying, bag opening unfolding, and material transfer, resulting in frequent machine shutdowns. This invention provides an automated rice packaging system based on machine vision intelligent monitoring.

[0005] The secondary objective of this invention is to address the following issues in the prior art: (1) the packaging bags are unstable in position during transport to various workstations, making it difficult to maintain accurate centering. Existing equipment lacks effective limiting and centering structures during bag transfer, alignment, and posture adjustment, causing the bags to easily shift or tip over, affecting the smooth connection of subsequent bag opening and unloading processes; (2) the quantitative control of rice unloading is inaccurate, making it difficult to ensure consistent weight for each bag. Traditional unloading devices lack precision in flow regulation, sealing control, and unloading stability, causing fluctuations in rice unloading volume and making it difficult to achieve stable and reliable quantitative bagging.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] One aspect of the present invention provides an automated rice packaging system based on machine vision intelligent monitoring, including a bag feeding device, a conveying device, a feeding device, a computer vision detection array, and a main electrical control box.

[0008] The bag feeding device includes a bag storage mechanism, a bag clamping mechanism, and a bag opening mechanism; the bag storage mechanism is used to store and sequentially distribute packaging bags; the bag clamping mechanism and the bag opening mechanism are connected to the bag storage mechanism in sequence and are used to perform bag picking, clamping, and bag opening operations.

[0009] The conveying device includes a bag mouth clamping mechanism, a bag mouth folding machine, a sewing mechanism, and a conveying mechanism; the bag mouth clamping mechanism is used to clamp the bag mouth after the rice is filled; the bag mouth folding machine is used to neatly fold the edges of the bag mouth; the sewing mechanism is used to sew and seal the neatly folded bag mouth; the conveying mechanism is used to transport the finished rice package that has been sewn and sealed to the downstream process.

[0010] The feeding device is used to measure the weight of the rice to be packaged and to fill the measured rice into the opened packaging bag.

[0011] The computer vision detection array is used to collect images during the rice packaging process. The images record the bag posture and bag opening status, as well as the flow state of the rice during filling.

[0012] The main electrical control box is used to control the operation of the bag feeding device, conveying device and feeding device to complete the automated rice packaging;

[0013] The main electrical control box is equipped with a deep learning recognition network, which is used to perform semantic segmentation, key point extraction and temporal change analysis on the images acquired by the computer vision detection array to obtain state information; it is also used to determine whether the bag body posture and bag opening status of the packaging bag and the material flow state during rice filling are abnormal based on the state information. If abnormal, the main electrical control box controls and adjusts the bag feeding device, conveying device and feeding device.

[0014] As a preferred technical solution, the computer vision detection array includes several cameras, which are respectively installed in the bag feeding device, the conveying device, and the feeding device.

[0015] As a preferred technical solution, the deep learning recognition network is built on an improved convolutional neural network architecture, including a semantic segmentation module, a key point extraction module, and a temporal change analysis module;

[0016] The semantic segmentation module uses a U-Net or DeepLabV3+ architecture to perform pixel-level classification on the input packaging bag image, segmenting the bag body region and the material dropping region. The material dropping velocity is obtained by calculating the rate of change of the pixel area of ​​the material dropping region between consecutive frames. ;

[0017] The key point extraction module adopts an HRNet structure to locate the corner points and the center feature points of the bag opening. Compare it with the preset standard reference point Perform Euclidean distance comparison and calculate bag offset. ;

[0018] The time-series change analysis module employs a long short-term memory network to analyze the bag opening degree over continuous time steps. Regression analysis was performed to predict and determine whether the bag opening would rebound or close.

[0019] The method for determining whether the bag's posture and opening status, and the flow pattern of rice during filling are abnormal, based on status information, specifically includes:

[0020] Set the bag offset threshold, bag opening threshold, and material flow rate threshold;

[0021] When the bag body offset exceeds the bag body offset threshold, a signal is sent to the main electrical control box to control the bag feeding device to adjust the bag body posture.

