Marine product sorting system, sorting method and marine product sorting model training method

Abstract

The invention relates to the technical field of marine product sorting, and discloses a marine product sorting system and method and a marine product sorting model training method. The sorting device of the sorting system comprises a sorting manipulator, and the sorting manipulator is used for grabbing the marine products from the feeding device and placing the grabbed marine products in packaging boxes of the boxing device end to end in a distinguished mode, so that the marine products are placed in the packaging boxes end to end in a crossed mode. In other words, sorting and boxing operation of the marine products can be automatically achieved, the labor cost can be saved, and the production efficiency and the production quality can be improved. Besides, in order to meet the preset boxing strategy, the marine products are boxed in a head-to-tail distinguishing mode through the sorting mechanical arm, the marine products are automatically placed in a packaging box in a head-to-tail crossed mode, sorting and boxing operation of the marine products can be automatically achieved, and different boxing strategies of the marine products can be met.

Description

Technical Field

[0001] This application relates to the technical field of seafood sorting, and specifically relates to a seafood sorting system, a sorting method, and a method for training a seafood sorting model. Background Art

[0002] Seafood refers to edible fish / other marine products (such as squid, etc.). The production of seafood mostly follows the model of large-scale farming / fishing, centralized processing, and repackaging for sale. Specifically, the process steps of processing and packaging involve: weighing, classification / sorting, boxing, packing, etc. Currently in the industry, the above-mentioned several processes are usually carried out manually. The specific processes of the classification / sorting and boxing links are as follows: manually classify the products according to their sizes based on the weighing results, and then place the products into the packing boxes. Because it requires selection and placement, and different products have different standards and rules, it results in a large labor intensity, consumes a lot of time and manpower when carried out manually, and is also prone to errors and omissions, which greatly affects production efficiency and production quality. Summary of the Invention

[0003] This application provides a seafood sorting system, a sorting method, and a method for training a seafood sorting model, which can automatically realize the sorting and boxing operations of seafood.

[0004] This application provides a seafood sorting system. The sorting system includes: a feeding device for conveying the seafood to be sorted; a sorting device including a sorting manipulator for grasping the seafood from the feeding device; and a boxing device for conveying packing boxes. The sorting manipulator is configured to place the grasped seafood head-to-tail in the packing box, so that the seafood is arranged head-to-tail in the packing box.

[0005] In an embodiment of the present application, the seafood is arranged in a preset direction in the packing box, and in the preset direction, the head of at least one seafood is oppositely arranged to the tail of the adjacent seafood.

[0006] In an embodiment of the present application, the sorting device further includes: a vision detection device for acquiring image information of the seafood on the feeding device, so as to use a seafood sorting model to extract the head feature and the tail feature of the seafood based on the image information. The sorting manipulator is configured to place the grasped seafood head-to-tail in the packing box based on the head feature and the tail feature.

[0007] In an embodiment of the present application, the sorting system further includes: a film placing device for placing a diaphragm; the sorting manipulator is configured to grab the diaphragm from the film placing device and lay it above the seafood when it places the seafood in the packing box and fills one layer.

[0008] In an embodiment of the present application, the sorting manipulator includes a robotic arm, and a first adsorption assembly and a second adsorption assembly connected to the robotic arm. The first adsorption assembly is configured to grasp seafood in an adsorption manner, and the second adsorption assembly is configured to grasp diaphragms in an adsorption manner.

[0009] In an embodiment of the present application, the first adsorption assembly includes: a first connecting member connected to the robotic arm; a buffer rod connected to the first connecting member; and a first suction cup connected to the buffer rod. The first suction cup is used for adsorbing seafood, and the buffer rod is configured to make the first suction cup tend to move away from the robotic arm to closely adhere to the seafood.

[0010] In an embodiment of the present application, the second adsorption assembly includes: a second connecting member connected to the robotic arm; and a second suction cup connected to the second connecting member. The second suction cup is configured to adsorb diaphragms.

[0011] In an embodiment of the present application, the sorting manipulator further includes: a driving cylinder. The robotic arm is connected to the second adsorption assembly, and the second adsorption assembly is connected to the first adsorption assembly through the driving cylinder. The driving cylinder is configured to drive the first adsorption assembly to move so that the first adsorption assembly adsorbs seafood.

[0012] In an embodiment of the present application, the boxing device has a first horizontal direction and a second horizontal direction that are perpendicular to each other. The boxing device includes: a box conveyor line configured to convey packing boxes along the first horizontal direction; a first sensor disposed on the box conveyor line, and the first sensor is configured to sense that the packing box is conveyed to the boxing position; a first stopping mechanism disposed on the box conveyor line, and the first stopping mechanism is configured to stop the packing box in the first horizontal direction when the first sensor senses that the packing box is conveyed to the boxing position; and a positioning mechanism disposed on the box conveyor line, and the positioning mechanism is configured to fix the position of the packing box in the second horizontal direction when the first sensor senses that the packing box is conveyed to the boxing position.

[0013] In an embodiment of the present application, the boxing device further includes: a second sensor disposed on the box conveyor line, and the second sensor is configured to sense that the packing box is conveyed to the buffer position, and the buffer position is upstream of the boxing position; and a second stopping mechanism disposed on the box conveyor line, and the second stopping mechanism is configured to stop the packing box at the buffer position when the first sensor senses that there is a packing box at the boxing position and the second sensor senses that there is a packing box at the buffer position.

