Clean packaging integrated device for black cabbage and processing method thereof

Abstract

This invention discloses an integrated clean packaging device for black cabbage, comprising an aluminum profile frame, and a cleaning device, a cutting device, a packaging device, an electrical control system, and a large turntable mounted on it. The cleaning device includes a water gun and an electric ball valve. The water gun is vertically downward-facing and aimed at the rotating station. During cleaning, the rotating station causes the black cabbage to rotate, achieving uniform rinsing. The cutting device includes a K230 camera, a shearing mechanism, a blade assembly drive control module, and a lead screw motor. The lead screw motor drives the shearing mechanism to the target position, and the blade assembly drive control module drives the blade assembly with grid-like fine teeth to open and close, performing a relative linear reciprocating motion to complete the cutting, achieving dynamic positioning of the cutting position. The packaging device includes a pen-shaped cylinder top, a vegetable basket, and a cross guide rail. The PLC coordinates the timing of each process, realizing integrated automatic operation of cleaning, cutting, and packaging, effectively improving production efficiency and reducing labor intensity and cutting loss rate.

Description

Technical Field

[0001] This invention relates to the fields of mechanical design and agricultural automation equipment, specifically to an integrated device for clean packaging of black cabbage and its processing method. Background Technology

[0002] Black tatsoi, a distinctive leafy vegetable, requires multiple post-harvest processing steps, including washing, cutting, and packaging. Existing leafy vegetable processing equipment is mostly single-function, with washing, cutting, and packaging equipment operating independently. This necessitates manual transfer of vegetables between machines, resulting in high labor intensity, low production efficiency, and a high risk of secondary contamination or mechanical damage during transport. Currently, there is a lack of integrated equipment that combines washing, cutting, and packaging processes into a single unit.

[0003] In terms of cleaning, existing cleaning equipment mostly adopts a slapping or multi-nozzle three-dimensional spray cleaning method. The vegetables are in a floating or static state during the cleaning process, lacking an effective positioning and rotation mechanism. Black tatsoi has thick leaves and strong soil adhesion on the stems. Static cleaning makes it difficult for the water flow to evenly wash both sides of the leaves and around the stems, resulting in incomplete removal of soil from the stems and leaf folds, and uneven cleaning effect.

[0004] In terms of cutting, existing cutting equipment typically uses a fixed-stroke reciprocating cutting method, with the cutting position preset and non-adjustable. However, black cabbage has an irregular shape and significant individual differences, and a fixed cutting position cannot adapt to the actual shape of each vegetable, easily resulting in over- or under-cutting and a high loss rate. Current technology lacks an intelligent cutting device that can dynamically determine the cutting position based on the actual shape of the vegetable. Summary of the Invention

[0005] This invention provides an integrated device for the clean packaging of black cabbage, which aims to automate the cleaning, cutting, and packaging processes of black cabbage.

[0006] To achieve the above objectives, the present invention proposes the following technical solution: an integrated clean packaging device for black cabbage, comprising an aluminum profile frame, characterized in that it includes a cleaning device, a cutting device, a packaging device, and an electrical control system disposed on the aluminum profile frame; it also includes a large turntable, on which a rotating station is provided;

[0007] The cleaning device includes a water gun and an electric ball valve. The water gun is fixed to the upper part of the aluminum profile frame and vertically downwards aligned with the rotating station. The electric ball valve is installed in the water inlet pipe of the water gun to control the water flow. During cleaning, the rotating station drives the black cabbage to rotate, so that the water flow washes the leaves and stems in sequence.

[0008] The cutting device includes a camera, a shearing mechanism, a blade assembly drive control module, and a lead screw motor. The camera is installed below the cutting station to capture images of the black cabbage and locate the cutting position coordinates. The shearing mechanism is mounted on the lead screw motor, which drives the shearing mechanism to move horizontally to the target cutting position. The shearing mechanism includes a set of blades with grid-like fine teeth. The blade assembly drive control module controls the opening and closing of the blade assembly with grid-like fine teeth to perform relative linear reciprocating motion to complete the cutting.

[0009] The packaging device includes a pen-shaped cylinder top, a basket, and a cross guide rail, with the basket placed on the cross guide rail;

[0010] The electrical control system includes a PLC, which is connected to the electric ball valve, the blade assembly drive control module, the lead screw motor, and the camera, respectively, and is used to coordinate the timing of the cleaning, cutting, and packaging processes.

[0011] Furthermore, in this invention, the cleaning device also includes a waterproof shell, which is made of transparent acrylic sheet and covers the entire cleaning area. The front of the waterproof shell is provided with an openable observation window, and the bottom is provided with a drain outlet.

[0012] Furthermore, in this invention, the water gun is connected to a water source via a water pipe, and the tail of the water gun is equipped with a manual adjustment knob for switching between three water spray patterns: columnar, fan-shaped, and mist-shaped.

[0013] Furthermore, in this invention, the large turntable is mounted on the aluminum profile frame via a hollow rotating platform, and a servo motor drives the hollow rotating platform to rotate the large turntable; the rotating station is driven to rotate by a 42CM06 motor via gears.

[0014] Furthermore, in this invention, the camera is connected to the PLC via an Ethernet interface, and a deep learning-based target detection model is used to identify the contour features of the black cabbage and locate the coordinates of key points at the junction of the stem and leaves; the PLC controls the lead screw motor to drive the shearing mechanism to move horizontally to the target cutting position according to the coordinates.

[0015] Furthermore, in this invention, the cutting device also includes a fixing device, which is disposed at the cutting station and is used to fix the black cabbage during cutting.

[0016] Furthermore, in this invention, the cross guide rail is a two-degree-of-freedom cross slide structure; the pen-shaped cylinder top body is controlled by the PLC and is used to feed the cut black cabbage into the vegetable basket.

[0017] The processing method using the above-mentioned integrated clean packaging device for black cabbage is characterized by the following steps:

[0018] S1. Loading: Place the black cabbage on the rotating station of the large turntable;

[0019] S2. Cleaning: The PLC controls the large turntable to rotate, transporting the black cabbage to the cleaning station. It controls the electric ball valve to open so that the water gun sprays water. At the same time, the rotating station drives the black cabbage to rotate, so that the water flow washes the leaves and stems in sequence. After cleaning, the electric ball valve is closed.

[0020] S3. Cutting: The large turntable continues to rotate, transporting the black cabbage to the cutting station. The camera captures images of the black cabbage and identifies the cutting position coordinates at the junction of the stem and leaves. After receiving the coordinates, the PLC controls the lead screw motor to drive the shearing mechanism to move to the target cutting position. The blade group drive control module drives the grid-like fine tooth blade group of the shearing mechanism to open and close and make relative linear reciprocating motion to complete the cutting. After the cutting is completed, the mechanism is reset.

[0021] S4. Packaging: The PLC controls the pen-shaped cylinder top to feed the cut black cabbage into the vegetable basket on the cross guide rail, completing the packaging.

[0022] Furthermore, in this invention, in step S2, before cleaning, the PLC controls the camera to pre-scan and capture images of the black cabbage on the rotating station, acquiring images of the cabbage surface and calculating a contamination coverage rate index; the PLC adaptively adjusts the cleaning parameters based on the contamination coverage rate index, the cleaning parameters including the cleaning duration and the rotation speed of the rotating station; wherein, the contamination coverage rate index is obtained by extracting the area ratio of soil color gamut pixels to the total outline pixels of the cabbage after color space conversion of the pre-scanned image; when the contamination coverage rate index is greater than a preset heavy contamination threshold, the PLC sets the cleaning duration to an upper limit and increases the rotation speed of the rotating station; when the contamination coverage rate index is less than a preset light contamination threshold, the PLC sets the cleaning duration to a lower limit and decreases the rotation speed of the rotating station; when the contamination coverage rate index is between the light contamination threshold and the heavy contamination threshold, the PLC performs linear interpolation calculation between the upper and lower limits of the cleaning duration based on the contamination coverage rate index to determine the cleaning duration.

