A vegetable sorting and packing system and method

Abstract

The application relates to a vegetable sorting and packing system and method, and relates to the technical field of food processing automation. The system comprises a disclumping module, a screening module, a packing module and a detection module. The disclumping module is used for disclumping the clumped vegetables into disclumped vegetables. The screening module is used for removing first problem vegetables in the disclumped vegetables. A visual detection assembly is used for acquiring a first real-time image of the disclumped vegetables. A screening assembly is used for dividing the disclumped vegetables into the first problem vegetables, first normal vegetables and first pending vegetables. The screening assembly is further used for removing the first problem vegetables, conveying the first pending vegetables to the packing module and conveying the first pending vegetables to a rechecking assembly. The rechecking assembly is used for conveying the first pending vegetables to the inlet of the screening module. The packing module is used for packing the first normal vegetables into vegetable bags. The detection module is used for detecting whether there is a foreign matter in the vegetable bags. The application has the effect of improving the efficiency of vegetable sorting and packing.

Description

Technical Field

[0001] This application relates to the field of food processing automation technology, and in particular to a vegetable sorting and packaging system and method. Background Technology

[0002] The food processing industry is currently undergoing a transformation towards automation and intelligentization, aiming to improve production efficiency, reduce operating costs, and ensure consistent food quality and safety. With technological advancements, automated equipment such as conveyor belts and testing instruments are increasingly widely used in food processing, significantly enhancing the flexibility and responsiveness of production lines.

[0003] Vegetable processing production lines using relevant technologies, especially when handling quick-frozen vegetables, typically consist of key steps such as preliminary screening, weighing, and re-inspection. The preliminary screening step uses mechanical stirring or manual beating to loosen clumps of vegetables and remove those that are yellowed, spotted, broken, or have foreign matter adhering to them. The weighing step uses an automatic weighing machine to control the weight of each bag of vegetables, ensuring that each bag has the same weight. The re-inspection step uses detection equipment to check for foreign objects (such as metal fragments, glass shards, etc.) inside the packaged vegetables.

[0004] Regarding the aforementioned technologies, the accuracy of the screening process is affected by human judgment, making it difficult to achieve high consistency and repeatability, resulting in low efficiency in vegetable sorting and packaging. Summary of the Invention

[0005] To improve the efficiency of vegetable sorting and packaging, this application provides a vegetable sorting and packaging system and method.

[0006] In the first aspect, this application provides a vegetable sorting and packaging system, which adopts the following technical solution: A vegetable sorting and packaging system includes a de-clustering module, a screening module, a packaging module, and an inspection module; the screening module includes a visual inspection component, a screening component, and a re-inspection component. The unclogging module is used to break up clumps of vegetables into unclogged vegetables through vibration. The filtering module is used to remove the first problematic vegetable from the unpacked vegetables; wherein, the visual detection component is used to acquire a first real-time image of the unpacked vegetables; the filtering component is used to divide the unpacked vegetables into the first problematic vegetable, the first normal vegetable, and the first undetermined vegetable according to the first real-time image; the filtering component is also used to remove the first problematic vegetable, and transport the first normal vegetable to the packaging module, and transport the first undetermined vegetable to the re-inspection component; the re-inspection component is used to transport the first undetermined vegetable to the entrance of the filtering module; The packaging module is used to package the first normal vegetables into vegetable packages; The detection module is used to detect whether there are foreign objects inside the vegetable package; if there are, the vegetable package is transported to the scrap area; if there are no foreign objects, the vegetable package is transported to the shipping area.

[0007] By adopting the above technical solution, the unpacking module uses vibration to unpack vegetables, preventing them from sticking together and affecting sorting; the screening module uses visual inspection and screening components to classify vegetables into problematic, normal, and undetermined categories, and a re-inspection component sends undetermined vegetables back for re-screening, forming a closed-loop feedback to reduce false rejection; the packaging module only packages the first batch of normal vegetables; and the detection module detects foreign objects in the vegetable packages and directs them to the scrap or shipment area accordingly. Overall, this achieves efficient and accurate sorting from unpacking to finished product, improving packaging quality and safety.

[0008] Optionally, the packaging module includes n packaging devices, n baffles, n-1 distributors, and an input conveyor belt; the input conveyor belt has n outlets, and the n outlets are connected to the inlets of the n packaging devices; the n baffles are rotatably connected to the n outlets; with the forward direction of the input conveyor belt as a reference, the n-1 distributors correspond one-to-one with the first n-1 outlets, and the i-th opening is located in front of the i-th distributor, and the distance between the i-th opening and the i-th distributor is a preset distance.

[0009] By adopting the above technical solution, and by setting up n packaging devices, n baffles, n-1 distributors, and an input conveyor belt with n outlets, and controlling the positional relationship between the distributors and their corresponding outlets at preset distances, vegetables can be flexibly allocated to different packaging devices according to their type or grade. The rotation of the baffles controls the opening and closing of the outlets, and the distributors guide the vegetable flow in advance, avoiding congestion and misallocation. This achieves multi-channel parallel packaging, significantly improving packaging efficiency while ensuring balanced load on each packaging device.

[0010] Optionally, the packaging module further includes n output conveyor belts, n movable base plates located at the ends of the n output conveyor belts, and a combined conveyor belt; the ends of the n output conveyor belts are at different horizontal heights, and the ends of the n output conveyor belts and the combined conveyor belt are located on the same vertical plane; the n movable base plates are rotatably connected to the ends of the n output conveyor belts.

[0011] By adopting the above technical solution, and by setting up n output conveyor belts with different end heights, a movable base plate, and a combined conveyor belt, with each end of the conveyor belt and the combined conveyor belt located on the same vertical plane, vegetable bags from different output conveyor belts can fall sequentially onto the combined conveyor belt when the movable base plate flips, achieving convergence. The height difference avoids collisions and stacking during convergence, and the movable base plate controls the timing of material falling, thus orderly merging vegetable bags produced by multiple packaging devices into a single conveyor belt, facilitating subsequent centralized inspection or shipment.