[0014] In an embodiment of the present application, the feeding device includes: a feeding conveyor line for conveying seafood to be sorted; and an identification device disposed at the feeding end of the feeding conveyor line. The identification device is configured to identify the quantity and specifications of the seafood at the feeding end to determine the boxing strategy for the seafood; wherein the boxing strategy includes the quantity of seafood in the packing box and the stacking layers.

[0015] In one embodiment of the present application, the feeding device includes: a feeding conveyor for conveying the seafood to be sorted; and an ice glazing machine located upstream of the feeding conveyor, and the seafood is conveyed to the feeding conveyor after being ice-glazed by the ice glazing machine.

[0016] Correspondingly, the present application also provides a method for sorting seafood. The sorting method includes: determining the head and / or tail of the seafood; controlling the sorting manipulator to grasp the seafood and placing the grasped seafood in the packaging box with the head and tail distinguished, so that the seafood is placed with the head and tail crossed in the packaging box.

[0017] In one embodiment of the present application, the step of determining the head and / or tail of the seafood includes: using a seafood sorting model to identify the head feature and the tail feature of the seafood; the step of placing the grasped seafood in the packaging box with the head and tail distinguished includes: based on the head feature and the tail feature, controlling the sorting manipulator to place the grasped seafood in the packaging box with the head and tail distinguished.

[0018] In one embodiment of the present application, the step of controlling the sorting manipulator to grasp the seafood includes: determining the grasping point coordinates based on the coordinates of the head feature and the coordinates of the tail feature, and guiding the sorting manipulator to grasp the seafood at the position represented by the grasping point coordinates.

[0019] In one embodiment of the present application, the step of using a seafood sorting model to identify the head feature and the tail feature of the seafood includes: obtaining an original image containing the seafood; preprocessing the original image to obtain a binary image; the contour of the seafood is marked in the binary image; the seafood sorting model identifies the head feature and the tail feature of the seafood based on the contour of the seafood in the binary image.

[0020] In one embodiment of the present application, after the step of obtaining the binary image, it further includes: determining the size of the seafood based on the contour of the seafood in the binary image, and determining whether the size of the current seafood meets a preset size range, so as to control the sorting manipulator to grasp the current seafood and place it in the packaging box when the size of the current seafood meets the preset size range.

[0021] In one embodiment of the present application, the step of controlling the sorting manipulator to grasp the seafood and placing the grasped seafood in the packaging box with the head and tail distinguished includes: controlling the sorting manipulator to place the seafood in the packaging box and cover one layer; controlling the sorting manipulator to grasp the diaphragm and lay it above the seafood, and then controlling the sorting manipulator to grasp the seafood and place it with the head and tail distinguished above the diaphragm.

[0022] Accordingly, the present application also provides a method for training a seafood sorting model. The training method includes: determining a training set; wherein the training set includes image information of seafood, and the image information is marked with the head feature and / or tail feature of the seafood; based on the training set, training the seafood sorting model so that the seafood sorting model can identify the head feature and / or tail feature of the seafood from the image information; wherein, the seafood sorting model is configured to guide the sorting manipulator to place the grasped seafood in the packing box with the head and tail distinguished, so that the seafood is placed with the head and tail crossed in the packing box.

[0023] In an embodiment of the present application, after the step that the seafood sorting model identifies the head feature and / or tail feature of the seafood from the image information, it includes: determining the grasping point coordinates based on the head feature and tail feature of the seafood, and the grasping point coordinates are configured for the sorting manipulator to grasp the seafood at the position represented by the grasping point coordinates.

[0024] In an embodiment of the present application, the step of determining the training set includes: acquiring the original image containing seafood; preprocessing the original image to obtain a binary image, wherein the binary image is marked with the head feature and tail feature of the seafood.

[0025] The beneficial effects of the present application are as follows: Different from the prior art, the present application provides a seafood sorting system, a sorting method and a method for training a seafood sorting model. The sorting device of the sorting system includes a sorting manipulator, and the sorting manipulator is used to grasp the seafood from the feeding device and place the grasped seafood in the packing box of the packing device with the head and tail distinguished, so that the seafood is placed with the head and tail crossed in the packing box. In other words, the present application can automatically realize the sorting and packing operations of seafood, which is beneficial to saving labor costs and improving production efficiency and quality. Moreover, in order to meet the predetermined packing strategy, the present application packs the seafood with the head and tail distinguished by the sorting manipulator, and automatically realizes the head and tail crossing placement of the seafood in the packing box. The present application can not only automatically realize the sorting and packing operations of seafood, but also meet different packing strategies of seafood. The head and tail of the seafood in the packing box of the present application are crossed, which can also improve the space utilization rate of the packing box. Description of the Drawings

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those skilled in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0027] Figure 1It is a schematic structural diagram of an embodiment of the sorting system for seafood in this application;

[0028] Figure 2 It is a schematic diagram of an embodiment of the seafood in this application placed in a packing box;

[0029] Figure 3 It is a schematic structural diagram of an embodiment of the sorting device in this application;

[0030] Figure 4 It is a schematic diagram of another embodiment of the seafood in this application placed in a packing box;

[0031] Figure 5 It is a schematic structural diagram of an embodiment of the sorting manipulator in this application;

[0032] Figure 6 It is a schematic structural diagram of an embodiment of the packing device in this application;

[0033] Figure 7 It is Figure 6 A schematic structural diagram of another perspective of the shown packing device;

[0034] Figure 8 It is Figure 6 A front view structural diagram of the shown packing device;

[0035] Figure 9 It is a schematic flowchart of an embodiment of the sorting method for seafood in this application;

[0036] Figure 10 It is a schematic flowchart of another embodiment of the sorting method for seafood in this application;

[0037] Figure 11 It is a practical machine detection diagram of an embodiment of the seafood in this application;

[0038] Figure 12 It is a schematic flowchart of an embodiment of the training method for the seafood sorting model in this application;

[0039] Figure 13 It is a schematic diagram of an embodiment of marking the head feature and tail feature of the seafood in the image information in this application.