[0023] Furthermore, in this invention, in step S3, after the camera acquires a bottom-view image from below the cutting station, the PLC performs an affine transformation mapping from the pixel coordinate system to the physical coordinate system on the image coordinates. Specifically, this includes: first, performing radial distortion correction on the original image to eliminate barrel distortion caused by the lens; then, using a pre-calibrated affine transformation matrix to map the distortion-corrected pixel coordinates to physical coordinates in the ball screw motor travel coordinate system; the affine transformation matrix is ​​obtained by placing a calibration plate with a known spacing at the cutting station, acquiring an image of the calibration plate, and then performing least-squares fitting to solve for the pixel coordinates of the corner points of the calibration plate and their corresponding physical coordinates.

[0024] In step S4, the PLC calculates the optimal placement coordinates of the current vegetable in the basket based on the outer rectangle size of the current black cabbage outline obtained by the camera in the cutting step, combined with the known internal dimensions of the basket and the cumulative occupied area of ​​the already loaded vegetables, using a greedy rectangular packing strategy. The PLC controls the cross guide rail to drive the basket to move according to the optimal placement coordinates, so that the target placement position of the basket is aligned with the unloading point before performing the unloading action. The greedy rectangular packing strategy is as follows: vegetables are arranged sequentially along the long side of the basket. When the remaining width of the current row is insufficient to accommodate the width of the current vegetable, the arrangement is changed to a new row. When changing rows, the row height is taken as the maximum length value of the vegetables already arranged in that row.

[0025] Furthermore, in this invention, the large turntable is equipped with at least two rotating stations. The PLC uses a pipeline parallel scheduling algorithm to control the black cabbage at each rotating station to operate in parallel between the washing, cutting, and packaging stations. The pipeline parallel scheduling algorithm includes: the PLC maintaining a process status identifier for each rotating station, the process status identifier including idle, washing in progress, washing completed and awaiting transfer, cutting in progress, cutting completed and awaiting transfer, and packaging in progress; within each scheduling cycle, the PLC traverses the process status identifiers of all rotating stations to determine whether there is a transfer conflict between the stations; when there is no transfer conflict, the PLC controls the large turntable to rotate by one station spacing angle, so that each rotating station synchronously enters the next process; when there is a transfer conflict, the PLC delays the rotation of the turntable until the current process of the conflicting station is completed; the scheduling cycle is equal to the maximum value among the washing duration, cutting duration, and packaging duration.

[0026] Beneficial effects: The technical solution of this application has the following technical effects:

[0027] This invention integrates a cleaning device, a cutting device, and a packaging device onto an aluminum profile frame, and uses a PLC to coordinate the timing of each process. This integrates the three processes of cleaning, cutting, and packaging into a single device, achieving continuous, integrated operation. The black cabbage is automatically transferred between workstations via a large turntable, eliminating the need for manual transfer, effectively reducing labor intensity, improving production efficiency, and avoiding secondary contamination and mechanical damage caused by inter-process transfer.

[0028] The cleaning device of this invention drives the black cabbage to rotate during the cleaning process by rotating the workstation. Combined with the water gun that sprays vertically downwards, the water flow can sequentially wash the front and back of the leaves and the area around the stem, which makes up for the coverage limitation of spraying in one direction. It effectively solves the problem of incomplete removal of dirt from the stem and leaf folds of the black cabbage, and improves the uniformity and cleanliness of the cleaning.

[0029] The cutting device of the present invention acquires images of black cabbage through a camera and locates the cutting position coordinates at the junction of the stem and leaves. The PLC controls the lead screw motor to drive the shearing mechanism to move to the target cutting position to complete the cutting, thereby realizing dynamic positioning of the cutting position. It can adapt to the individual morphological differences of black cabbage and effectively reduce the loss caused by improper cutting position.

[0030] In summary, this invention does not simply mechanically superimpose the three functional modules of cleaning, cutting, and packaging. Instead, it addresses the actual processing characteristics of black cabbage, such as "leaf drooping and unfolding, more yellow or mixed leaves on the outside, relatively intact leaves inside, and easy accumulation of mud in the stem and leaf folds," and forms a collaborative processing scheme with a large turntable as the core of the process flow, a rotating workstation as the core of the self-rotation and positioning of a single black cabbage, and a PLC as the core of the timing control. Specifically, the large turntable is used to accurately transport the black cabbage sequentially to the washing, cutting, and packaging stations, solving the problems of manual transfer, low efficiency, easy contamination, and easy damage that are inherent in traditional decentralized equipment. The rotating station is equivalent to a small turntable-type self-rotating station. At the washing station, the fixed downward spray of water can sequentially cover the front and back of the black cabbage leaves, the leaf edges, and the circumference of the stem, avoiding the problem of local dead corners in static washing. At the cutting station, it can work with the camera to obtain more complete peripheral image information, enabling visual recognition to more accurately capture the outer area to be cut and the junction of stem and leaf. The cutting mechanism only needs to perform radial compensation positioning under the action of the lead screw motor, without the need for a large-scale cross slide or complex multi-axis cutter mechanism. It can complete 360-degree continuous cutting of the outer area as the rotating station rotates. Thus, the large turntable connects the three work areas, while the small turntable rotating stations amplify the functions of cleaning coverage, visual recognition, and shearing positioning. The PLC then coordinates the timing of actions of the actuators, such as electric ball valves, cameras, lead screw motors, blade assembly drive control modules, and pen-shaped cylinder tops, creating a closed-loop collaborative relationship between the functional modules: "cleaning and purification improves the accuracy of visual recognition, visual positioning improves the accuracy of shearing, precise shearing reduces the workload of subsequent rejection and packaging, and automatic unloading and packaging reduces damage from manual handling."

[0031] Practical application verification shows that, compared to individual module operation or simple series connection, this invention can reduce the processing time of a single black cabbage plant from approximately 38 seconds to approximately 25.5 seconds, increasing processing efficiency by approximately 32.9%; the overall processing qualification rate increases from approximately 85.2% to approximately 98.5%, and the positioning error can be controlled within ±0.5mm; by eliminating the need for additional transfer mechanisms, large-scale multi-axis shearing mechanisms, and independent rejection mechanisms, the overall machine cost can be controlled within 10,000 yuan, a cost reduction of approximately 28.6%, and processing loss can be reduced to approximately 1.5%; at the same time, the rotation combined with precise spraying reduces water consumption by approximately 20%, and the PLC closed-loop timing control significantly improves the overall machine operation stability. This technical solution is particularly suitable for vegetable cooperatives, clean vegetable processing centers, community fresh food delivery stations, pre-processing workshops for agricultural products, and primary processing scenarios in small and medium-sized black cabbage production areas. It can achieve integrated continuous operation of black cabbage cleaning, outer yellow leaf trimming, and packaging at a relatively low equipment cost. This not only improves the standardization of clean vegetable processing and product consistency, but also reduces reliance on manual labor and raw material loss. It has high industrialization promotion value and market application prospects.

[0032] It should be understood that all combinations of the foregoing concepts and the additional concepts described in more detail below can be considered part of the inventive subject matter of this disclosure, provided that such concepts do not contradict each other.