[0022] When the opening of the packaging bag is less than the opening threshold, a signal is sent to the main electrical control box to control the conveying device to adjust the opening state of the packaging bag.

[0023] When the rice flow rate during filling is less than the flow rate threshold, a signal is sent to the main control box to control the feeding device to adjust the rice flow rate.

[0024] As a preferred technical solution, in the bag feeding device, the main body of the bag storage mechanism is a bag storage frame, and it is also equipped with a pallet moving mechanism, a bag dispensing mechanism, a bag feeding mechanism, a flipping bag loading mechanism, a bag pressing mechanism and a bag picking mechanism for cooperative use.

[0025] The pallet moving mechanism is used to carry and adjust the stacking position of the packaging bags; the bag separating mechanism is used to separate the packaging bags one by one; the bag feeding mechanism is used to directionally transport the separated packaging bags to the flipping bag loading mechanism; the flipping bag loading mechanism is used to adjust the packaging bags from a flat state to an upright state; the bag pressing mechanism and the bag picking mechanism are used to perform the adsorption and gripping of the packaging bags.

[0026] As a preferred technical solution, the bag feeding device further includes a bag limiting mechanism, a bag pressing adjustment mechanism, a packaging bag limiting plate mounting block, a packaging bag limiting plate, a bag storage air circuit box, and a vacuum generator;

[0027] The bag limiting mechanism, the bag pressing adjustment mechanism, the packaging bag limiting plate mounting block, and the packaging bag limiting plate are used together to constrain and correct the position and posture of the packaging bag.

[0028] The vacuum generator is used to provide negative pressure to the bag storage air circuit box; the bag storage air circuit box is used to manage the pneumatic control actions of the bag feeding device.

[0029] As a preferred technical solution, the feeding device is constructed on an independent frame and also includes a hopper and a weighing system. The hopper is suspended within the frame via a sensor connecting plate and a connecting rod, and is flexibly connected to the weighing system using a weighing connecting shaft. The feeding device's discharge control section includes a curved plate for forming a discharge channel and a switch plate for controlling the material flow. The switch plate is assembled via a shaft, bushing, and hinge block to achieve rotation and opening / closing. The weighing system uses pins to position the weighing module, and is equipped with sensor covers and locking accessories to protect the stainless steel hopper and weighing components. The weight signal is collected by a weighing sensor installed on the sensor connecting plate and drives the switch plate through a fisheye bearing to achieve force transmission and buffering. Finally, a cylinder controls the opening and closing of the discharge gate.

[0030] As a preferred technical solution, the feeding device further includes a weighing hopper and a dust removal device for reducing dust overflow installed at the discharge end of the weighing hopper; the opening and closing action of the weighing hopper is driven by an actuator installed on a cylinder support.

[0031] As a preferred technical solution, the main body of the conveying device is a frame assembly with casters supporting the bottom, which can be moved and adjusted horizontally; the frame assembly integrates a vertical lifting mechanism to match packaging bags of different sizes and heights; the bottom of the frame assembly is also provided with a finished product conveying and guiding mechanism to guide the finished rice packaging that has been sewn and sealed to the downstream process through the conveying mechanism.

[0032] As a preferred technical solution, the sewing mechanism includes a sewing machine head unit, a thread holder, a sewing drive motor, a sewing transmission mechanism, and a thread tensioner.

[0033] The thread placement rack, thread tension feeder, and sewing machine head unit are connected in sequence to control the thread tension and sew the opening of the packaging bag; the sewing drive motor and sewing transmission mechanism are used to provide sewing power.

[0034] As a preferred technical solution, a touch screen component connected to the main electrical control box is also included for human-computer interaction and parameter setting.

[0035] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0036] (1) Intelligent monitoring and closed-loop adjustment significantly reduce the failure rate. This invention introduces a computer vision detection array and a deep learning recognition network, which can acquire key information such as bag offset and bag opening degree in real time. When an abnormality is detected, the system can automatically send a signal to adjust the bag feeding or conveying device, correcting deviations before physical failures occur and avoiding downtime caused by inaccurate positioning.