[0040] Explanation of reference numerals:

[0041] 10 Loading device; 11 Loading conveyor line; 12 Identification device; 13 Loading end; 20 Sorting device; 21 Sorting manipulator; 211 Robot arm; 212 First adsorption component; 2121 First connecting piece; 2122 Buffer rod; 2123 First suction cup; 213 Second adsorption component; 2131 Second connecting piece; 2132 Second suction cup; 214 Driving cylinder; 22 Vision detection device; 23 Electrical control device; 24 Mounting bracket; 30 Boxing device; 31 Box conveyor line; 321 First sensor; 322 Second sensor; 323 Third sensor; 331 First stopping mechanism; 332 Second stopping mechanism; 333 Third stopping mechanism; 34 Positioning mechanism; 40 Seafood; 50 Packaging box; 60 Film placing device; 61 Diaphragm. Detailed implementation manners

[0042] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the protection scope of the present application. In addition, it should be understood that the specific implementation manners described herein are only used to illustrate and explain the present application, and are not used to limit the present application. In the present application, unless otherwise stated, the orientation words such as "upper", "lower", "left", and "right" usually refer to the upper, lower, left, and right in the actual use or working state of the device, specifically the drawing direction in the accompanying drawings.

[0043] In the present application, unless otherwise clearly defined and limited, the terms "connected", "connected", "stacked", etc. should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or integrated; it can be directly connected, or indirectly connected through an intermediate medium, and can be the internal connection of two components or the interaction relationship between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations.

[0044] The present application provides a sorting system, a sorting method, and a training method for a seafood sorting model of seafood, which will be described in detail below. It should be noted that the description order of the following embodiments does not limit the preferred order of the embodiments of the present application. And in the following embodiments, each embodiment is described with emphasis. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0045] Please refer to Figure 1 and Figure 2 , Figure 1 is a schematic structural diagram of an embodiment of the sorting system of the seafood of the present application, Figure 2It is a schematic diagram of an embodiment where the seafood products of the present application are placed in a packing box.

[0046] In one embodiment, the sorting system for seafood products includes a feeding device 10, a sorting device 20, and a packing device 30. The feeding device 10 is used to convey the seafood products 40 to be sorted. The seafood products 40 are conveyed to the sorting device 20 through the feeding device 10. The packing device 30 is used to convey the packing boxes 50. The sorting device 20 is used to grab the seafood products 40 from the feeding device 10 and place the grabbed seafood products 40 in the packing boxes 50, thus realizing the sorting and packing operations of the seafood products 40. In this embodiment, the sorting device 20 automatically grabs the seafood products 40 and packs them, which can automatically realize the sorting and packing operations of the seafood products 40, and is beneficial to saving labor costs and improving production efficiency and quality.

[0047] In one embodiment, the feeding device 10 includes a feeding conveyor line 11, and the feeding conveyor line 11 is used to convey the seafood products 40 to be sorted. The staff places the seafood products 40 that have been weighed and preliminarily classified by weight range at the feeding end 13 of the feeding conveyor line 11, and the seafood products 40 at the feeding end 13 are conveyed to the sorting device 20 through the feeding conveyor line 11. Optionally, the feeding conveyor line 11 can be in the form of a belt conveyor line or the like, which is not limited herein.

[0048] Furthermore, the feeding device 10 further includes an identification device 12, and the identification device 12 is arranged at the feeding end 13 of the feeding conveyor line 11. The identification device 12 is configured to identify the quantity and specifications (such as the size of the seafood products 40, etc.) of the seafood products 40 at the feeding end 13 to determine the packing strategy. The packing strategy includes the quantity of the seafood products 40 in the packing box 50 and the stacking layers. In this embodiment, the identification device 12 can identify the quantity and specifications of the seafood products 40 at the feeding end 13 through machine vision, and select a packing strategy that adapts to the current quantity and specifications of the seafood products 40 from the pre-set packing strategies, which can further improve the automation degree of the sorting and packing operations of the seafood products 40 in this embodiment, and is further beneficial to saving labor costs and improving production efficiency and quality. It can be understood that the identification device 12 is an optional device for the feeding conveyor line 11 in the embodiment of the present application. For the feeding method of manually identifying the quantity and specifications of the seafood products 40, the identification device 12 can be omitted, which is not limited herein.