[0033] The foregoing and other aspects, embodiments, and features of the teachings of the present invention will be more fully understood from the following description in conjunction with the accompanying drawings. Other additional aspects of the invention, such as features and / or beneficial effects of exemplary embodiments, will become apparent from the following description or may be learned through practice of specific embodiments according to the teachings of the present invention. Attached Figure Description

[0034] The accompanying drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component shown in the various figures may be denoted by the same reference numeral. For clarity, not every component is labeled in each figure. Embodiments of various aspects of the invention will now be described by way of example and with reference to the accompanying drawings, wherein:

[0035] Figure 1 This is a schematic diagram of the structure of the present invention;

[0036] Figure 2 This is a schematic diagram of the structure of the present invention.

[0037] Figure 3 This is a schematic diagram of the shearing mechanism of the present invention.

[0038] The meanings of the labels in the attached figures are as follows: 1. Water gun; 2. Electric ball valve; 3. Aluminum profile frame; 4. Water pipe; 5. Shearing mechanism; 6. Blade assembly drive control module; 7. Waterproof shell; 8. Pen-shaped cylinder top body; 9. Large turntable; 10. Fixing device; 11. Rotary station; 12. Vegetable basket; 13. Cross guide rail; 14. Hollow rotating platform; 15. Servo motor; 16. 42CM06 motor; 17. Gear; 18. PLC; 19. Lead screw motor; 20. Camera. Detailed Implementation

[0039] The embodiments of the invention are described in detail below with reference to the accompanying drawings to clearly illustrate the structure, purpose, advantages, positional relationships, and connection methods of each component. It should be noted that the directional indications (such as "front," "back," "up," and "down") involved in this embodiment are based on the posture shown in the drawings and are only used to describe the relative positional relationships and movement of the components. If the posture changes, the directional indications will be adjusted accordingly. The term "connection" includes mechanical connections and electrical connections, and can be fixed connections, detachable connections, or indirect connections through an intermediate medium. The specific meaning is understood by those skilled in the art based on the context.

[0040] Example 1

[0041] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings. Figures 1 to 3 As shown, this embodiment provides an integrated clean packaging device for black cabbage, including an aluminum profile frame 3, and a cleaning device, a cutting device, a packaging device, and an electrical control system mounted on the aluminum profile frame 3. It also includes a large turntable 9 with a rotating workstation 11. The aluminum profile frame 3 serves as the load-bearing foundation of the entire machine, constructed from aluminum alloy profiles, with overall dimensions of 1000mm in length, 1000mm in width, and 1000mm in height. It provides a stable installation benchmark and structural support for each functional module. Furthermore, the aluminum alloy profiles offer advantages such as light weight, corrosion resistance, and ease of disassembly and adjustment, making them suitable for the humid and watery conditions encountered in vegetable processing environments.

[0042] The large turntable 9 is mounted in the middle of the aluminum profile frame 3 via a hollow rotating platform 14. A servo motor 15 drives the hollow rotating platform 14 to intermittently rotate the large turntable 9 around its central axis. As the core conveying mechanism for material flow between workstations, the large turntable 9 sequentially transports the black cabbage from the loading position to the washing, cutting, and packaging workstations, achieving automatic connection between processes and eliminating manual transfer. The hollow rotating platform 14 has a hollow structure, facilitating the passage of pipes and cables through the center and preventing entanglement during rotation. The servo motor 15 features high positioning accuracy and fast response speed, precisely controlling the rotation angle and stopping position of the large turntable 9 to ensure the black cabbage accurately reaches each functional workstation.

[0043] The rotating station 11 is installed on the upper surface of the large turntable 9 to support the black cabbage and drive it to rotate around its own axis. The rotating station 11 is driven by a 42CM06 motor 16 via gear 17. The 42CM06 motor 16 is fixed below the large turntable 9, and its output shaft meshes with the transmission gear of the rotating station 11 through gear 17, transmitting power to the rotating station 11. The rotation function of the rotating station 11 is key to achieving uniform cleaning. When the black cabbage is placed on the rotating station 11, the station slowly rotates it, allowing the vertically downward water flow from the water gun 1 to sequentially wash the front and back of the cabbage leaves and the surrounding area of ​​the stem, thus compensating for the limited coverage of the fixed-direction water jet. The gear 17 transmission method is compact and smooth, suitable for achieving rotational drive in a limited space. The surface of the rotating station 11 can be textured with anti-slip material to prevent the black cabbage from slipping during rotation.

[0044] The cleaning device is positioned at the corresponding cleaning station on the large turntable 9, and includes a water gun 1, an electric ball valve 2, a water pipe 4, and a waterproof housing 7. The water gun 1 is fixed to the upper crossbeam of the aluminum profile frame 3 and vertically downwards, aligned with the center of the rotating station 11, allowing water to flow vertically from directly above and wash the black cabbage placed on the rotating station 11. The water gun 1 is an adjustable high-pressure water gun with a manual adjustment knob at its tail. The operator can switch between three water spray patterns—columnar, fan-shaped, and mist-shaped—depending on the degree of soil adhesion and leaf characteristics of the black cabbage. The columnar water spray has a strong impact, suitable for removing stubborn soil from the stems; the fan-shaped water spray covers a larger area, suitable for cleaning the leaf surface; and the mist-shaped water spray is gentler, suitable for a final fine rinse, avoiding damage to the leaves. The manual knob adjustment method is intuitive and simple to operate, requiring no complex electronic control parameter settings, thus lowering the operating threshold.

[0045] Water gun 1 is connected to an external water source via water pipe 4, and an electric ball valve 2 is connected in series in water pipe 4. The electric ball valve 2 is controlled by PLC 18 to open and close, thus achieving automatic control of water flow. When PLC 18 sends an open signal, the electric ball valve 2 opens, and water flows through water pipe 4 into water gun 1 for cleaning; when the cleaning time reaches the preset time, PLC 18 sends a close signal, the electric ball valve 2 closes, and the water flow is interrupted. The electric ball valve 2 automates the cleaning process, eliminating the need for manual water supply switching and ensuring consistent cleaning time.

[0046] The waterproof shell 7 is made of transparent acrylic sheet, completely covering the cleaning area and fixed to the aluminum profile frame 3. The main function of the waterproof shell 7 is to prevent water from splashing during cleaning, keeping the working environment clean, and protecting surrounding electrical components and control circuits from water corrosion. The transparent material of the waterproof shell 7 allows the operator to easily observe the cleaning process and assess the cleaning effect. The front of the waterproof shell 7 has an openable observation window, through which the operator can place or remove food items, and also open the window to check the cleaning progress during the process. The bottom of the waterproof shell 7 has a drain outlet, through which wastewater generated during cleaning is collected and discharged. A drain pipe can be connected to guide the wastewater to a wastewater collection container for subsequent treatment.

[0047] The working principle of the cleaning device is as follows: After the black cabbage is placed in the center of the rotating station 11, the large turntable 9 rotates and transports it to the cleaning station. The PLC 18 controls the electric ball valve 2 to open, and water flows through the water pipe 4 and sprays vertically downwards from the water gun 1 onto the surface of the black cabbage. At the same time, the PLC 18 controls the 42CM06 motor 16 to drive the rotating station 11 to rotate through the gear 17. The black cabbage rotates slowly with the rotating station 11, and the vertical water flow washes away the dirt and impurities on all sides of the leaves and stems. After cleaning for 10-15 seconds, the PLC 18 controls the electric ball valve 2 to close, and the cleaning process is completed. Through the combination of the rotation of the rotating station 11 and the fixed vertical water flow, a relatively uniform cleaning coverage effect is achieved under the premise of simple structure and controllable cost.

[0048] The cutting device is positioned at the corresponding cutting station on the large turntable 9, and includes a K230 camera 20, a shearing mechanism 5, a blade assembly drive control module 6, a fixing device 10, and a lead screw motor 19. The K230 camera 20 is installed directly below the cutting station with its lens facing upwards, used to capture a bottom-view image of the black tartare. The K230 camera 20 is connected to the PLC 18 via an Ethernet interface for real-time transmission of image data. The design intent of installing the K230 camera 20 below the cutting station is that shooting from a bottom-view angle allows for a clear observation of the boundary area between the black tartare stem and leaves. Compared to shooting from above, the bottom-view angle is not affected by leaf obstruction, which is beneficial for accurately identifying the cutting target point.