[0037] (2) Monitoring the flow of falling material to improve filling accuracy. By visually monitoring the changes in the flow rate of falling material, and in conjunction with the precision weighing and cylinder control of the feeding device, the flow of falling rice is accurately controlled, effectively solving the problems of easy clogging and overflow in traditional equipment.

[0038] (3) Strong structural coordination. The independent feeding device frame design and the sensor coordination between various mechanisms solve the problem of process connection caused by deformation during the handling and forming of flexible packaging bags. Attached Figure Description

[0039] Figure 1 This is a three-dimensional structural diagram of an automated rice packaging system based on machine vision intelligent monitoring according to an embodiment of the present invention;

[0040] Figure 2 This is a three-dimensional structural schematic diagram of the bag feeding device according to an embodiment of the present invention;

[0041] Figure 3 This is a three-dimensional structural schematic diagram of the feeding device according to an embodiment of the present invention;

[0042] Figure 4 This is a three-dimensional structural diagram of the conveying device according to an embodiment of the present invention;

[0043] Figure 4 This is a three-dimensional structural diagram of the conveying device according to an embodiment of the present invention;

[0044] Figure 5 This is a schematic diagram of the visual algorithm detection control process according to an embodiment of the present invention.

[0045] Explanation of icon numbers:

[0046] 1. Equipment frame; 2. Bag storage mechanism; 3. Bag clamping mechanism; 4. Bag opening mechanism; 5. Bag mouth clamping mechanism; 6. Bag mouth folding machine; 7. Sewing machine head unit; 8. Thread placement rack; 9. Weighing system; 10. Conveying mechanism; 11. Touch screen assembly; 12. Main electrical control box; 13. First camera; 14. Second camera; 15. Third camera; 16. Bag storage frame; 17. Pallet moving mechanism; 18. Bag dispensing mechanism; 19. Bag feeding mechanism; 20. Flipping bag loading mechanism; 21. Bag pressing mechanism; 22. Bag retrieval mechanism; 23. Bag limiting mechanism; 24. Bag pressing adjustment mechanism; 25. Packaging bag limiting plate mounting block; 26. Packaging bag limiting plate; 27. Bag storage air circuit box; 28. Vacuum generator; 29. ​​Frame; 30. Sensor connecting plate; 31. Connecting rod; 32. First weighing connecting shaft; 33. Second weighing... 34. Connecting shaft; 35. First hopper; 36. Bend plate; 37. First switch plate; 38. Second switch plate; 39. Shaft; 40. Bushing; 41. Hinge block; 42. Pin; 43. Sensor cover; 44. Locking accessories; 45. Second stainless steel hopper; 46. Cylinder support; 47. Weighing hopper 2; 48. Dust removal device for feeding pipe; 49. Weighing sensor; 50. Fisheye bearing; 51. Cylinder; 52. Frame assembly; 53. Caster support foot assembly; 54. Vertical lifting mechanism; 55. Thread guide rail assembly; 56. Sewing drive motor; 57. Sewing transmission mechanism; 58. Thread tension frame; 59. Finished product conveying and guiding mechanism. Detailed Implementation

[0047] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative effort are within the scope of protection of the present application.

[0048] Example:

[0049] This invention proposes an intelligent and automated rice packaging system based on machine vision intelligent monitoring, applicable to the grain processing field. Its overall structure is as follows: Figure 1 As shown, the equipment frame 1 serves as the main supporting foundation, including a bag feeding device, a feeding device, a conveying device, a computer vision detection array, and a main electrical control box; each functional module is arranged in an orderly manner around the frame and operates collaboratively.