[0049] In an alternative embodiment, an ice glazing machine (not shown in the figure) may be provided upstream of the loading conveyor line 11, and the seafood 40 is conveyed to the loading conveyor line 11 after being ice-glazed by the ice glazing machine. For the case where the seafood 40 has the requirement of chilled storage, in this embodiment, the manual ice-overloading method may be adopted. The staff places the seafood 40 into the ice glazing machine, and the seafood 40 is conveyed to the loading conveyor line 11 after being ice-glazed by the ice glazing machine. It can be understood that the ice glazing machine is an optional device for the loading conveyor line 11 in the embodiments of the present application. For the case where the seafood 40 does not have the requirement of chilled storage, the ice glazing machine may be omitted from the loading conveyor line 11 in the embodiments of the present application, which is not limited herein.

[0050] Please refer to Figure 3 , Figure 3 which is a schematic structural diagram of an embodiment of the sorting device of the present application.

[0051] In one embodiment, the sorting device 20 includes a sorting manipulator 21, and the sorting manipulator 21 is used to grab the seafood 40 from the loading device 10. The sorting manipulator 21 is configured to place the grabbed seafood 40 in the packing box 50 with the head and tail distinguished, so that the seafood 40 is arranged with the head and tail crossed in the packing box 50. In this embodiment, to meet the predetermined packing strategy, the sorting manipulator 21 packs the seafood 40 with the head and tail distinguished, and automatically realizes the arrangement of the seafood 40 with the head and tail crossed in the packing box 50. This embodiment can not only automatically realize the sorting and packing operations of the seafood 40, but also meet different packing strategies of the seafood 40. The arrangement of the seafood 40 with the head and tail crossed in the packing box 50 in this embodiment can also improve the space utilization rate of the packing box 50.

[0052] It should be noted that, as Figure 2 shown, the arrangement of the seafood 40 with the head and tail crossed in the packing box 50 can be specifically understood as that the seafood 40 is arranged along the preset direction P in the packing box 50, and in the preset direction P, the head of at least one seafood 40 is oppositely arranged to the tail of the adjacent seafood 40.

[0053] In one embodiment, the sorting device 20 further includes a vision detection device 22. The vision detection device 22 is used to obtain the image information of the seafood 40 on the loading device 10, so that the seafood 40 sorting model extracts the head feature and the tail feature based on the image information. The sorting manipulator 21 is configured to place the grabbed seafood 40 in the packing box 50 with the head and tail distinguished based on the head feature and the tail feature.

[0054] Specifically, the sorting device 20 further includes a mounting bracket 24. The sorting manipulator 21 and the vision detection device 22 are both mounted on the mounting bracket 24. The part of the feeding device 10 close to the sorting device 20 is the position where the sorting manipulator 21 grabs the seafood 40. The vision detection device 22 is specifically arranged above this part of the feeding device 10, so that the vision detection device 22 can obtain the image information of the seafood 40 to be grabbed by the sorting manipulator 21. Light sources can also be arranged around the mounting bracket 24, and the light sources play a role in supplementary lighting to ensure that the vision detection device 22 obtains clear images of the seafood 40.

[0055] In one embodiment, the sorting device 20 further includes an electrical control device 23. The electrical control device 23 uses a PLC (Programmable Logic Controller) as the main controller. According to the feedback of each sensor and the vision system, it controls components such as each motor and pneumatic unit of the device to automatically operate according to the parameters set manually, and is equipped with an industrial touch display screen as a man-machine interaction interface.

[0056] Please refer to Figure 4 , Figure 4 which is a schematic diagram of another embodiment of the seafood in the present application placed in the packing box.

[0057] In one embodiment, the sorting system further includes a film placing device 60, and the film placing device 60 is used to place the diaphragm 61. The packing strategy of the seafood 40 in this embodiment can specifically be that the seafood 40 and the diaphragm 61 are alternately stacked in the packing box 50, and the heads and tails of the seafood 40 in each layer are placed crosswise. In view of this, in this embodiment, the sorting manipulator 21 is configured such that when the sorting manipulator 21 places the seafood 40 in the packing box 50 and fills one layer, the sorting manipulator 21 grabs the diaphragm 61 from the film placing device 60 and lays it above the seafood 40, and then continues to grab the seafood 40 and place it above the diaphragm 61 until the quantity and / or weight of the seafood 40 in the packing box 50 meet the requirements.

[0058] By setting the film placing device 60 in this embodiment of the sorting system, it can adapt to the packing strategy of alternately stacking the seafood 40 and the diaphragm 61 in the packing box 50. Of course, the sorting system in this embodiment can also adapt to the packing strategy that does not require using the diaphragm 61 to separate adjacent layers of seafood 40. At this time, the film placing device 60 does not participate in the sorting and packing operations of the seafood 40. Of course, in other embodiments of the present application, for the case where it is not necessary to use the diaphragm 61 to separate adjacent layers of seafood 40, the sorting system can omit the film placing device 60 to simplify the composition of the sorting system.

[0059] Please refer to Figure 5 , Figure 5 which is a schematic structural diagram of one embodiment of the sorting manipulator of the present application.

[0060] In one embodiment, the sorting manipulator 21 includes a robotic arm 211, a first adsorption assembly 212 and a second adsorption assembly 213 connected to the robotic arm 211. The first adsorption assembly 212 is configured to grasp the seafood 40 in an adsorption manner, and the second adsorption assembly 213 is configured to grasp the diaphragm 61 in an adsorption manner.