[0049] The K230 camera 20 uses a deep learning-based object detection model to process the acquired images. This model can identify the contour features of black tartary cabbage and automatically locate the coordinates of key points at the junction of the stem and leaves. In this embodiment, the YOLOv8 object detection model is preferred. This model performs excellently in terms of both detection speed and accuracy, with a single image processing time of approximately 50ms, which meets the requirements of real-time processing. The model has been pre-trained with a large number of black tartary cabbage image samples collected in the field, achieving a recognition accuracy of 96.8%. It can adapt to the individual morphological differences of black tartary cabbage and dynamically determine the optimal cutting position for each plant. After recognition, the key point coordinates are transmitted to the PLC 18 via Ethernet.

[0050] The shearing mechanism 5 is mounted on the slider of the lead screw motor 19, which is fixed to the aluminum profile frame 3. Its slider can move linearly in a single horizontal direction, thereby driving the shearing mechanism 5 to move back and forth to the target cutting position. The lead screw motor 19 is controlled and driven by a PLC 18. When the PLC 18 receives the target cutting position coordinates sent by the K230 camera 20, it calculates the distance and direction the lead screw motor 19 needs to move, and controls the lead screw motor 19 to drive the shearing mechanism 5 to move precisely to the target cutting position. The lead screw motor 19 uses lead screw transmission, which has the advantages of high positioning accuracy and smooth operation. The positioning accuracy can reach ±0.1mm, ensuring that the shearing mechanism 5 accurately reaches the target position.

[0051] The shearing mechanism 5 is a fine-tooth grid-type electric shear, comprising a pair of oppositely arranged blades. Each blade has fine teeth arranged in a grid pattern on its cutting edge. The fine teeth of the two blades are staggered along the cutting edge direction, forming a blade assembly with grid-like fine teeth. The shearing mechanism 5 also includes a cylinder gripper, a hollow cup motor, an eccentric shaft, and a linear slide structure. The cylinder gripper drives the two blades to open and close relative to each other. The hollow cup motor acts as the prime mover, driving the eccentric shaft to rotate. The eccentric shaft converts the rotational motion into linear motion, which is limited by the linear slide structure, causing the blade assembly with grid-like fine teeth to perform rapid relative linear reciprocating motion along the cutting edge direction. The opening and closing action and relative linear reciprocating motion of the blade assembly with grid-like fine teeth are controlled by the blade assembly drive control module 6. The blade assembly drive control module 6 is connected to the output of the PLC 18. When the shearing mechanism 5 reaches the target cutting position, the PLC 18 sends a signal to the blade assembly drive control module 6. The blade assembly drive control module 6 drives the cylinder gripper to clamp and open the two blades at a macroscopic level to shear the stem of the black cabbage. Simultaneously, it drives the hollow cup motor to make the grid-like fine-toothed blade assembly perform rapid relative linear reciprocating motion at a microscopic level, repeatedly crushing the strong and robust veins of the black cabbage. The simultaneous action of macroscopic shearing and microscopic crushing avoids the situation of tangled threads or incomplete cuts, ensuring high cutting accuracy and high cutting efficiency. After cutting, the blade assembly drive control module 6 controls the shearing mechanism 5 to open and reset, and the lead screw motor 19 drives the shearing mechanism 5 back to its initial position, ready for the next cut. The fixing device 10 is installed at the cutting station to fix the black cabbage during the cutting process, preventing displacement or shaking and ensuring cutting accuracy and cut quality.

[0052] The cutting device works by combining visual recognition technology with an automated actuator. First, a K230 camera 20 captures real-time images of the black cabbage. A deep learning target detection model automatically identifies the stem-leaf junction and outputs the cutting coordinates. Then, the PLC 18 controls the lead screw motor 19 based on these coordinates for precise linear positioning, moving the shearing mechanism 5 to the target position. Finally, the blade assembly drive control module 6 drives the grid-like fine-toothed blade assembly to open and close, performing relative linear reciprocating motion to complete the cutting. This solution achieves dynamic positioning and precise control of the cutting position, adapting to individual morphological differences in black cabbage and effectively reducing cutting loss.

[0053] The packaging device is located downstream of the cutting station and includes a pen-shaped cylinder top body 8, a vegetable basket 12, and a cross guide rail 13. The pen-shaped cylinder top body 8 is installed near the cutting station and its extension and retraction are controlled by a PLC 18. After cutting, the PLC 18 controls the pen-shaped cylinder top body 8 to extend, pushing the cut black cabbage out of the cutting station and discharging it into the vegetable basket 12 below. The pen-shaped cylinder top body 8 has a short stroke and fast action, making it suitable for simple pushing and discharging actions.

[0054] The vegetable basket 12 is placed on the cross guide rail 13. The cross guide rail 13 is a two-degree-of-freedom cross slide structure, driven by two servo motors for linear motion in the X and Y axes respectively, enabling two-dimensional positioning in the horizontal plane. The function of the cross guide rail 13 is to transport the vegetable basket 12 containing black cabbage from the receiving position to the packaging machine or the discharging position, realizing automatic connection between the cutting and packaging processes. The two-degree-of-freedom design of the cross guide rail 13 allows the vegetable basket 12 to move flexibly to any required planar position, adapting to different packaging layout requirements.

[0055] The packaging device works as follows: After cutting, the pen-shaped cylinder top 8 pushes the black cabbage out of the packaging and into the vegetable basket 12 on the cross guide rail 13; the cross guide rail 13 then transports the vegetable basket 12 to the packaging position to complete the packaging process. Through the coordinated operation of the pen-shaped cylinder top 8 and the cross guide rail 13, the automatic flow from cutting to packaging is achieved without manual intervention.

[0056] The electrical control system, including the PLC18, is the core that coordinates the timing of actions of all devices in the machine. The PLC18 is installed in the lower part of the aluminum profile frame 3 and works in conjunction with the touchscreen, which serves as the human-machine interface. Operators can start the equipment, set parameters, and monitor the operating status via the touchscreen. The digital output terminals of the PLC18 are connected to the electric ball valve 2 of the cleaning device and the blade assembly drive control module 6 of the cutting device, respectively, to control the water flow and the opening and closing of the grid-like fine-tooth blade assembly and its relative linear reciprocating motion. The pulse output terminals are connected to the drive motors of the servo motor 15, the 42CM06 motor 16, the lead screw motor 19, and the cross guide rail 13, respectively, to control the movement of each motion mechanism. The PLC18 is connected to the K230 camera 20 via an Ethernet interface to receive coordinate data output from the vision recognition unit. The PLC18 internally stores a pre-programmed control program, which, according to a preset workflow and timing logic, sequentially controls the automatic execution of each process—feeding, cleaning, cutting, and packaging—to achieve integrated automatic operation of cleaning, cutting, and packaging.

[0057] The overall working principle of this device is as follows: The operator places the black cabbage to be processed on the rotating station 11 of the large turntable 9. After starting the equipment via the touch screen, the PLC 18 automatically controls the execution of all processes according to the preset program. First, the servo motor 15 drives the large turntable 9 to rotate, conveying the black cabbage to the washing station. The PLC 18 controls the electric ball valve 2 to open, and the 42CM06 motor 16 drives the rotating station 11 to rotate. The water jet from the water gun 1 sprays water vertically in conjunction with the rotation of the black cabbage to complete the washing. Then, the large turntable 9 continues to rotate, conveying the black cabbage to the cutting station. The K230 camera 20 collects images and identifies the cutting coordinates. The PLC 18 controls the lead screw motor 19 to position the cuttings, and the blade group drive control module 6 drives the shearing mechanism 5 to complete the cutting. Finally, the pen-shaped cylinder top body 8 discharges the cut black cabbage into the vegetable basket 12, and the cross guide rail 13 conveys the vegetable basket 12 to the packaging position to complete the unloading. Throughout the entire process, the PLC 18 coordinates the timing of the actions of each device, realizing integrated continuous automatic operation of washing, cutting, and packaging.