[0050] 1. Bag feeding device. It includes a bag storage mechanism 2, a bag clamping mechanism 3, and a bag opening mechanism 4, which are installed above the equipment frame 1 for storing and sequentially distributing packaging bags. The bag storage mechanism is connected to the bag clamping mechanism 3 and the bag opening mechanism 4 in sequence to complete the bag picking, clamping, and bag opening operations, thereby providing a stable and reliable bag opening state for subsequent filling processes.

[0051] like Figure 2 As shown, the bag feeding device also includes a pallet moving mechanism 17, a bag dispensing mechanism 18, a bag feeding mechanism 19, a flipping bag feeding mechanism 20, a bag pressing mechanism 21, a bag taking mechanism 22, a bag limiting mechanism 23, a bag pressing adjustment mechanism 24, a packaging bag limiting plate mounting block 25, a packaging bag limiting plate 26, a bag storage air circuit box 27, and a vacuum generator 28.

[0052] The main body of the bag storage mechanism 2 is the bag storage frame 16, which serves as the supporting foundation; the pallet moving mechanism 17 is used to carry and adjust the stacking position of the packaging bags; the bag separating mechanism 18 is used to separate the packaging bags one by one; the bag feeding mechanism 19 is used to directionally transport the separated packaging bags to the flipping bag loading mechanism 20; the flipping bag loading mechanism 20 is used to adjust the packaging bags from a flat position to an upright position; during the bag retrieval process, the bag pressing mechanism 21 and the bag retrieval mechanism 22 work together to perform the adsorption and gripping of the packaging bags; the bag limiting mechanism 23, the bag pressing adjustment mechanism 24, the packaging bag limiting plate mounting block 25, and the packaging bag limiting plate 26 are used together to constrain and correct the position and posture of the packaging bags; the vacuum generator 27 is used to provide a stable negative pressure to the bag storage air circuit box 28 to ensure the reliability of the bag retrieval process; the bag storage air circuit box 28 is used to manage the pneumatic control of the above-mentioned actions of the bag feeding device.

[0053] II. Feeding device. For example... Figure 3 The feeding device shown is built on an independent frame 29 and includes a first hopper 34 and a weighing system 9, which are used to weigh the rice to be put into the packaging bag and fill the weighed rice into the opened packaging bag.

[0054] The first hopper 34 is suspended within the frame 29 via a sensor connecting plate 30 and a connecting rod 31, and is flexibly connected to the weighing system 9 via a weighing connecting shaft (in a preferred embodiment, there are two: a first weighing connecting shaft 32 and a second weighing connecting shaft 33). The feeding device's feeding control part includes a curved plate 35 for forming a feeding channel, and a switch plate for controlling the material flow (in a preferred embodiment, there are two: a first switch plate 36 and a second switch plate 37). The switch plate is assembled via a shaft 38, a bushing 39, and a hinge block 40 to achieve flexible rotation, opening and closing, and stable connection. The weighing system 9 uses pins 41 to precisely position the weighing module, and is equipped with sensor covers 42 and locking accessories 43 to protect the second stainless steel hopper 44 and the weighing components. The core weight signal is collected by the weighing sensor 48 installed on the sensor connection plate 30, and is driven by the fisheye bearing 49 to drive the switch plate to achieve force transmission and buffering. Finally, the cylinder 50 completes the precise control of the material unloading and opening / closing door.

[0055] The feeding device also includes a weighing hopper 46 and a dust removal device 47 for reducing dust overflow installed at the discharge end of the weighing hopper 46; the opening and closing action of the weighing hopper 46 is driven by an actuator installed on the cylinder support 45.

[0056] III. Conveying Device. This includes a bag opening clamping mechanism 5, a bag opening folding machine 6, a sewing mechanism (including a sewing machine head unit 7 and a thread holder 8), and a conveying mechanism 10. The bag opening clamping mechanism 5 is used to clamp the bag opening after rice filling; the bag opening folding machine 6 is used to neatly fold the edges of the bag opening; the sewing mechanism is used to sew and seal the neatly folded bag opening; and the conveying mechanism 10 is used to transport the finished rice package to the downstream process.