[0061] Optionally, the robotic arm 211 can be a six-axis manipulator or the like. The robotic arm 211 has a high degree of flexibility, enabling the robotic arm 211 to drive the first adsorption assembly 212 and the second adsorption assembly 213 to move flexibly, ensuring that the first adsorption assembly 212 can grasp the seafood 40 as required and place the seafood 40 in the packing box 50 with the head and tail distinguished, so that the seafood 40 is placed in the packing box 50 with the head and tail crossed, and ensuring that the second adsorption assembly 213 can grasp the diaphragm 61 as required and lay it above the seafood 40.

[0062] The first adsorption assembly 212 includes a first connecting member 2121, a buffer rod 2122 and a first suction cup 2123. The first connecting member 2121 is connected to the robotic arm 211. The buffer rod 2122 is connected to the first connecting member 2121. The first suction cup 2123 is connected to the buffer rod 2122. The first adsorption assembly 212 forms a negative pressure in the first suction cup 2123 through a vacuum generator, so that the first suction cup 2123 is used to adsorb the seafood 40. The buffer rod 2122 is configured to make the first suction cup 2123 tend to move away from the robotic arm 211 to closely adhere to the seafood 40. The second adsorption assembly 213 includes a second connecting member 2131 and a second suction cup 2132. The second connecting member 2131 is connected to the robotic arm 211. The second suction cup 2132 is connected to the second connecting member 2131, and the second suction cup 2132 is configured to adsorb the diaphragm 61.

[0063] The sorting manipulator 21 further includes a driving cylinder 214. The robotic arm 211 is connected to the second adsorption assembly 213, and the second adsorption assembly 213 is connected to the first adsorption assembly 212 through the driving cylinder 214. The driving cylinder 214 is configured to drive the first adsorption assembly 212 to move so that the first adsorption assembly 212 adsorbs the seafood 40. During the process that the robotic arm 211 drives the first adsorption assembly 212 to adsorb the seafood 40, the driving cylinder 214 drives the whole first adsorption assembly 212 to move away from the robotic arm 211. When the first suction cup 2123 contacts the seafood 40, the driving cylinder 214 further drives the whole first adsorption assembly 212 to move away from the robotic arm 211. At this time, the buffer rod 2122 can provide an elastic restoring force to make the first suction cup 2123 tend to move away from the robotic arm 211 to closely adhere to the seafood 40, ensuring that the first suction cup 2123 reliably adsorbs the seafood 40.

[0064] Please refer to Figures 6 to 8 , Figure 6 which is a schematic structural diagram of an embodiment of the packing device of the present application, Figure 7 and Figure 6 which is a schematic structural diagram of another perspective of the packing device shown in Figure 8 and Figure 6 which is a front view structural diagram of the packing device shown in

[0065] In one embodiment, the packing device 30 has a first horizontal direction X and a second horizontal direction Y that are perpendicular to each other. The packing device 30 includes a box conveyor line 31, and the box conveyor line 31 is configured to convey the packing box 50 along the first horizontal direction X. An area of the box conveyor line 31 close to the sorting manipulator 21 has a packing position M. The sorting manipulator 21 transports the seafood 40 it grabs to the packing position M to place the seafood 40 in the packing box 50.

[0066] Specifically, the packing device 30 further includes a first sensor 321, a first stop mechanism 331, and a positioning mechanism 34. The first sensor 321 is disposed on the box conveyor line 31, and the first sensor 321 is configured to sense that the packing box 50 is conveyed to the packing position M. The first stop mechanism 331 is disposed on the box conveyor line 31, and the first stop mechanism 331 is configured to stop the packing box 50 in the first horizontal direction X when the first sensor 321 senses that the packing box 50 is conveyed to the packing position M. The positioning mechanism 34 is disposed on the box conveyor line 31, and the positioning mechanism 34 is configured to fix the position of the packing box 50 in the second horizontal direction Y when the first sensor 321 senses that the packing box 50 is conveyed to the packing position M. When the packing box 50 is conveyed to the packing position M via the box conveyor line 31, the first sensor 321 senses that the packing box 50 is conveyed to the packing position M. At this time, the first stop mechanism 331 stops the packing box 50 in the first horizontal direction X, and the positioning mechanism 34 fixes the position of the packing box 50 in the second horizontal direction Y. In this embodiment, the packing box 50 is fixed at the packing position M by the cooperation of the first stop mechanism 331 and the positioning mechanism 34, which can eliminate the position recognition operation of the packing box 50, simplify the sorting and packing operations of the seafood 40, and further facilitate improving the production efficiency.

[0067] Exemplarily, the box conveyor line 31 can be a roller conveyor line or the like. The positioning mechanism 34 pushes the packing box 50 along the second horizontal direction Y to the side edge of the box conveyor line 31 close to the feeding device 10 to fix the position of the packing box 50 in the second horizontal direction Y. Of course, in other embodiments of the present application, the positioning mechanism 34 can also fix the packing box 50 at other positions in the second horizontal direction Y, which is not limited herein. The first stopping mechanism 331 is located below the box conveyor line 31. When the first sensor 321 senses that the packing box 50 is conveyed to the packing position M, the first stopping mechanism 331 rises from bottom to top, so as to stop the packing box 50 in the first horizontal direction X.