[0058] In summary, this invention does not simply mechanically superimpose the three functional modules of cleaning, cutting, and packaging. Instead, it addresses the actual processing characteristics of black cabbage, such as "leaf drooping and unfolding, more yellow or mixed leaves on the outside, relatively intact leaves inside, and easy accumulation of mud in the stem and leaf folds," by forming a collaborative processing scheme with a large turntable 9 as the core of the process flow, a rotating station 11 as the core of the self-rotation and positioning of a single black cabbage plant, and a PLC 18 as the core of the timing control. Specifically, the large turntable 9 is used to accurately transport the black cabbage to the washing, cutting, and packaging stations in sequence, solving the problems of manual transfer, low efficiency, easy contamination, and easy damage in traditional decentralized equipment. The rotating station 11 is equivalent to a small turntable-type self-rotating station. At the washing station, the fixed downward spray of water can sequentially cover the front and back of the black cabbage leaves, the leaf edges, and the circumferential position of the stem, avoiding the problem of local dead corners in static washing. At the cutting station, it can cooperate with the camera 20 to obtain more complete peripheral image information, enabling visual recognition to more accurately capture the outer area to be cut and the junction of stem and leaf. The cutting mechanism 5 only needs to perform radial compensation positioning under the action of the lead screw motor 19, without the need to use a large-scale cross slide or complex multi-axis cutter mechanism. It can complete the 360-degree continuous cutting of the outer area with the rotation of the rotating station 11.

[0059] Thus, the large turntable 9 is responsible for connecting the three work areas, while the small turntable rotary workstations 11 amplify the functions of cleaning coverage, visual recognition, and shearing positioning. The PLC 18 then coordinates the timing of actions of the actuators such as the electric ball valve 2, camera 20, lead screw motor 19, blade group drive control module 6, and pen-shaped cylinder top body 8, so that a closed-loop collaborative relationship is formed between the functional modules: "cleaning and purification improves the accuracy of visual recognition, visual positioning improves the accuracy of shearing, precise shearing reduces the workload of subsequent rejection and packaging, and automatic unloading and packaging reduces damage from manual handling."

[0060] Practical application verification shows that, compared to individual module operation or simple series connection, this invention can reduce the processing time of a single black cabbage plant from approximately 38 seconds to approximately 25.5 seconds, increasing processing efficiency by approximately 32.9%; the overall processing qualification rate increases from approximately 85.2% to approximately 98.5%, and the positioning error can be controlled within ±0.5mm; by eliminating the need for additional transfer mechanisms, large-scale multi-axis shearing mechanisms, and independent rejection mechanisms, the overall machine cost can be controlled within 10,000 yuan, a cost reduction of approximately 28.6%, and processing loss can be reduced to approximately 1.5%; at the same time, the rotation combined with precise spraying reduces water consumption by approximately 20%, and the PLC closed-loop timing control significantly improves the overall machine operation stability. This technical solution is particularly suitable for vegetable cooperatives, clean vegetable processing centers, community fresh food delivery stations, pre-processing workshops for agricultural products, and primary processing scenarios in small and medium-sized black cabbage production areas. It can achieve integrated continuous operation of black cabbage cleaning, outer yellow leaf trimming, and packaging at a relatively low equipment cost. This not only improves the standardization of clean vegetable processing and product consistency, but also reduces reliance on manual labor and raw material loss. It has high industrialization promotion value and market application prospects.

[0061] Example 2

[0062] This embodiment, based on the integrated clean packaging device and processing method for black cabbage described in Embodiment 1, further details the specific implementation methods of the washing adaptive control algorithm, cutting coordinate transformation algorithm, intelligent packing algorithm, and pipeline parallel scheduling algorithm. All of the following algorithms are executed by the internal program of PLC18 or the embedded processing unit of K230 camera 20.

[0063] The adaptive control algorithm for cleaning involves the PLC18 controlling the large turntable 9 to rotate the rotating station 11 carrying the black cabbage to the acquisition area of ​​the K230 camera 20 before the cleaning process in step S2. The K230 camera 20 then pre-scans and captures a color image of the cabbage surface. This pre-scanning step reuses the camera hardware from the cutting station. By adjusting the rotation timing of the large turntable 9, the camera can be reused before cleaning without adding additional sensor hardware, demonstrating the coordinated design of the device structure and algorithm.

[0064] After acquiring the pre-scanned image, it is first converted from the RGB color space to the HSV color space to facilitate the separation of hue, saturation, and brightness information. In the HSV space, the hue values ​​of the soil region are concentrated in the yellowish-brown range, which has a significant hue difference from the green of the black tatsoi leaves and the light green of the stems. Therefore, the soil region can be segmented and extracted by setting a hue threshold.

[0065] The extraction criteria for pixels in the soil region are: the hue channel value must be between the preset lower and upper boundaries of the soil hue, and the saturation channel value must be greater than the preset saturation threshold. Pixels meeting these criteria are marked as soil pixels, and the total number of soil pixels is counted. Simultaneously, the total number of pixels in the outline region of the black tumbleweed is extracted by performing foreground-background segmentation on the image. The pollution coverage rate is calculated using the following formula:

[0066]

[0067] in, Pollution coverage is expressed as a percentage. This refers to the total number of soil pixels, which is the number of pixels in the image that meet the soil color gamut filtering criteria. This represents the total number of pixels in the outline region of the vegetable body, i.e., the total number of pixels belonging to the black cabbage body after foreground segmentation.

[0068] PLC18 based on pollution coverage The value adaptively adjusts the cleaning parameters. A light contamination threshold is set. and heavy pollution threshold In this embodiment Take 15%, Take 45%. Cleaning duration. and the rotation speed of the rotary workstation The rules for determining it are as follows:

[0069] When the contamination coverage exceeds the heavy contamination threshold, the cleaning duration and the rotation speed of the rotary station are both set to their maximum values.

[0070]

[0071] in, The maximum cleaning duration is set to 15 seconds in this embodiment; The upper limit of the rotation speed of the rotary workstation is 60 revolutions per minute in this embodiment.

[0072] When the contamination coverage is less than the light contamination threshold, the cleaning duration and the rotation speed of the rotary station are both set to the lower limit.

[0073]

[0074] in, The minimum cleaning duration is set to 8 seconds in this embodiment; The lower limit of the rotation speed of the rotary workstation is 30 revolutions per minute in this embodiment.

[0075] When the contamination coverage falls between the light and heavy contamination thresholds, the cleaning duration is calculated using linear interpolation:

[0076]

[0077] in, The cleaning duration is obtained through linear interpolation. The pollution coverage calculated for the current image; The threshold for mild pollution; The threshold for severe pollution; This represents the lower limit of the cleaning duration. This represents the upper limit of the cleaning duration.

[0078] The corresponding rotational speed of the rotary workstation is also calculated using linear interpolation:

[0079]

[0080] in, The rotational speed of the rotary workstation is obtained through linear interpolation. This is the lower limit of the rotation speed of the rotary workstation; This represents the upper limit of the rotation speed of the rotary workstation; the meanings of the other symbols are the same as above.