[0012] Optionally, the screening assembly includes an upper conveyor belt, a lower conveyor belt, a first screening device, a second screening device, and a third screening device; the upper conveyor belt is located above the lower conveyor belt. The first screening device is used to move the first problematic vegetable and the undetermined vegetable located on the upper conveyor belt to the lower conveyor belt; The second screening device is used to identify the second problematic vegetable, the second normal vegetable, and the second undetermined vegetable on the lower conveyor belt, and to separate the second problematic vegetable; The third screening device is used to remove the second problematic vegetable from the lower conveyor belt.

[0013] By adopting the above technical solution, a double-layered conveyor belt system is used in conjunction with three screening devices: the first screening device moves problematic and undetermined vegetables from the upper conveyor belt to the lower conveyor belt; the second screening device separates problematic vegetables from other vegetables; and the third screening device finally removes the problematic vegetables. This structure achieves multi-level separation of problematic vegetables, preventing them from mixing with the normal vegetable stream; at the same time, undetermined vegetables can continue along the upper conveyor belt to the packaging module, while normal vegetables remain on the lower conveyor belt. This clear division of labor reduces duplicate identification and erroneous actions, improving screening accuracy.

[0014] Secondly, this application provides a method for sorting and packaging vegetables, employing the following technical solution: A method for sorting and packaging vegetables includes: After the clump of vegetables is broken down into smaller pieces, a first real-time image of the smaller pieces of vegetables is obtained. Based on the first real-time image, the unblocked vegetables are divided into first problem vegetables, first normal vegetables, and first undetermined vegetables; The first problematic vegetable is removed, the first normal vegetable is transported to the packaging module, and the first undetermined vegetable is transported to the re-inspection component; Pack the first normal vegetables into vegetable packages; Detect whether there are foreign objects inside the vegetable package; If present, the vegetable package will be transported to the disposal area; If not, the vegetable package is transported to the shipping area.

[0015] By adopting the above technical solution, real-time images are acquired after sorting, and vegetables are classified into problem vegetables, normal vegetables, and vegetables awaiting determination. Vegetables awaiting determination are sent to the re-inspection component for recycling, while only normal vegetables are packaged, and foreign object detection is performed on the vegetable packages. This process avoids simply rejecting potentially qualified vegetables, reducing the false detection rate; the addition of a foreign object detection step after packaging ensures that the shipped vegetable packages are free of foreign objects. The entire method has a clear logic, balances sorting efficiency and finished product safety, and is suitable for continuous operation on automated production lines.

[0016] Optionally, the first problematic vegetable and the first undetermined vegetable on the upper conveyor belt are blown off to the lower conveyor belt by the first screening device, and the first normal vegetable moves along the upper conveyor belt to the packaging module; Acquire a second real-time image of the lower conveyor belt; In the second real-time image, the fallen vegetables that have dropped from the upper conveyor belt to the lower conveyor belt are identified; Based on the second real-time image, the fallen vegetables are divided into second problem vegetables, second normal vegetables, and second undetermined vegetables; The second problematic vegetable on the conveyor belt is blown to the scrap area by the third screening device, and the two undetermined vegetables are blown to the re-inspection component, while the second normal vegetable moves to the packaging module along the lower conveyor belt.

[0017] By adopting the above technical solution, the first screening device blows problematic and undetermined vegetables from the upper conveyor belt to the lower conveyor belt, while the undetermined vegetables continue to move along the upper conveyor belt to the packaging module. Real-time images are acquired on the lower conveyor belt to identify and re-sort the fallen vegetables. The third screening device blows the second batch of problematic vegetables to the scrap level and the second batch of undetermined vegetables back for inspection. This process performs a second precise sorting of the initially undetermined vegetables, while the normal vegetables on the lower conveyor belt can still be packaged, achieving parallel processing through dual channels and improving the overall sorting rate and resource utilization.

[0018] Optionally, based on the positions of the first problematic vegetable and the first undetermined vegetable in the first real-time image, the first position coordinates of the first problematic vegetable and the first undetermined vegetable on the upper conveyor belt are obtained; Based on the first position coordinates and the moving speed of the upper conveyor belt, calculate the target time for the first problematic vegetable and the first undetermined vegetable to arrive at the first screening device; Based on the second real-time image and the moving speed of the lower conveyor belt, it is determined whether there are vegetables at the target position of the lower conveyor belt at the target time. The target position is located on the lower conveyor belt and below the first screening device. If so, the exposed area is determined in the second real-time image; The target speed is obtained based on the location of the exposed area and the speed range of the lower conveyor belt. Adjust the lower conveyor belt to move at the target speed.

[0019] By employing the above technical solution, the location coordinates of problematic and unselected vegetables are located based on the first real-time image. The arrival time of these vegetables at the first screening device is calculated by combining the conveyor belt speed, and it is determined whether there are vegetables at the corresponding position on the conveyor belt at that moment. If so, the exposed area is identified, and the conveyor belt speed is adjusted accordingly to ensure that falling vegetables avoid existing vegetables. This control strategy avoids the stacking or collision of vegetables on the upper and lower conveyor belts, ensuring the clarity of the secondary imaging and the accuracy of sorting, and reducing misjudgments caused by overlap.

[0020] Optionally, based on the second real-time image, determine the second position coordinates of the fallen vegetable on the lower conveyor belt; Based on the second position coordinates, the second screening device is run to adjust the posture of the fallen vegetables, thereby changing the posture of the fallen vegetables; During the operation of the second screening device, images of the fallen vegetables are continuously acquired to obtain an image sequence; Based on the image sequence, the fallen vegetables are classified into the second problem vegetables, the second normal vegetables, and the second undetermined vegetables.

[0021] By employing the above technical solution, after obtaining the position coordinates of vegetables falling on the lower conveyor belt, a second screening device is used to adjust their posture, causing them to rotate or move, and continuously acquiring image sequences. Secondary classification is then performed based on images from multiple angles or in dynamic postures. Compared to a single static image, this method can more comprehensively observe multiple surfaces of the vegetables, effectively identify hidden problem features, and thus improve the accuracy of classifying vegetables into problem vegetables, normal vegetables, or vegetables awaiting classification.