[0068] In one embodiment, the box conveyor line 31 further has a buffer position N, and the buffer position N is located upstream of the packing position M. When there is a packing box 50 undergoing a packing operation at the packing position M, the upstream packing box 50 is temporarily stored at the buffer position N. Specifically, the packing device 30 further includes a second sensor 322 and a second stopping mechanism 332. The second sensor 322 is arranged on the box conveyor line 31, and the second sensor 322 is configured to sense that the packing box 50 is conveyed to the buffer position N. The second stopping mechanism 332 is arranged on the box conveyor line 31, and the second stopping mechanism is configured to stop the packing box 50 at the buffer position N when the first sensor 321 senses that there is a packing box 50 at the packing position M and the second sensor 322 senses that there is a packing box 50 at the buffer position N. When there is a packing box 50 undergoing a packing operation at the packing position M, the upstream packing box 50 is conveyed to the buffer position N via the box conveyor line 31. At this time, the second sensor 322 senses that a packing box 50 is conveyed to the buffer position N, and the second stopping mechanism stops the packing box 50 at the buffer position N in the first horizontal direction X, preventing the upstream packing box 50 from being further conveyed to the packing position M and affecting the ongoing packing operation.

[0069] In one embodiment, the box conveyor line 31 further has a sealing position L, and the sealing position L is located downstream of the packing position M. The packing box 50 that has completed the packing operation is conveyed to the sealing position L via the box conveyor line 31, and the packing box 50 that has completed the packing operation is packed and sealed at the sealing position L. Specifically, the packing device 30 further includes a third sensor 323 and a third stopping mechanism 333. The third sensor 323 is arranged on the box conveyor line 31, and the third sensor 323 is configured to sense that the packing box 50 is conveyed to the sealing position L. The third stopping mechanism 333 is arranged on the box conveyor line 31, and the third stopping mechanism is configured to stop the packing box 50 at the sealing position L when the third sensor 323 senses that there is a packing box 50 at the sealing position L, so as to pack and seal the packing box 50 that has completed the packing operation at the sealing position L.

[0070] In one embodiment, the packing device 30 further includes an adjusting plate 35. The adjusting plate 35 is movably arranged on the box conveyor line 31 and is located upstream of the packing position M. The distance between the adjusting plate 35 and the side edge of the box conveyor line 31 close to the feeding device 10 can be adjusted, and this distance can be set according to the size of the packing box 50. This distance can match the length or width of the packing box 50. For example, this distance is slightly larger than the length or width of the packing box 50. The packing box 50 needs to pass through the gap between the adjusting plate 35 and the side edge of the box conveyor line 31 with its length direction or width direction parallel to the first horizontal direction X, so as to achieve the effect of preventing the packing box 50 from being misaligned, ensure that the packing box 50 is conveyed to the packing position M in the correct posture, and further ensure that the seafood 40 is packed according to the established packing strategy. If the packing box 50 at the packing position M is skewed, it will cause the seafood 40 to be unable to be correctly placed in the packing box 50.

[0071] Optionally, the number of the adjusting plates 35 can be one or two or more. When the number of the adjusting plates 35 is two or more, the adjusting plates 35 are distributed at intervals along the first horizontal direction X.

[0072] The working process of the sorting system for seafood in the embodiment of the present application is described below.

[0073] The staff places the seafood products 40 that have been weighed and preliminarily classified by weight range at the loading end 13 of the loading conveyor 11. The diaphragms 61 and packing boxes 50 required for packing are placed at the film placing device 60 and the head end of the box conveyor 31. After confirming that the above materials are all in place, the staff selects the classification corresponding to the seafood products 40 on the human-machine interface of the electrical control device 23, sets the relevant production parameters, and presses the "Automatic Operation" button, and the whole system can start automatic operation. The loading end 13 of the loading conveyor 11 is provided with an identification device 12 for photographing and identifying the quantity of the seafood products 40 at the loading end 13 and feeding the result back to the PLC and the server database for subsequent packing schemes and statistical information. After confirming the quantity of the seafood products 40, the system conveys the seafood products 40 one by one to the position to be grabbed through the loading conveyor 11. After the seafood products 40 are conveyed to the position to be grabbed on the loading conveyor 11, they are photographed by the vision detection device 22 at this position and sent to the algorithm server, and are detected by the deep learning model (i.e., the seafood sorting model) to distinguish the head and tail and the size of the seafood products 40, and guide the sorting manipulator 21 to the coordinates where the seafood products 40 are located and adjust the angle to grab the seafood products 40. The sorting manipulator 21 will grab the seafood products 40 at the photographed position one by one and neatly place the seafood products 40 with the head and tail distinguished in the packing box 50 according to the set packing strategy. According to the pre-set packing strategy, after one layer is filled in the packing box 50, the sorting manipulator 21 will pause grabbing the seafood products 40, instead grab the diaphragm 61 for layering and place it above the layer of seafood products 40 that has been placed, and then continue to grab the seafood products 40. According to the pre-set packing strategy, after the quantity of products in the packing box 50 is sufficient, the box conveyor 31 will send the box of products to the sealing position L, and then the staff can choose manual sealing or send it to a sealing and packing machine, etc. for subsequent processing methods according to the actual situation. Thus, the sorting and packing of a box of seafood products 40 is completed, and then the sorting and packing operations of the next box of seafood products 40 continue.