[0081] The technical advantages of this algorithm are as follows: by reusing the camera at the cutting station for contamination assessment before cleaning, it achieves deep coupling between the device hardware structure and the control algorithm, enabling the cleaning process to have adaptive adjustment capabilities without adding additional sensors. Compared with fixed parameter cleaning, it reduces unnecessary cleaning time and water consumption for lightly contaminated vegetables, enhances the cleaning intensity for heavily contaminated vegetables, and increases the overall cleaning pass rate to over 96%, while saving approximately 20% of water.

[0082] In the cutting coordinate transformation algorithm, during the cutting process in step S3, the K230 camera 20 captures a bottom-view image of the black cabbage from directly below the cutting station, with the target position in the image represented by pixel coordinates. However, when the PLC 18 controls the lead screw motor 19 to perform positioning actions, it requires the travel value in the physical coordinate system. Therefore, it is necessary to map the image pixel coordinates to the physical travel coordinates of the lead screw motor 19. Due to the radial distortion of the K230 camera 20 lens, directly using the original pixel coordinates for mapping will result in positioning deviation; therefore, distortion correction is required first.

[0083] The radial distortion correction model is as follows. Let the ideal distortion-free normalized coordinates of a point in the image be the projection position of that point under the ideal pinhole model, and its radial distance squared from the optical center be:

[0084]

[0085] in, This is the normalized radial distance from the point to the optical center of the image; The normalized x-coordinate of the point is equal to the pixel x-coordinate of the point minus the x-coordinate of the optical center pixel, divided by the number of pixels at the horizontal focal length. The normalized ordinate of the point is equal to the pixel ordinate of the point minus the ordinate of the optical center pixel, divided by the number of pixels at the vertical focal length.

[0086] Due to lens distortion, the actual observed pixel position is offset from the ideal position. Using a second-order radial distortion model, the corrected normalized coordinates are calculated by the following formula:

[0087] ;

[0088] in, and These are the normalized x and y coordinates after distortion correction; and These are the normalized x-axis and y-axis values ​​with distortion, as actually observed; This is the normalized radial distance from the distortion point to the optical center; The first-order radial distortion coefficient; It is the second-order radial distortion coefficient. and This is obtained in advance through the camera calibration program, in this embodiment. The value is -0.23. The value is 0.05.

[0089] After distortion correction, the corrected pixel coordinates are mapped to physical coordinates in the 19-stroke coordinate system of the lead screw motor through an affine transformation. The mathematical model of the affine transformation is as follows:

[0090]

[0091] in, The physical horizontal coordinate obtained by mapping is in millimeters, corresponding to the horizontal stroke position of the lead screw motor 19; The physical ordinate obtained from the mapping is in millimeters; The x-coordinate of the pixel after distortion correction; The vertical coordinate of the pixel after distortion correction; , , , The four coefficients of the affine transformation matrix represent the scaling and rotation relationship between the pixel coordinate system and the physical coordinate system; , The translation component of the affine transformation represents the offset between the origins of the two coordinate systems.

[0092] The six parameters (four matrix coefficients and two translation components) in the affine transformation matrix are solved through a calibration process. During calibration, a checkerboard calibration board with known spacing is placed at the cutting station. Images of the calibration board are captured by a K230 camera 20, and the pixel coordinates of the checkerboard corner points are extracted. A correspondence is established between these pixel coordinates and the known physical coordinates of the corner points. Let the total number of extracted... There are three corner points, each providing two equations. The six unknown parameters can be solved using at least three corner points (six equations). When the number of corner points is greater than three, the system of equations becomes overdetermined, and the least squares method is used to find the parameter values ​​that minimize the sum of squared residuals.

[0093]

[0094] in, This is the optimal affine transformation parameter matrix in the least squares sense (including rotation scaling coefficients and translation components). To determine the total number of corner points; For the first The known physical coordinate vectors of the corner points; For the first The homogeneous vector of pixel coordinates (including x-coordinate, y-coordinate and constant 1) after distortion correction of each corner point; This represents the Euclidean distance norm.

[0095] In actual operation, the PLC18 obtains physical coordinates. Then, a pulse command is sent directly to the target stroke position of the lead screw motor 19, and the lead screw motor 19 drives the shearing mechanism 5 to move to that position to complete the cutting.

[0096] The technical advantages of this algorithm are as follows: it eliminates the nonlinear distortion error of the K230 camera lens 20 through distortion correction, and achieves accurate mapping between image space and physical execution space through affine transformation, thereby improving the cutting positioning accuracy from approximately ±2 mm before correction to within ±0.3 mm. Combined with the ±0.1 mm mechanical positioning accuracy of the lead screw motor 19 itself, the overall cutting positioning accuracy of the system reaches ±0.4 mm, which significantly reduces the cutting loss caused by positioning deviation.

[0097] In the intelligent packing algorithm, during the packaging process in step S4, PLC18 no longer uses fixed-position feeding. Instead, it calculates the optimal placement coordinates based on the actual size of each black cabbage and the remaining space in the vegetable basket 12, controls the cross guide rail 13 to drive the vegetable basket 12 to the corresponding position, and then performs feeding.

[0098] The size information of each black cabbage stem comes from the recognition results of the K230 camera 20 during the cutting process in step S3. While performing YOLOv8 object detection, the model outputs the bounding rectangle of the black cabbage outline. The length and width of the bounding rectangle are converted into physical dimensions after the aforementioned affine transformation mapping, and recorded as the current length and width of the cabbage. The PLC18 then transmits this size data to the packing algorithm module.

[0099] The internal dimensions of the vegetable basket 12 are known constants, and the internal length and width of the vegetable basket are denoted as basket length and basket width, respectively. PLC18 maintains a current load status register, which records the starting vertical coordinate of the current row, the cumulative horizontal width occupied in the current row, and the maximum vegetable body length value of the current row.

[0100] The specific process of the greedy rectangular bin packing algorithm is as follows: For the current dish to be placed, first determine whether the remaining width of the current row is sufficient to accommodate the width of the dish:

[0101]

[0102] in, The remaining available width for the current row; The internal length of the basket (along the direction of the vegetables). This represents the total width occupied by the dishes already placed in the current row.

[0103] If the remaining width of the current row is greater than or equal to the width of the current menu item Then the dish body will be placed in the current row, and its center coordinates will be:

[0104] ;

[0105] in, The x-coordinate of the current dish's placement center; This represents the cumulative width already occupied by the current row before placement; The width of the current dish; The vertical coordinate of the current dish placement center; This is the starting y-coordinate of the current row; This is the maximum length of the dishes already placed in the current row, used to determine the row height.

[0106] Update the cumulative width and row height after placement:

[0107] ;

[0108] in, The preset gap value between adjacent vegetable pieces is 10 mm in this embodiment, which is used to prevent damage caused by squeezing adjacent vegetable pieces. This represents the length of the current dish.

[0109] If the remaining width of the current row is insufficient to accommodate the width of the current dish, a line break operation is performed, and the starting vertical coordinate of the row is updated.

[0110]

[0111] In this process, the starting ordinate of the new line after a line break is equal to the starting ordinate of the original line plus the original line height plus the spacing value. After a line break, the cumulative width and line height are reset to zero, and then the current menu item is placed in the new line according to the above rules.

[0112] If the starting vertical coordinate of the new line after the line break plus the length of the current vegetable body exceeds the internal width of the vegetable basket, it is determined that the current vegetable basket is full. PLC18 sends a full vegetable basket signal, controls the cross guide 13 to move the full vegetable basket to the discharge position, moves the empty vegetable basket to the receiving position, resets the loading status register, and continues feeding.

[0113] The PLC18 controls the two servo motors of the cross guide rail 13 to drive the basket 12 to move in the X and Y axes respectively according to the calculated placement center coordinates, so that the target placement position in the basket 12 is precisely aligned with the unloading point of the pen-shaped cylinder top 8. Then, the PLC18 controls the pen-shaped cylinder top 8 to extend and complete the unloading.