[0057] like Figure 4 As shown, the main body of the conveying device consists of a frame assembly 51, with casters and support feet 52 at the bottom for movement and horizontal adjustment. The frame integrates a vertical lifting mechanism 53 to accommodate packaging bags of different sizes and heights. A cable routing track assembly 54 on the left side is used for orderly cable and air pipe routing, improving the safety and cleanliness of the equipment. The sewing system is powered by a sewing drive motor 55 at the top, which transmits power to the sewing execution end via a sewing transmission mechanism 56. This transmission, combined with a thread tension frame 57, precisely controls the thread tension. Finally, the sealed rice bags are smoothly output under the guidance of the finished product conveying guide mechanism 58 at the bottom, thus achieving fully automated and intelligent packaging operations from bag feeding, metering, filling, sealing to conveying.

[0058] IV. Computer Vision Inspection Array. The computer vision inspection array includes several cameras, which are respectively installed in the bag feeding device, conveying device, and feeding device, for acquiring images during the rice packaging process. The images record the bag posture and bag opening status, as well as the flow pattern of rice during filling.

[0059] Furthermore, the computer vision detection array includes a first camera 13, a second camera 14, and a third camera 15, which are respectively positioned above the bag feeding path, in front of the bag opening area, and around the material dropping channel to acquire the center position of the bag. , bag opening angle With material flow area These key parameters are used to monitor and intelligently judge the bag supply status, bag opening posture, filling and sealing quality in real time, thereby improving the stability of the overall operation and packaging quality.

[0060] V. Main Electrical Control Box. The main electrical control box is used to control the operation of the bag feeding device, conveying device, and feeding device to complete the automated rice packaging.

[0061] To enhance the system's intelligence, this invention embeds a deep learning recognition network based on computer vision into key nodes of the entire packaging process. The visual algorithm detection and control flow shown in Figure 5 embodies this core logic. This deep learning recognition network performs semantic segmentation, key point extraction, and temporal change analysis on images acquired by the computer vision detection array to obtain state information. It also determines the bag's posture and opening state, as well as the flow pattern of rice during filling, based on this state information. If abnormalities are found, the main control box controls and adjusts the bag feeding device, conveying device, and feeding device, forming an integrated closed-loop control system combining vision and mechanics, ensuring stable and coordinated operation of each module.

[0062] The deep learning recognition network is built on an improved convolutional neural network architecture, including a semantic segmentation module, a key point extraction module, and a temporal change analysis module.

[0063] The semantic segmentation module uses a U-Net or DeepLabV3+ architecture to perform pixel-level classification on the input packaging bag image, segmenting the bag body region and the material dropping region. The material dropping velocity is obtained by calculating the rate of change of the pixel area of ​​the material dropping region between consecutive frames. , For material flow area;

[0064] The key point extraction module adopts an HRNet structure to locate the corner points and the center feature points of the bag opening. Compare it with the preset standard reference point Perform Euclidean distance comparison and calculate bag offset. ;

[0065] The time-series change analysis module employs a long short-term memory network to analyze the bag opening degree over continuous time steps. Regression analysis was performed to predict and determine whether the bag opening would rebound or close.

[0066] The method for determining whether the bag's posture and opening status, and the flow pattern of rice during filling are abnormal, based on status information, specifically includes:

[0067] Set the bag offset threshold, bag opening threshold, and material flow rate threshold;

[0068] When the bag body offset exceeds the bag body offset threshold, a signal is sent to the main electrical control box to control the bag feeding device to adjust the bag body posture.

[0069] When the opening of the packaging bag is less than the opening threshold, a signal is sent to the main electrical control box to control the conveying device to adjust the opening state of the packaging bag.

[0070] When the rice flow rate during filling is less than the flow rate threshold, a signal is sent to the main control box to control the feeding device to adjust the rice flow rate.