[0074] The sorting system of the seafood products in the embodiment of the present application performs technological processes such as quantity identification, head and tail distinction, and classification packing of the seafood products through a pre-trained seafood sorting model and program parameter setting, and realizes fully automated production during the process, and the manual work only needs to assist in placing the production materials in the loading area. In the production and processing process of the seafood products in the embodiment of the present application, the human resources in the sorting / packing link are effectively liberated, and the production efficiency and stability of the seafood sorting and packing operations are significantly improved.

[0075] Please refer to Figure 9 , Figure 9 is a flow schematic diagram of an embodiment of the sorting method of the seafood products of the present application. The sorting method of the seafood products in this embodiment is based on the sorting system of the seafood products described in the above embodiment.

[0076] S101: Determine the head and / or tail of the seafood.

[0077] In this embodiment, the seafood is required to be placed head-to-tail in the packing box to improve the space utilization rate of the packing box. This embodiment automatically realizes the packing operation of the seafood and makes the seafood placed head-to-tail in the packing box. By determining the head and / or tail of the seafood, the seafood is placed in the packing box with the head and tail distinguished, so as to ensure that the seafood is placed head-to-tail in the packing box.

[0078] S102: Control the sorting manipulator to grab the seafood and place the grabbed seafood in the packing box with the head and tail distinguished, so that the seafood is placed head-to-tail in the packing box.

[0079] In this embodiment, after determining the head and / or tail of the seafood, control the sorting manipulator to grab the seafood and control the sorting manipulator to place the grabbed seafood in the packing box with the head and tail distinguished, so that the seafood is placed head-to-tail in the packing box.

[0080] By the above method, this embodiment can automatically realize the sorting and packing operations of the seafood, which is beneficial to saving labor costs and improving production efficiency and quality. Moreover, in order to meet the predetermined packing strategy, this embodiment packs the seafood with the head and tail distinguished by the sorting manipulator, and automatically realizes the head-to-tail placement of the seafood in the packing box. This embodiment can not only automatically realize the sorting and packing operations of the seafood, but also meet different packing strategies of the seafood. The head-to-tail placement of the seafood in the packing box of this embodiment can also improve the space utilization rate of the packing box.

[0081] Please refer to Figure 10 , Figure 10 which is a schematic flowchart of another embodiment of the sorting method of the seafood in this application.

[0082] S201: Obtain the original image containing the seafood.

[0083] In this embodiment, the seafood is required to be placed head-to-tail in the packing box to improve the space utilization rate of the packing box. In view of this, this embodiment obtains the original image containing the seafood to identify the head characteristics and tail characteristics of the seafood through machine vision and digital image processing technology, so that the seafood is placed in the packing box with the head and tail distinguished, and further ensures that the seafood is placed head-to-tail in the packing box.

[0084] S202: Preprocess the original image.

[0085] In this embodiment, after obtaining the original image containing seafood, the original image is preprocessed to obtain a binary image. The outline of the seafood is marked in the binary image, and then the head features and tail features of the seafood are identified through the outline of the seafood in the binary image.

[0086] Exemplarily, preprocessing the original image may include preprocessing processes such as color component extraction, image grayscale conversion, binarization, image enhancement, and contour extraction of the original image using digital image processing techniques to obtain a binary image of the seafood, which is not limited herein.

[0087] S203: Use the seafood sorting model to identify the head features and tail features of the seafood.

[0088] In this embodiment, there are differences in the shape, structure, size, etc. between the head and tail of the seafood. The differences between the head features and tail features reflect the differences between the head and tail of the seafood. Therefore, the head and tail of the seafood can be accurately identified through the head features and tail features of the seafood. Based on this, in this embodiment, after preprocessing the original image to obtain a binary image, the seafood sorting model is used to identify the head features and tail features of the seafood. In this embodiment, through the machine vision detection and recognition technology of the deep learning model (i.e., the seafood sorting model), the recognition accuracy of the head and tail of the seafood can be guaranteed.

[0089] Exemplarily, Figure 11 Taking the yellow croaker as an example, the head feature O1 and tail feature O2 of the seafood identified by the seafood sorting model are shown.

[0090] S204: Based on the head features and tail features, control the sorting manipulator to place the grabbed seafood in the packing box with the head and tail distinguished.

[0091] In this embodiment, after identifying the head features and tail features of the seafood, based on the head features and tail features, control the sorting manipulator to place the grabbed seafood in the packing box with the head and tail distinguished. This embodiment can automatically realize the sorting and packing operations of the seafood, which is beneficial to saving labor costs and improving production efficiency and quality. Moreover, in order to meet the predetermined packing strategy, in this embodiment, the seafood is packed with the head and tail distinguished by the sorting manipulator, and the head and tail of the seafood are automatically placed crosswise in the packing box. This embodiment can not only automatically realize the sorting and packing operations of the seafood, but also meet different packing strategies of the seafood. The head and tail of the seafood are placed crosswise in the packing box in this embodiment, which can also improve the space utilization rate of the packing box.

[0092] Further, the steps of controlling the sorting manipulator to grasp the seafood further include: determining the grasping point coordinates based on the coordinates of the head feature and the coordinates of the tail feature, and guiding the sorting manipulator to grasp the seafood at the position represented by the grasping point coordinates. There is a preset relative position relationship between the grasping point coordinates, the coordinates of the head feature, and the coordinates of the tail feature. After identifying the head feature and the tail feature of the seafood, the coordinates of the head feature and the coordinates of the tail feature can be determined, and then the grasping point coordinates can be determined according to the preset relative position relationship. The grasping point coordinates are defined as the position where the sorting manipulator grasps the seafood. Therefore, the sorting manipulator is guided to grasp the seafood at the position represented by the grasping point coordinates.