[0022] Optionally, based on the last image in the image sequence, it can be determined whether there is a second normal vegetable or a second undetermined vegetable within the periphery of the second problem vegetable; If so, the position of the second normal vegetable or the second undetermined vegetable is recorded as the third position coordinate, and the fourth position coordinate of the second problem vegetable is determined; Based on the third and fourth position coordinates, the operating mode of the second screening device is set; According to the operating mode, control the second screening device to keep the second normal vegetable or the second undetermined vegetable away from the second problem vegetable.

[0023] By adopting the above technical solution, based on the last image in the image sequence, it is determined whether there are normal or undetermined vegetables around the second problematic vegetable. If so, their respective position coordinates are recorded, and the operation mode of the screening equipment is set to keep the normal or undetermined vegetables away from the problematic vegetable. This operation avoids accidentally damaging nearby qualified vegetables when removing the problematic vegetable, ensuring the selectivity of the sorting action. By intelligently adjusting the effective area and intensity of the screening equipment, precise separation is achieved, reducing the waste of normal vegetables.

[0024] Optionally, if the number of vegetables in the second problem is greater than one, calculate the shortest distance between the vegetables in the second problem; If the shortest distance is less than a preset distance threshold, then a collection path for the second problem vegetables is generated based on the fourth location coordinates. The second screening device is controlled to blow the second problematic vegetables onto the lower conveyor belt to the collection point according to the collection path.

[0025] By adopting the above technical solution, when multiple problematic vegetables exist, the shortest distance between them is calculated. If it is less than a distance threshold, a convergence path is generated, and the screening device is controlled to blow each problematic vegetable along the path to the same convergence point. This method concentrates multiple scattered problematic vegetables in one place, facilitating subsequent unified collection or processing, reducing the dispersion of multiple rejection actions, and improving rejection efficiency. At the same time, it avoids mutual interference due to excessively close distances, making the blowing process orderly and controllable, and reducing the probability of omission.

[0026] Thirdly, this application provides a smart terminal, which adopts the following technical solution: A smart terminal includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any of the preceding methods.

[0027] Fourthly, this application provides a computer storage medium capable of storing corresponding programs, which facilitates improving the efficiency of vegetable sorting and packaging, and adopts the following technical solution: A computer-readable storage medium storing a computer program that can be loaded by a processor and executed in any of the above-described vegetable sorting and packaging methods.

[0028] In summary, this application includes at least one of the following beneficial technical effects: 1. The unpacking module uses vibration to unpack vegetables, preventing them from sticking together and affecting sorting. The sorting module uses visual inspection and sorting components to classify vegetables into problem vegetables, normal vegetables, and undetermined vegetables. A re-inspection component sends undetermined vegetables back for re-sorting, forming a closed-loop feedback to reduce false rejections. The packaging module only packages the first batch of normal vegetables. The detection module detects foreign objects in the vegetable packages and directs them to the scrap or shipment area accordingly. Overall, this achieves efficient and accurate sorting from unpacking to finished product, improving packaging quality and safety. 2. A double-layered conveyor belt system with three screening devices is employed: the first screening device moves problematic and undetermined vegetables from the upper conveyor belt to the lower conveyor belt; the second screening device separates problematic vegetables from other vegetables; and the third screening device finally removes the problematic vegetables. This structure achieves multi-level separation of problematic vegetables, preventing them from mixing with the normal vegetable stream; simultaneously, undetermined vegetables can continue along the upper conveyor belt to the packaging module, while normal vegetables remain on the lower conveyor belt. This clear division of labor reduces duplicate identification and erroneous actions, improving screening accuracy. 3. The first sorting device blows problematic and undetermined vegetables from the upper conveyor belt to the lower conveyor belt, while the undetermined vegetables continue along the upper conveyor belt to the packaging module. Real-time images are acquired on the lower conveyor belt to identify and re-sort the fallen vegetables. The third sorting device blows the second batch of problematic vegetables to the scrapped state and the second batch of undetermined vegetables back for inspection. This process performs a second round of precise sorting of the initially undetermined vegetables, while the normal vegetables on the lower conveyor belt can still be packaged. This achieves parallel processing through two channels, improving the overall sorting rate and resource utilization. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of an overall control system disclosed in an embodiment of this application.

[0030] Figure 2 This is a schematic diagram of a vegetable sorting and packaging system disclosed in an embodiment of this application.

[0031] Figure 3 This is a schematic diagram of a unpacking module disclosed in an embodiment of this application.

[0032] Figure 4 This is a schematic diagram of a screening module disclosed in an embodiment of this application.

[0033] Figure 5 This is a schematic diagram of an output conveyor belt disclosed in an embodiment of this application.

[0034] Figure 6 This is a flowchart illustrating a vegetable sorting and packaging method disclosed in an embodiment of this application.

[0035] Figure 7 This is a flowchart illustrating a vegetable screening method disclosed in an embodiment of this application.

[0036] Figure 8 This is a flowchart illustrating a method for adjusting the falling position of vegetables disclosed in an embodiment of this application.

[0037] Figure 9 This is a flowchart illustrating a vegetable classification method disclosed in an embodiment of this application.

[0038] Figure 10 This is a flowchart illustrating a screening device control method disclosed in an embodiment of this application.

[0039] Figure 11 This is a flowchart illustrating a method for collecting problematic vegetables disclosed in an embodiment of this application. Detailed Implementation

[0040] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figures 1 to 11 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.

[0041] This application discloses a schematic diagram of an overall control system. (Refer to...) Figure 1 The system includes a vegetable sorting and packaging system 1, a controller 2, and a memory 3.

[0042] The vegetable sorting and packaging system 1 includes a vision inspection component 11, a screening component 12, a drive component 13, and an interaction component 14. The controller 2 can be a programmable logic controller, an industrial computer, or an embedded industrial control computer, which integrates an image processing unit and an actuator drive unit.