[0071] By leveraging machine vision for status monitoring and adaptive control, this invention upgrades the automated rice packaging process from structure-driven to perception-driven, significantly improving bagging stability, quantitative accuracy, and operational continuity. The bag feeding device uses visual recognition to ensure the bag's posture is corrected before entering the conveyor belt; the conveying device relies on visual feedback to ensure the bag opening angle meets feeding requirements; and the feeding device utilizes flow detection to ensure continuous and smooth material delivery. This invention ultimately constructs a rationally structured, highly autonomous, and high-speed-operation-suitable intelligent packaging equipment, providing a more efficient and reliable automation solution for the grain processing industry.

[0072] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure can be divided into different functional modules to complete all or part of the functions described above.

[0073] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0074] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. An automated rice packaging system based on machine vision intelligent monitoring, characterized in that, It includes a bag feeding device, a conveying device, a feeding device, a computer vision inspection array, and a main electrical control box; The bag feeding device includes a bag storage mechanism, a bag clamping mechanism, and a bag opening mechanism; the bag storage mechanism is used to store and sequentially distribute packaging bags. The bag clamping mechanism and the bag opening mechanism are connected in sequence to the bag storage mechanism to perform bag picking, clamping and bag opening operations; The conveying device includes a bag mouth clamping mechanism, a bag mouth folding machine, a sewing mechanism, and a conveying mechanism; the bag mouth clamping mechanism is used to clamp the bag mouth after the rice is filled; the bag mouth folding machine is used to neatly fold the edges of the bag mouth; the sewing mechanism is used to sew and seal the neatly folded bag mouth; the conveying mechanism is used to transport the finished rice package that has been sewn and sealed to the downstream process. The feeding device is used to measure the weight of the rice to be packaged and to fill the measured rice into the opened packaging bag. The computer vision detection array is used to collect images during the rice packaging process. The images record the bag posture and bag opening status, as well as the flow state of the rice during filling. The main electrical control box is used to control the operation of the bag feeding device, conveying device and feeding device to complete the automated rice packaging; The main electrical control box is equipped with a deep learning recognition network, which is used to perform semantic segmentation, key point extraction and temporal change analysis on the images acquired by the computer vision detection array to obtain state information; it is also used to determine whether the bag body posture and bag opening status of the packaging bag and the material flow state during rice filling are abnormal based on the state information. If abnormal, the main electrical control box controls and adjusts the bag feeding device, conveying device and feeding device.

2. The automated rice packaging system based on machine vision intelligent monitoring according to claim 1, characterized in that, The computer vision detection array includes several cameras, which are respectively installed in the bag feeding device, the conveying device, and the feeding device.

3. The automated rice packaging system based on machine vision intelligent monitoring according to claim 1, characterized in that, The deep learning recognition network is built on an improved convolutional neural network architecture, including a semantic segmentation module, a key point extraction module, and a temporal change analysis module. The semantic segmentation module uses a U-Net or DeepLabV3+ architecture to perform pixel-level classification on the input packaging bag image, segmenting the bag body region and the material dropping region. The material dropping velocity is obtained by calculating the rate of change of the pixel area of ​​the material dropping region between consecutive frames. ; The key point extraction module adopts an HRNet structure to locate the corner points and the center feature points of the bag opening. Compare it with the preset standard reference point Perform Euclidean distance comparison and calculate bag offset. ; The time-series change analysis module employs a long short-term memory network to analyze the bag opening degree over continuous time steps. Regression analysis was performed to predict and determine whether the bag opening would rebound or close. The method for determining whether the bag's posture and opening status, and the flow pattern of rice during filling are abnormal, based on status information, specifically includes: Set the bag offset threshold, bag opening threshold, and material flow rate threshold; When the bag body offset exceeds the bag body offset threshold, a signal is sent to the main electrical control box to control the bag feeding device to adjust the bag body posture. When the opening of the packaging bag is less than the opening threshold, a signal is sent to the main electrical control box to control the conveying device to adjust the opening state of the packaging bag. When the rice flow rate during filling is less than the flow rate threshold, a signal is sent to the main control box to control the feeding device to adjust the rice flow rate.