[0114] The technical advantages of this algorithm are as follows: by transferring the vegetable size information obtained in the cutting process to the packaging process, and combining it with the dual-degree-of-freedom positioning capability of the cross guide rail 13, an orderly and compact arrangement of vegetables within the basket is achieved. Compared with disordered feeding at fixed positions, the basket space utilization rate is increased by approximately 35%, the single basket loading capacity is increased, the frequency of basket replacement is reduced, and packaging efficiency is improved. At the same time, the orderly arrangement avoids mechanical damage caused by the stacking and compression of vegetables.

[0115] The pipeline parallel scheduling algorithm is used when there are multiple rotating stations 11 on the large turntable 9 (in this embodiment there are four rotating stations, evenly distributed along the circumference, with an angle of 90 degrees between adjacent stations). The PLC18 uses the pipeline parallel scheduling algorithm to coordinate the black cabbage on each station to perform different processes simultaneously, so as to improve the overall machine capacity.

[0116] PLC18 maintains a process status identifier variable for each rotary station. ,in This is the workstation number. The optional values ​​include: zero indicates idle, one indicates cleaning in progress, two indicates cleaning completed and ready for transfer, three indicates cutting in progress, four indicates cutting completed and ready for transfer, and five indicates packaging in progress.

[0117] Define the single execution time of each process as: (Cleaning time) (Cutting time) and (Packaging time). Production line scheduling cycle. Equal to the maximum of the three:

[0118]

[0119] in, The duration of a single scheduling cycle; The execution time of a single cleaning process (dynamically determined by the cleaning adaptive algorithm, with its maximum possible value used as the scheduling benchmark); This refers to the execution time of a single cutting process, including the total time spent on image acquisition, model reasoning, coordinate transformation, lead screw positioning, and blade assembly shearing. This refers to the execution time of a single packaging process, including the total time spent on box coordinate calculation, cross guide positioning, and cylinder unloading.

[0120] At the start of each scheduling cycle, the PLC18 executes the following scheduling judgment logic: iterates through the process status identifiers of all rotating stations and checks for any transfer conflicts. The condition for determining a transfer conflict is: there is at least one station whose process status identifier is "cleaning," "cutting," or "packaging" (i.e., the current process at that station has not yet been completed).

[0121] The rotation angle of the turntable when there is no collision is:

[0122]

[0123] in, The rotation angle of turntable 9 during each dispatch and transfer; In this embodiment, the total number of rotating stations on the large turntable 9 is [number missing]. Taking four corresponds to a rotation angle of 90 degrees.

[0124] PLC18 controls servo motor 15 to drive large turntable 9 to rotate. After the angle is adjusted, each rotating station advances synchronously to the next station position: the vegetables originally at the loading station enter the washing station, the vegetables originally at the washing station enter the cutting station, the vegetables originally at the cutting station enter the packaging station, and the finished vegetables originally at the packaging station are discharged. PLC18 then updates the process status indicators of each station.

[0125] The production capacity of the production line during steady-state operation is determined by the scheduling cycle, and the output per unit time is:

[0126]

[0127] in, The number of black cabbage plants processed per unit time (plants per second). The scheduling period is in seconds. In this embodiment, The maximum duration is 15 seconds. Approximately 3 seconds Approximately 4 seconds, therefore The cycle time is 15 seconds, and the steady-state output is approximately 4 trees per minute. Compared to a single-station sequential operation (cleaning, cutting, and packaging are performed sequentially, with a single cycle time of approximately 22 seconds and an output of approximately 2.7 trees per minute), the parallel scheduling of the production line increases capacity by approximately 48%.

[0128] When the adaptive cleaning algorithm allocates a shorter cleaning time to a lightly contaminated vegetable, that station may complete cleaning ahead of schedule. At this point, the PLC18 detects that the station's status has changed to "cleaned and ready for transfer," but other stations may still be in the process of executing their processes. The PLC18 marks this station as ready and waits for the other stations to complete their processes before transferring the vegetables. This waiting mechanism ensures the synchronization of the production line and avoids disruptions to the turntable rotation sequence caused by the early completion of individual stations.

[0129] The technical advantages of this algorithm are as follows: By fully utilizing the multi-station structure of the large turntable with 9 workstations through parallel scheduling of the pipeline, the three processes of washing, cutting, and packaging are executed in overlapping time, significantly shortening the equivalent processing cycle of a single vegetable and increasing the overall machine capacity. The scheduling algorithm works in conjunction with the adaptive washing algorithm, using the longest process time as the scheduling benchmark, which ensures the full execution of each process while maximizing parallel efficiency.

[0130] The processing method of this invention introduces an adaptive cleaning control algorithm and a cutting coordinate transformation algorithm into the cleaning and cutting processes, achieving dynamic adaptive adjustment of processing parameters according to individual differences in the vegetables. In the cleaning stage, the camera at the cutting station is reused to pre-scan and assess the degree of surface contamination of the vegetables, and the cleaning time and rotation speed of the rotating station are automatically adjusted based on the contamination coverage. This ensures that heavily contaminated vegetables receive more thorough cleaning while lightly contaminated vegetables are prevented from being over-washed, improving the cleaning pass rate while reducing water consumption. In the cutting stage, radial distortion correction and affine transformation mapping accurately convert the camera image pixel coordinates into the physical stroke coordinates of the lead screw motor, eliminating the influence of lens distortion and coordinate system deviation on cutting positioning accuracy. This allows the visual recognition results to accurately drive the actuator to complete precise cutting, further reducing the cutting loss rate caused by positioning deviation. Both algorithms are deeply coupled with the device's own hardware structure. Without adding additional sensors or actuators, the software algorithms release the reuse value of the existing hardware, achieving a synergistic effect between the device structure and the control algorithm.

[0131] The processing method of this invention further introduces intelligent packing algorithms and parallel scheduling algorithms into the packaging and end-to-end scheduling stages, achieving orderly packaging processes and parallel operation of multiple workstations. In the packaging stage, utilizing the circumscribed rectangle dimensions of the vegetables obtained during the cutting process, combined with the dual-degree-of-freedom positioning capability of the cross-guide rails, a greedy rectangular packing strategy is used to calculate the optimal placement position of each vegetable in the basket. This ensures that the vegetables are arranged in an orderly and compact manner within the basket, improving basket space utilization, reducing basket replacement frequency, and avoiding squeezing damage to the vegetables caused by disorderly stacking. In the end-to-end scheduling stage, utilizing the multi-rotating workstation structure of the large turntable, the parallel scheduling algorithm ensures that vegetables at different workstations simultaneously perform washing, cutting, and packaging processes. The longest process time is used as the unified scheduling cycle, ensuring that each process is fully executed while allowing for overlapping and parallel operation in time, significantly improving overall machine capacity. The two algorithms mentioned above are functionally coupled with the dual-degree-of-freedom motion capability of the cross guide rail and the multi-station rotation structure of the large turntable, respectively, transforming the structural advantages of the packaging device and the conveying mechanism into process advantages that can be controlled by the algorithm, thus realizing the synergistic amplification effect of the device's structural potential and the intelligent algorithm.

[0132] While the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the invention. Those skilled in the art can make various modifications and refinements without departing from the spirit and scope of the invention. Therefore, the scope of protection of the present invention shall be determined by the claims.