[0043] The vision inspection component 11 includes an industrial camera. The controller 2 is connected to the vision inspection component 11 and periodically reads the image data returned by the vision inspection component 11. The controller 2 performs inference on the image data and outputs the bounding box, defect confidence score, and preliminary classification result for each vegetable. Then, the controller 2 determines whether the sorting component 12 should be activated based on the bounding box, defect confidence score, and preliminary classification result, and sends a running signal to the corresponding sorting component 12 to start its operation. On the other hand, the controller 2 also sends a running signal to the drive component 13 to operate, thereby driving the conveyor belt within the vegetable sorting and packaging system 1.

[0044] In addition, the interactive component 14 includes at least one of a touch screen display, an alarm indicator, and a start / stop button. The controller 2 periodically sends system operating parameters (conveyor belt speed, load of each packaging device, current output, defect detection rate, etc.) to the interactive component 14 for real-time display.

[0045] The memory 3 can record various types of data and instructions generated by the controller 1.

[0046] This application discloses a vegetable sorting and packaging system. (Refer to...) Figure 2 The vegetable sorting and packaging system includes a de-clustering module 101, a screening module 102, a packaging module 103, and a detection module 104.

[0047] The de-clumping module 101 is used to break up clumps of vegetables into individual pieces through vibration. For frozen vegetables, when they leave the cold storage, they often contain ice crystals and fragments, and the vegetables are often stuck together by ice. Therefore, using the de-clumping module 101 allows for the extraction of individual, unclumped vegetables, which is beneficial for subsequently identifying and removing problematic vegetables. Please refer to... Figure 3 The unclogging module 101 consists of a vibrating plate 111 and a vibrating belt 112 located below the vibrating plate 111. After the clumps of vegetables are placed on the vibrating plate 111, the vegetables fall through the through holes on the vibrating plate 111 onto the vibrating belt 112. At the same time, the vibrating belt 112 itself is also vibrating, so that ice chips and ice scraps can fall through the vibrating belt 122 to the bottom, while the unclogging vegetables will stay on the vibrating belt 122.

[0048] The filtering module 12 is used to remove the first problematic vegetables from the unblocked vegetables. In this application, please refer to... Figure 3 The screening module 12 includes a visual inspection component, a screening component, and a re-inspection component. The visual inspection component acquires a first real-time image of the disassembled vegetables. The screening component, based on the first real-time image, classifies the disassembled vegetables into first problematic vegetables, first normal vegetables, and first undetermined vegetables. The screening component also removes the first problematic vegetables, conveys the first normal vegetables to the packaging module 13, and conveys the first undetermined vegetables to the re-inspection component. The re-inspection component conveys the first undetermined vegetables to the inlet of the screening module 12.

[0049] The visual detection component can be a camera or a video camera. For example, vegetables move together with the conveyor belt, and the camera is positioned above the conveyor belt so that the camera can capture images of the vegetables on the conveyor belt and obtain a first real-time image.

[0050] Furthermore, the screening component can call the visual inspection module 14 to detect the vegetables in the first real-time image and obtain the defect confidence score of each disassembled vegetable. The defect confidence score represents the probability that a defect exists on the disassembled vegetable; the higher the defect confidence score, the greater the probability that the disassembled vegetable has a defect. Based on the specific value of the defect confidence score, the disassembled vegetables are divided into first problem vegetables, first normal vegetables, and first pending vegetables. For example, vegetables with a defect confidence score of 85% to 100% are classified as first problem vegetables, vegetables with a defect confidence score of 10% to 85% are classified as first pending vegetables, and vegetables with a defect confidence score of 0% to 10% are classified as first normal vegetables. The aforementioned defects include, but are not limited to, insect infestation, damage, breakage, and discoloration.

[0051] For the filtering components, please refer to Figure 4 The screening assembly includes an upper conveyor belt 121, a lower conveyor belt 122, a first screening device, a second screening device, and a third screening device. The upper conveyor belt 121 is located above the lower conveyor belt 122. The first screening device is used to move first problematic vegetables and first undetermined vegetables located on the upper conveyor belt 121 to the lower conveyor belt 122. The second screening device is used to identify second problematic vegetables, second normal vegetables, and second undetermined vegetables on the lower conveyor belt 122, and to separate the second problematic vegetables. The third screening device is used to remove the second problematic vegetables from the lower conveyor belt 122.

[0052] For example, after detecting the first problematic vegetable and the first undetermined vegetable, the first problematic vegetable and the second undetermined vegetable are blown off the upper conveyor belt 121 by a first screening device (e.g., an air nozzle) and fall onto the lower conveyor belt 122. Then, a second screening device re-identifies the second problematic vegetable, the second normal vegetable, and the second undetermined vegetable on the lower conveyor belt 122, and separates the second problematic vegetable for subsequent removal. Finally, a third screening device removes the second problematic vegetable. Simultaneously, the third screening device can also transport the second undetermined vegetable to the entrance of the screening module 12, thereby achieving a secondary screening to ensure complete removal of any problematic vegetables.

[0053] Packaging module 13 is used to package the first normal vegetables into vegetable bags. Optional, please refer to... Figure 5 The packaging module 13 includes n packaging devices, n baffles, n-1 distributors, and an input conveyor belt. The input conveyor belt has n outlets, each connected to an inlet of one of the n packaging devices. The n baffles are rotatably connected to the n outlets. Taking the forward direction of the input conveyor belt as a reference, the n-1 distributors correspond one-to-one with the first n-1 outlets, and the i-th opening is located in front of the i-th distributor, with a preset distance between the i-th opening and the i-th distributor.

[0054] For example, please refer to Figure 5The packaging module 13 also includes n output conveyor belts 131, n movable base plates 132 located at the ends of the n output conveyor belts 131, and a connecting conveyor belt 133. The ends of the n output conveyor belts 131 are at different horizontal heights, and the ends of the n output conveyor belts 131 and the connecting conveyor belt 133 are located on the same vertical plane. The n movable base plates 132 are rotatably connected to the ends of the n output conveyor belts 131. When vegetable packages on two of the output conveyor belts fall onto the connecting conveyor belt at the same time, the movable base plate corresponding to one of the output conveyor belts can be rotated to adjust the falling time of the vegetable packages on that output conveyor belt.