4. The automated rice packaging system based on machine vision intelligent monitoring according to claim 1, characterized in that, In the bag feeding device, the main body of the bag storage mechanism is the bag storage frame, and it is also equipped with a pallet moving mechanism, a bag dispensing mechanism, a bag feeding mechanism, a flipping bag loading mechanism, a bag pressing mechanism, and a bag picking mechanism for cooperative use. The pallet moving mechanism is used to carry and adjust the stacking position of the packaging bags; the bag separating mechanism is used to separate the packaging bags one by one; the bag feeding mechanism is used to directionally transport the separated packaging bags to the flipping bag loading mechanism; the flipping bag loading mechanism is used to adjust the packaging bags from a flat state to an upright state; the bag pressing mechanism and the bag picking mechanism are used to perform the adsorption and gripping of the packaging bags.

5. The automated rice packaging system based on machine vision intelligent monitoring according to claim 4, characterized in that, The bag feeding device also includes a bag limiting mechanism, a bag pressing adjustment mechanism, a packaging bag limiting plate mounting block, a packaging bag limiting plate, a bag storage air circuit box, and a vacuum generator. The bag limiting mechanism, the bag pressing adjustment mechanism, the packaging bag limiting plate mounting block, and the packaging bag limiting plate are used together to constrain and correct the position and posture of the packaging bag. The vacuum generator is used to provide negative pressure to the bag storage air circuit box; the bag storage air circuit box is used to manage the pneumatic control actions of the bag feeding device.

6. The automated rice packaging system based on machine vision intelligent monitoring according to claim 1, characterized in that, The feeding device is built on an independent frame and also includes a hopper and a weighing system; The hopper is suspended within the frame via a sensor connecting plate and a connecting rod, and is flexibly connected to the weighing system via a weighing connecting shaft; the feeding device's discharge control part includes a curved plate for forming a discharge channel and a switch plate for controlling the flow of materials. The switch plate is assembled via a shaft, bushing, and hinge block to achieve rotation and opening / closing. The weighing system uses pins to position the weighing module, and is equipped with sensor covers and locking accessories to protect the stainless steel hopper and weighing components. The weight signal is collected by the weighing sensor installed on the sensor connection plate, and the switch plate is driven by the fisheye bearing to realize force transmission and buffering. Finally, the cylinder controls the opening and closing of the unloading door.

7. An automated rice packaging system based on machine vision intelligent monitoring according to claim 6, characterized in that, The feeding device also includes a weighing hopper and a dust removal device for reducing dust overflow installed at the discharge end of the weighing hopper; the opening and closing action of the weighing hopper is driven by an actuator installed on a cylinder support.

8. The automated rice packaging system based on machine vision intelligent monitoring according to claim 1, characterized in that, The main body of the conveying device is a frame assembly with casters supporting the bottom, which can be moved and adjusted horizontally; the frame assembly integrates a vertical lifting mechanism to match packaging bags of different sizes and heights; the bottom of the frame assembly is also equipped with a finished product conveying guide mechanism to guide the finished rice packaging that has been sewn and sealed to the downstream process through the conveying mechanism.

9. The automated rice packaging system based on machine vision intelligent monitoring according to claim 1, characterized in that, The sewing mechanism includes a sewing machine head unit, a thread holder, a sewing drive motor, a sewing transmission mechanism, and a thread tension feeder. The thread placement rack, thread tension feeder, and sewing machine head unit are connected in sequence to control the thread tension and sew the opening of the packaging bag; the sewing drive motor and sewing transmission mechanism are used to provide sewing power.

10. An automated rice packaging system based on machine vision intelligent monitoring according to claim 1, characterized in that, It also includes a touch screen component connected to the main electrical control box for human-machine interaction and parameter setting.

Images