Claims

1. A clean packaging integrated device for black cabbage, comprising an aluminum profile frame (3), characterized in that, The aluminum profile frame (3) includes a cleaning device, a cutting device, a packaging device and an electrical control system; it also includes a large turntable (9) with a rotating station (11) on the turntable (9). The cleaning device includes a water gun (1) and an electric ball valve (2). The water gun (1) is fixed to the upper part of the aluminum profile frame (3) and vertically downward aligned with the rotating station (11). The electric ball valve (2) is installed in the water inlet pipe of the water gun (1) to control the flow of water. During cleaning, the rotating station (11) drives the black cabbage to rotate, so that the water flows to wash the leaves and stems in sequence. The cutting device includes a camera (20), a shearing mechanism (5), a blade group drive control module (6), and a lead screw motor (19). The camera (20) is installed below the cutting station to collect images of black cabbage and locate the coordinates of the cutting position. The shearing mechanism (5) is installed on the lead screw motor (19), and the lead screw motor (19) drives the shearing mechanism (5) to move horizontally to the target cutting position. The shearing mechanism (5) includes a set of blades with grid-like fine teeth. The blade group drive control module (6) controls the blade group with grid-like fine teeth to open and close and perform relative linear reciprocating motion to complete the cutting. The packaging device includes a pen-shaped cylinder top (8), a basket (12) and a cross rail (13), with the basket (12) placed on the cross rail (13); The electrical control system includes a PLC (18), which is connected to the electric ball valve (2), the blade group drive control module (6), the lead screw motor (19) and the camera (20) respectively, and is used to coordinate the timing of the cleaning, cutting and packaging processes.

2. The integrated clean packaging device for black cabbage according to claim 1, characterized in that, The cleaning device also includes a waterproof shell (7), which is made of transparent acrylic sheet and covers the entire cleaning area. The front of the waterproof shell (7) is provided with an openable observation window and the bottom is provided with a drain outlet.

3. The integrated clean packaging device for black cabbage according to claim 1, characterized in that, The water gun (1) is connected to a water source through a water pipe (4). The tail of the water gun (1) is equipped with a manual adjustment knob for switching between three water spray patterns: columnar, fan-shaped, and mist-shaped.

4. The integrated clean packaging device for black cabbage according to claim 1, characterized in that, The large turntable (9) is mounted on the aluminum profile frame (3) via a hollow rotating platform (14). The servo motor (15) drives the hollow rotating platform (14) to rotate the large turntable (9). The rotating station (11) is driven to rotate by a 42CM06 motor (16) via a gear (17).

5. The integrated clean packaging device for black cabbage according to claim 1, characterized in that, The camera (20) is connected to the PLC (18) via an Ethernet interface. It uses a deep learning-based target detection model to identify the outline features of the black cabbage and locate the coordinates of key points at the junction of the stem and leaves. The PLC (18) controls the lead screw motor (19) to drive the shearing mechanism (5) to move horizontally to the target cutting position according to the coordinates.

6. The integrated clean packaging device for black cabbage according to claim 1, characterized in that, The cutting device also includes a fixing device (10), which is located at the cutting station and is used to fix the black cabbage during cutting. The cross guide rail (13) is a two-degree-of-freedom cross slide structure; the pen-shaped cylinder top body (8) is controlled by the PLC (18) and is used to feed the cut black cabbage into the vegetable basket (12).

7. A treatment method using the integrated clean packaging device for black cabbage as described in any one of claims 1-6, characterized in that, Includes the following steps: S1. Loading: Place the black cabbage on the rotating station (11) on the large turntable (9); S2. Cleaning: PLC (18) controls the large turntable (9) to rotate, transporting the black cabbage to the cleaning station, controlling the electric ball valve (2) to open so that the water gun (1) sprays water, and at the same time the rotating station (11) drives the black cabbage to rotate, so that the water flow washes the leaves and stems in sequence. After cleaning, the electric ball valve (2) is closed. S3, Cutting: The large turntable (9) continues to rotate to transport the black cabbage to the cutting station. The camera (20) collects images of the black cabbage and identifies the cutting position coordinates at the junction of the stem and the leaves. After receiving the coordinates, the PLC (18) controls the screw motor (19) to drive the shearing mechanism (5) to move to the target cutting position. The blade group drive control module (6) drives the grid-shaped fine tooth blade group of the shearing mechanism (5) to open and close and make relative linear reciprocating motion to complete the cutting. After the cutting is completed, the device is reset. S4. Packaging: PLC (18) controls the top body (8) of the pen-shaped cylinder to feed the cut black cabbage into the vegetable basket (12) on the cross guide rail (13) to complete the packaging.

8. The processing method according to claim 7, characterized in that, In step S2, before cleaning, the PLC controls the camera to pre-scan and photograph the black cabbage on the rotating station, acquiring an image of the cabbage surface and calculating a contamination coverage rate index. The PLC adaptively adjusts the cleaning parameters based on the contamination coverage rate index, including the cleaning duration and the rotation speed of the rotating station. The contamination coverage rate index is obtained by extracting the area ratio of soil color gamut pixels to the total outline pixels of the cabbage after color space conversion of the pre-scanned image. When the contamination coverage rate index is greater than a preset heavy contamination threshold, the PLC sets the cleaning duration to an upper limit and increases the rotation speed of the rotating station. When the contamination coverage rate index is less than a preset light contamination threshold, the PLC sets the cleaning duration to a lower limit and decreases the rotation speed of the rotating station. When the contamination coverage rate index is between the light and heavy contamination thresholds, the PLC performs linear interpolation calculations between the upper and lower limits of the cleaning duration based on the contamination coverage rate index to determine the cleaning duration.

9. The processing method according to claim 8, characterized in that, In step S3, after the camera acquires a bottom-view image from below the cutting station, the PLC performs an affine transformation mapping from the pixel coordinate system to the physical coordinate system on the image coordinates. Specifically, this includes: first, performing radial distortion correction on the original image to eliminate barrel distortion caused by the lens; then, using a pre-calibrated affine transformation matrix to map the distortion-corrected pixel coordinates to physical coordinates in the ball screw motor travel coordinate system; the affine transformation matrix is ​​obtained by placing a calibration plate with a known spacing at the cutting station, acquiring an image of the calibration plate, and then performing least-squares fitting on the pixel coordinates of the corner points of the calibration plate and their corresponding physical coordinates. In step S4, the PLC calculates the optimal placement coordinates of the current vegetable in the basket based on the outer rectangle size of the current black cabbage outline obtained by the camera in the cutting step, combined with the known internal dimensions of the basket and the cumulative occupied area of ​​the already loaded vegetables, using a greedy rectangular packing strategy. The PLC controls the cross guide rail to drive the basket to move according to the optimal placement coordinates, so that the target placement position of the basket is aligned with the unloading point before performing the unloading action. The greedy rectangular packing strategy is as follows: vegetables are arranged sequentially along the long side of the basket. When the remaining width of the current row is insufficient to accommodate the width of the current vegetable, the arrangement is changed to a new row. When changing rows, the row height is taken as the maximum length value of the vegetables already arranged in that row.

10. The processing method according to claim 9, characterized in that, The large turntable is equipped with at least two rotating stations, and the PLC uses a pipeline parallel scheduling algorithm to control the black cabbage on each rotating station to work in parallel between the washing station, the cutting station and the packaging station. The parallel scheduling algorithm for the production line includes: the PLC maintains a process status identifier for each rotary station, the process status identifier including idle, cleaning in progress, cleaning completed and awaiting transfer, cutting in progress, cutting completed and awaiting transfer, and packaging in progress; within each scheduling cycle, the PLC traverses the process status identifiers of all rotary stations to determine whether there is a transfer conflict between the stations; when there is no transfer conflict, the PLC controls the large turntable to rotate by one station spacing angle, so that each rotary station synchronously enters the next process; when there is a transfer conflict, the PLC delays the turntable rotation until the current process of the conflicting station is completed; the scheduling cycle is equal to the maximum value among the cleaning duration, cutting duration, and packaging duration.

Images