[0055] The detection module 14 is used to detect whether there are foreign objects inside the vegetable package. If foreign objects are found, the vegetable package is transported to the disposal area. If no foreign objects are found, the vegetable package is transported to the shipping area.

[0056] Optionally, the detection module 14 includes a metal detection device and a non-metal detection device. The metal detection device can detect whether there are metal fragments or metal shavings mixed in with the vegetable package. For example, the metal detection device is a metal detector that uses the principle of electromagnetic induction to detect whether there are metal fragments such as iron or stainless steel mixed in with the vegetable package. The non-metal detection device is an X-ray foreign object detector, which can detect whether there are glass fragments mixed in with the vegetable package.

[0057] Please refer to Figure 6 This application discloses a method for sorting and packaging vegetables, the method comprising: Step S501: After the clump of vegetables is broken into smaller pieces, the first real-time image of the smaller pieces of vegetables is obtained.

[0058] The vegetables used in this application include green beans, peas, broad beans, cauliflower, and broccoli.

[0059] The first real-time image refers to the image of the disassembled vegetables that is captured in real time by the visual inspection component at the entrance of the screening module.

[0060] Step S502: Based on the first real-time image, divide the unblocked vegetables into first problem vegetables, first normal vegetables, and first undetermined vegetables.

[0061] The visual inspection module is invoked to detect the vegetables in the first real-time image and obtain the defect confidence score for each vegetable block. Based on the specific value of the defect confidence score, the vegetables block are divided into three categories: first problematic vegetables, first normal vegetables, and first undetermined vegetables.

[0062] Step S503: Remove the first problematic vegetable, transport the first normal vegetable to the packaging module, and transport the first undetermined vegetable to the re-inspection component.

[0063] The first batch of vegetables awaiting approval is transported to the re-inspection unit, which then sends them back to the entrance of the screening module via a conveyor belt. There, they are mixed with newly diced vegetables and subjected to visual inspection again to ensure a high detection rate of problematic vegetables. For example, if a section of green beans with insect holes is identified as the first batch of vegetables awaiting approval, it will enter the re-inspection unit and be sent back for re-photographing to prevent missed detection.

[0064] Step S504: Pack the first normal vegetables into a vegetable package.

[0065] Specifically, the packaging module receives the first batch of normal vegetables from the screening module. The packaging module includes multiple packaging devices, each of which can be independently set to a target weight. The first batch of normal vegetables is distributed to an idle packaging device via an input conveyor belt. The packaging device automatically completes the weighing, bag making, filling, and sealing operations, outputting sealed vegetable bags.

[0066] Step S505: Check if there are any foreign objects inside the vegetable package.

[0067] The detection module includes metal detection equipment and non-metal detection equipment. For example, the metal detection equipment is a metal detector, and the non-metal detection equipment is an X-ray foreign object detector.

[0068] Step S506: If present, transport the vegetable package to the disposal area.

[0069] Specifically, after foreign objects are detected inside the vegetable package, the vegetable package needs to be transported to the disposal area for disposal to prevent substandard products from entering the market.

[0070] Step S507: If not, transport the vegetable package to the shipping area.

[0071] The shipping area can be connected to an automatic palletizer or temporary storage station for subsequent packing, refrigeration, or shipping. For example, a batch of green beans that has been inspected and found to be free of foreign objects will be neatly arranged on the shipping conveyor belt, awaiting automatic palletizing.

[0072] By adopting the above technical solution, real-time images are acquired after sorting, and vegetables are classified into problem vegetables, normal vegetables, and vegetables awaiting determination. Vegetables awaiting determination are sent to the re-inspection component for recycling, while only normal vegetables are packaged, and foreign object detection is performed on the vegetable packages. This process avoids simply rejecting potentially qualified vegetables, reducing the false detection rate. The addition of a foreign object detection step after packaging ensures that the shipped vegetable packages are free of foreign objects. The entire method has a clear logic, balances sorting efficiency and finished product safety, and is suitable for continuous operation on automated production lines.

[0073] In real-world scenarios, sorting equipment typically uses air nozzles. Because the conveyor belt usually carries a large quantity of vegetables, the air nozzles, while removing problematic vegetables, may also blow nearby vegetables onto the next conveyor belt. Therefore, the vegetables need to be re-sorted on the next conveyor belt to ensure more accurate sorting. Please refer to [reference needed]. Figure 7 This application discloses a vegetable screening method, which includes: Step S601: The first problematic vegetable and the first undetermined vegetable on the upper conveyor belt are blown off to the lower conveyor belt by the first screening device, and the first undetermined vegetable moves along the upper conveyor belt to the packaging module.

[0074] Step S602: Obtain the second real-time image of the lower conveyor belt.

[0075] Optionally, a vision detection component is also installed above the lower conveyor belt to continuously capture images of the vegetables on the lower conveyor belt at a fixed frequency, thereby obtaining a second real-time image.

[0076] Step S603: In the second real-time image, identify the fallen vegetables that have fallen from the upper conveyor belt to the lower conveyor belt.

[0077] Optionally, the second real-time image can be analyzed in real time, and the individual vegetables in the image can be identified using the background subtraction method.

[0078] On the other hand, since there may be vegetables already placed on the lower conveyor belt, it is necessary to distinguish which vegetables have just fallen from the upper conveyor belt. This can be determined by a tracking algorithm. For example, the trigger time of the first screening device and the expected trajectory of the falling vegetables are recorded. Then, newly appearing vegetables are found at the corresponding positions in the second real-time image and recorded as fallen vegetables.

[0079] Step S604: Based on the second real-time image, the fallen vegetables are divided into second problem vegetables, second normal vegetables, and second pending vegetables.

[0080] The method of classifying fallen vegetables into second problem vegetables, second normal vegetables, and second pending vegetables is the same as the method of classifying unblocked vegetables into first problem vegetables, first normal vegetables, and first pending vegetables, and will not be repeated here.

[0081] Step S605: The second problematic vegetable on the lower conveyor belt is blown to the scrap area by the third screening device, and the second undetermined vegetable is blown to the re-inspection component, while the second normal vegetable moves to the packaging module along the lower conveyor belt.

[0082] Optionally, the third screening device includes multiple independently controllable air nozzles. For the second batch of normal vegetables, no blowing action is performed, allowing them to continue along the lower conveyor belt and eventually enter the packaging module for packaging. For the second batch of undetermined vegetables, the air nozzles blow them into the inlet of the re-inspection component, which then transports them via conveyor belt to the inlet of the screening module for secondary screening. For the second batch of problematic vegetables, the air nozzles blow them off the lower conveyor belt, causing them to fall onto the adjacent scrap conveyor belt and ultimately be sent to the scrap area.

[0083] By adopting the above technical solution, the first screening device blows problematic and undetermined vegetables from the upper conveyor belt to the lower conveyor belt, while the undetermined vegetables continue to move along the upper conveyor belt to the packaging module. Real-time images are acquired on the lower conveyor belt to identify and re-sort the fallen vegetables. The third screening device blows the second batch of problematic vegetables to the scrap level and the second batch of undetermined vegetables back for inspection. This process performs a second precise sorting of the initially undetermined vegetables, while the normal vegetables on the lower conveyor belt can still be packaged, achieving parallel processing through dual channels and improving the overall sorting rate and resource utilization.

[0084] In the following embodiments, when vegetables fall onto the lower conveyor belt, they may land on top of other vegetables, thus affecting subsequent processing. Therefore, this situation needs to be avoided as much as possible. Please refer to... Figure 8 This application discloses a method for adjusting the falling position of vegetables, the method comprising: Step S701: Based on the positions of the first problem vegetable and the first undetermined vegetable in the first real-time image, obtain the first position coordinates of the first problem vegetable and the first undetermined vegetable on the upper conveyor belt.

[0085] The first position coordinates refer to the positions of the first problem vegetable and the first undetermined vegetable on the upper conveyor belt. For example, the first position coordinates include the distance along the direction of movement of the conveyor belt and the lateral offset distance perpendicular to the direction of movement of the conveyor belt.

[0086] Step S702: Calculate the target time for the first problem vegetable and the first undetermined vegetable to arrive at the first screening device based on the first position coordinates and the moving speed of the upper conveyor belt.

[0087] For example, based on the first location coordinates and the location of the first screening device, the distance difference between the first location coordinates and the first screening device is obtained. The ratio of the distance difference to the moving speed is calculated to obtain the moving time. The sum of the current time and the moving time is calculated to obtain the target time.

[0088] Step S703: Based on the second real-time image and the moving speed of the lower conveyor belt, determine whether there are vegetables at the target position on the lower conveyor belt at the target time. The target position is on the lower conveyor belt and below the first screening device.

[0089] For example, the positions of each vegetable on the lower conveyor belt are obtained to form a position set. Based on the position set and the moving speed, the updated positions of each vegetable on the lower conveyor belt at the target time are obtained.

[0090] Step S704: If so, determine the exposed area in the second real-time image.

[0091] If vegetables are determined to be present at the target location, a new, vacant drop location needs to be found to avoid stacking. For example, candidate areas uncovered by vegetables are identified in the second real-time image; within these candidate areas, the region with an area greater than a preset area threshold and closest to the mapped coordinates of the first location on the lower conveyor belt is selected to obtain the exposed area. The preset area threshold is a preset empirical value; for example, the preset area threshold needs to be greater than three times the area of ​​a single piece of vegetable.

[0092] Step S705: Obtain the target speed based on the location of the exposed area and the speed range of the lower conveyor belt.

[0093] Optionally, based on the location of the exposed area and the target time, a candidate speed is obtained, such that when the lower conveyor belt moves at the candidate speed, the exposed area reaches the first position coordinate on the lower conveyor belt at the target time. It is then checked whether the candidate speed belongs to the moving speed range. If so, the candidate speed is used as the target speed. If not, the exposed area is updated.

[0094] Step S706: Adjust the lower conveyor belt to move at the target speed.

[0095] By employing the above technical solution, the location coordinates of problematic and unselected vegetables are located based on the first real-time image. The arrival time of these vegetables at the first screening device is calculated by combining the conveyor belt speed, and it is determined whether there are vegetables at the corresponding position on the conveyor belt at that moment. If so, the exposed area is identified, and the conveyor belt speed is adjusted accordingly to ensure that falling vegetables avoid existing vegetables. This control strategy avoids the stacking or collision of vegetables on the upper and lower conveyor belts, ensuring the clarity of the secondary imaging and the accuracy of sorting, and reducing misjudgments caused by overlap.

[0096] As vegetables fall from the upper conveyor belt to the lower conveyor belt, their posture changes. This can cause defects such as insect damage and discoloration to be obscured due to angle and temperature variations, leading to misjudgment. Therefore, this embodiment changes the orientation or spatial posture of the vegetables on the conveyor belt, allowing for observation from multiple angles. Please refer to... Figure 9 This application discloses a vegetable classification method, which includes: Step S801: Determine the second position coordinates of the fallen vegetables on the lower conveyor belt based on the second real-time image.

[0097] Optionally, an object detection algorithm is invoked to determine the bounding box of the fallen vegetable in the second real-time image. The coordinates of the geometric center of the bounding box are used as the second position coordinates.

[0098] Step S802: Based on the second position coordinates, run the second screening device to adjust the posture of the fallen vegetables, so that the posture of the fallen vegetables changes.

[0099] Optionally, the second screening device is an air nozzle. In this case, the air jet direction of the second screening device can be adjusted according to the second position coordinates, so that the air jet from the second screening device blows the fallen vegetables located at the second position coordinates.

[0100] Step S803: During the operation of the second screening device, images of the fallen vegetables are continuously acquired to obtain an image sequence.

[0101] For example, an industrial camera mounted above the lower conveyor belt begins capturing images at a preset frame rate and continues for a preset duration until the attitude adjustment action is completed. All images captured by the industrial camera are arranged in chronological order to form an image sequence.

[0102] Step S804: Based on the image sequence, divide the fallen vegetables into second problem vegetables, second normal vegetables, and second undetermined vegetables.

[0103] For example, the visual inspection module is invoked to inspect each image in the image sequence and obtain the defect confidence score for each image. If the defect confidence score of at least one image is greater than a first threshold, the fallen vegetable is classified as a second problem vegetable; if the defect confidence scores of all images are less than the first threshold and the defect confidence score of at least one image is greater than the second threshold, the fallen vegetable is classified as a second normal vegetable; if the defect confidence scores of all images are less than the second threshold, the fallen vegetable is classified as a second problem vegetable.

[0104] By employing the above technical solution, after obtaining the position coordinates of vegetables falling on the lower conveyor belt, a second screening device is used to adjust their posture, causing them to rotate or move, and continuously acquiring image sequences. Secondary classification is then performed based on images from multiple angles or in dynamic postures. Compared to a single static image, this method can more comprehensively observe multiple surfaces of the vegetables, effectively identify hidden problem features, and thus improve the accuracy of classifying vegetables into problem vegetables, normal vegetables, or vegetables awaiting classification.

[0105] To improve the accuracy of subsequent removal processes and avoid accidentally damaging adjacent, healthy vegetables while removing problematic ones, the problematic vegetables can be separated from the other vegetables. Please refer to... Figure 10 This application discloses a method for controlling a screening device, the method comprising: Step S901: Based on the last image in the image sequence, determine whether there is a second normal vegetable or a second undetermined vegetable within the periphery of the second problem vegetable.

[0106] Based on the bounding box or outline of each vegetable, calculate the center coordinates of the second problem vegetable, and draw a perimeter range with a preset radius centered on these coordinates. Then, iterate through the coordinates of all the second normal vegetables and the second problem vegetable to determine whether the center or edge of any other vegetable falls within this range.

[0107] Step S902: If so, record the position of the second normal vegetable or the second undetermined vegetable as the third position coordinate, and determine the fourth position coordinate of the second problem vegetable.

[0108] Step S903: Set the operating mode of the second screening device according to the third and fourth position coordinates.

[0109] The operation of the second screening device includes selecting which actuator, the air jet direction of the nozzle, the airflow intensity, and the action time. Optionally, a target offset can be preset. The required airflow direction and intensity are determined through geometric calculations. For example, pushing qualified vegetables away from problematic vegetables by at least 30mm, so that they are no longer within the range of action of the subsequent rejection nozzles.

[0110] Step S904: According to the operating mode, control the second screening device to keep the second normal vegetable or the second undetermined vegetable away from the second problem vegetable.

[0111] By adopting the above technical solution, based on the last image in the image sequence, it is determined whether there are normal or undetermined vegetables around the second problematic vegetable. If so, their respective position coordinates are recorded, and the operation mode of the screening equipment is set to keep the normal or undetermined vegetables away from the problematic vegetable. This operation avoids accidentally damaging nearby qualified vegetables when removing the problematic vegetable, ensuring the selectivity of the sorting action. By intelligently adjusting the effective area and intensity of the screening equipment, precise separation is achieved, reducing the waste of normal vegetables.

[0112] Please refer to Figure 11 This application discloses a method for collecting problematic vegetables, the method comprising: Step S1001: If the number of vegetables in the second problem is greater than one, calculate the shortest distance between the vegetables in the second problem.

[0113] Optionally, iterate through all pairwise combinations of vegetables from the second problem and calculate the Euclidean distance between each pair. Record the minimum value and its corresponding vegetable pair to obtain the shortest distance.

[0114] Step S1002: If the shortest distance is less than the preset distance threshold, then generate the aggregation path of the second problem vegetables based on the fourth position coordinates.

[0115] If the shortest distance is less than the preset distance threshold, it means that at least two problematic vegetables are close together. If they are removed separately, they may interfere with each other or be missed. Therefore, the method of collecting and removing them together is adopted.

[0116] Optionally, based on the order of the second problem vegetables along the conveyor belt, the plan is to blow the later second problem vegetables forward to the position of the first second problem vegetable, or blow the earlier second problem vegetables backward to the position of the last second problem vegetable, so that all the second problem vegetables eventually converge at the same coordinate.

[0117] Optionally, multiple nozzles can be planned to blow the corresponding second-problem vegetables from different directions toward a common midpoint.

[0118] Step S1003: Control the second screening device to blow the second problematic vegetables to the collection point on the lower conveyor belt according to the collection path.

[0119] By adopting the above technical solution, when multiple problematic vegetables exist, the shortest distance between them is calculated. If it is less than a distance threshold, a convergence path is generated, and the screening device is controlled to blow each problematic vegetable along the path to the same convergence point. This method concentrates multiple scattered problematic vegetables in one place, facilitating subsequent unified collection or processing, reducing the dispersion of multiple rejection actions, and improving rejection efficiency. At the same time, it avoids mutual interference due to excessively close distances, making the blowing process orderly and controllable, and reducing the probability of omission.

[0120] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0121] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as a vegetable sorting and packaging method.

[0122] Computer storage media include, for example, USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media that can store program code.

[0123] Based on the same inventive concept, this application provides a smart terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as a vegetable sorting and packaging method.

[0124] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0125] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A vegetable sorting and packaging system, characterized in that, It includes a de-clustering module, a filtering module, a packaging module, and an inspection module; the filtering module includes a visual inspection component, a filtering component, and a re-inspection component; The unclogging module is used to break up clumps of vegetables into unclogged vegetables through vibration. The filtering module is used to remove the first problematic vegetable from the unpacked vegetables; wherein, the visual detection component is used to acquire a first real-time image of the unpacked vegetables; the filtering component is used to divide the unpacked vegetables into the first problematic vegetable, the first normal vegetable, and the first undetermined vegetable according to the first real-time image; the filtering component is also used to remove the first problematic vegetable, and transport the first normal vegetable to the packaging module, and transport the first undetermined vegetable to the re-inspection component; the re-inspection component is used to transport the first undetermined vegetable to the entrance of the filtering module; The packaging module is used to package the first normal vegetables into vegetable packages; The detection module is used to detect whether there are foreign objects inside the vegetable package; if there are, the vegetable package is transported to the scrap area; if there are no foreign objects, the vegetable package is transported to the shipping area.

2. The vegetable sorting and packaging system according to claim 1, characterized in that, The packaging module includes n packaging devices, n baffles, n-1 distributors, and an input conveyor belt; the input conveyor belt has n outlets, and the n outlets are connected to the inlets of the n packaging devices; the n baffles are rotatably connected to the n outlets; with reference to the forward direction of the input conveyor belt, the n-1 distributors correspond one-to-one with the first n-1 outlets, and the i-th opening is located in front of the i-th distributor, and the distance between the i-th opening and the i-th distributor is a preset distance.

3. The vegetable sorting and packaging system according to claim 2, characterized in that, The packaging module also includes n output conveyor belts, n movable base plates located at the ends of the n output conveyor belts, and a combined conveyor belt; the ends of the n output conveyor belts are at different horizontal heights, and the ends of the n output conveyor belts and the combined conveyor belt are located on the same vertical plane; the n movable base plates are rotatably connected to the ends of the n output conveyor belts.

4. The vegetable sorting and packaging system according to claim 1, characterized in that, The screening assembly includes an upper conveyor belt, a lower conveyor belt, a first screening device, a second screening device, and a third screening device; the upper conveyor belt is located above the lower conveyor belt. The first screening device is used to move the first problematic vegetable and the first undetermined vegetable located on the upper conveyor belt to the lower conveyor belt; The second screening device is used to identify the second problematic vegetable, the second normal vegetable, and the second undetermined vegetable on the lower conveyor belt, and to separate the second problematic vegetable; The third screening device is used to remove the second problematic vegetable from the lower conveyor belt.

5. A method for sorting and packaging vegetables, characterized in that, The method is performed by the vegetable sorting and packaging system of claim 1, including: After the clump of vegetables is broken down into smaller pieces, a first real-time image of the smaller pieces of vegetables is obtained. Based on the first real-time image, the unblocked vegetables are divided into first problem vegetables, first normal vegetables, and first undetermined vegetables; The first problematic vegetable is removed, the first normal vegetable is transported to the packaging module, and the first undetermined vegetable is transported to the re-inspection component; Pack the first normal vegetables into vegetable packages; Detect whether there are foreign objects inside the vegetable package; If present, the vegetable package will be transported to the disposal area; If not, the vegetable package is transported to the shipping area.

6. The vegetable sorting and packaging method according to claim 5, characterized in that, The step of removing the first problematic vegetable, conveying the first normal vegetable to the packaging module, and conveying the undetermined vegetable to the re-inspection component includes: The first screening device blows the first problematic vegetable and the first undetermined vegetable from the upper conveyor belt to the lower conveyor belt, and the first normal vegetable moves along the upper conveyor belt to the packaging module. Acquire a second real-time image of the lower conveyor belt; In the second real-time image, the fallen vegetables that have dropped from the upper conveyor belt to the lower conveyor belt are identified; Based on the second real-time image, the fallen vegetables are divided into second problem vegetables, second normal vegetables, and second undetermined vegetables; The second problematic vegetable on the lower conveyor belt is blown to the scrap area by the third screening device, and the second undetermined vegetable is blown to the re-inspection component, while the second normal vegetable moves along the lower conveyor belt to the packaging module.

7. The vegetable sorting and packaging method according to claim 6, characterized in that, Before the first problematic vegetable and the first undetermined vegetable on the upper conveyor belt are blown off to the lower conveyor belt by the first screening device, the process further includes: Based on the positions of the first problematic vegetable and the first undetermined vegetable in the first real-time image, the first position coordinates of the first problematic vegetable and the first undetermined vegetable on the upper conveyor belt are obtained; Based on the first position coordinates and the moving speed of the upper conveyor belt, calculate the target time for the first problematic vegetable and the first undetermined vegetable to arrive at the first screening device; Based on the second real-time image and the moving speed of the lower conveyor belt, it is determined whether there are vegetables at the target position of the lower conveyor belt at the target time. The target position is located on the lower conveyor belt and below the first screening device. If so, the exposed area is determined in the second real-time image; The target speed is obtained based on the location of the exposed area and the speed range of the lower conveyor belt. Adjust the lower conveyor belt to move at the target speed.

8. The vegetable sorting and packaging method according to claim 6, characterized in that, The step of classifying the fallen vegetables into two categories—the second problem vegetables, the second normal vegetables, and the second undetermined vegetables—based on the second real-time image includes: Based on the second real-time image, determine the second position coordinates of the fallen vegetable on the lower conveyor belt; Based on the second position coordinates, the second screening device is run to adjust the posture of the fallen vegetables, thereby changing the posture of the fallen vegetables; During the operation of the second screening device, images of the fallen vegetables are continuously acquired to obtain an image sequence; Based on the image sequence, the fallen vegetables are classified into the second problem vegetables, the second normal vegetables, and the second undetermined vegetables.

9. The vegetable sorting and packaging method according to claim 8, characterized in that, The method further includes: Based on the last image in the image sequence, determine whether the second normal vegetable or the second undetermined vegetable exists within the periphery of the second problem vegetable; If so, the position of the second normal vegetable or the second undetermined vegetable is recorded as the third position coordinate, and the fourth position coordinate of the second problem vegetable is determined; Based on the third and fourth position coordinates, the operating mode of the second screening device is set; According to the operating mode, control the second screening device to keep the second normal vegetable or the second undetermined vegetable away from the second problem vegetable.

10. The vegetable sorting and packaging method according to claim 9, characterized in that, The method further includes: If the number of vegetables in the second problem is greater than one, calculate the shortest distance between the vegetables in the second problem; If the shortest distance is less than a preset distance threshold, then a collection path for the second problem vegetables is generated based on the fourth location coordinates. The second screening device is controlled to blow the second problematic vegetables onto the lower conveyor belt to the collection point according to the collection path.

Images