[0093] Further, based on the packing strategy of the seafood described in the above embodiment, the seafood and the diaphragm are alternately stacked in the packing box, and the heads and tails of the seafood in each layer are placed crosswise. In view of this, the steps of controlling the sorting manipulator to grasp the seafood and placing the grasped seafood in the packing box with the head and tail distinguished include: controlling the sorting manipulator to place the seafood in the packing box and cover a layer; controlling the sorting manipulator to grasp the diaphragm and lay it above the seafood, and then controlling the sorting manipulator to grasp the seafood and place it above the diaphragm with the head and tail distinguished.

[0094] In other embodiments of the present application, after the step of obtaining the binary image, it may further include: determining the size of the seafood based on the contour of the seafood in the binary image, and determining whether the size of the current seafood meets the preset size range, so as to control the sorting manipulator to grasp the current seafood and place it in the packing box when the size of the current seafood meets the preset size range. The preset size range should be understood as the requirements for the size of the seafood in the current packing operation. When the size of the current seafood meets the preset size range, then control the sorting manipulator to grasp the current seafood and place it in the packing box. If the size of the current seafood does not meet the preset size range, the current seafood will not be packed temporarily.

[0095] Please refer to Figure 12 , Figure 12 which is a schematic flowchart of an embodiment of the training method of the seafood sorting model of the present application.

[0096] S301: Determine the training set; where the training set includes image information of seafood, and the image information is marked with the head feature and / or tail feature of the seafood.

[0097] In this embodiment, the training set is determined, where the training set includes image information of seafood, and the image information is marked with the head feature and / or tail feature of the seafood. In this embodiment, the image information in the training set is used as the basic training data to train the seafood sorting model.

[0098] Further, obtain the original images containing seafood. Specifically, by simulating the shooting angle, lighting conditions, and background of the actual machine, a vision device collects a large amount of image information as the basic training data. Preprocess the original images to obtain binary images, where the binary images are marked with the head features and tail features of the seafood. In this embodiment, preprocessing such as normalization and image segmentation is performed on the collected training set, and the training set and the test set are divided. Both the training set and the test set include image information with seafood. The division rule of the training set and the test set can be to divide 80% of the basic training data into the training set and 20% into the test set. Moreover, in this embodiment, the head features and tail features of the seafood in the image information are manually marked. Figure 13 Exemplarily shows the marking of the head feature O1 and tail feature O2 of the seafood in the image information.

[0099] S302: Based on the training set, train the seafood sorting model so that the seafood sorting model can identify the head features and / or tail features of the seafood from the image information.

[0100] In this embodiment, after determining the training set, based on the training set, train the seafood sorting model so that the seafood sorting model can identify the head features and / or tail features of the seafood from the image information. The seafood sorting model is configured to guide the sorting manipulator to place the grabbed seafood with the head and tail distinguished in the packing box based on the head features and / or tail features of the seafood, so that the seafood is placed with the head and tail crossed in the packing box.

[0101] Further, the content of training the seafood sorting model also includes determining the grasping point coordinates based on the head features and tail features of the seafood. The grasping point coordinates are configured to enable the sorting manipulator to grab the seafood at the position represented by the grasping point coordinates. There is a preset relative position relationship between the grasping point coordinates and the coordinates of the head features and the coordinates of the tail features. By training the seafood sorting model to identify the head features and tail features of the seafood, and further training the seafood sorting model to determine the grasping point coordinates, it is used to guide the sorting manipulator to grab the seafood at the position represented by the grasping point coordinates in the future.

[0102] For example, use the YOLO model for deep learning training, set the number of iteration rounds to 200, and stop training when the loss function does not decrease for more than 20 rounds. Input the test set into the model to verify the training effect. Select the image information with poor recognition effect, re-mark the head features and tail features, and then add them to the training set. Repeat the above training steps to optimize the model to obtain the optimal model, that is, the seafood sorting model. The recognition accuracy of the final seafood sorting model can reach 99.3%. After the trained seafood sorting model is exported, it can be called by the seafood sorting system described in the above embodiment.

[0103] The above has introduced in detail the sorting system, sorting method and training method of the seafood sorting model provided by this application. Specific examples are used in this article to elaborate on the principle and implementation manner of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. A sorting system for seafood, characterized in that, The sorting system includes: A feeding device for conveying the seafood to be sorted; A sorting device including a sorting manipulator for grasping the seafood from the feeding device; and A boxing device for conveying packing boxes, wherein the sorting manipulator is configured to place the grasped seafood head-to-tail in the packing box, so that the seafood is arranged head-to-tail in the packing box.

2. The sorting system according to claim 1, wherein The seafood is arranged in the packing box along a preset direction, and in the preset direction, the head of at least one seafood is oppositely arranged to the tail of an adjacent seafood.

3. The sorting system according to claim 1 or 2, wherein The sorting device further includes: A vision detection device for acquiring image information of the seafood on the feeding device, so as to use a seafood sorting model to extract the head feature and the tail feature of the seafood based on the image information, and the sorting manipulator is configured to place the grasped seafood head-to-tail in the packing box based on the head feature and the tail feature.

4. The sorting system according to claim 1 or 2, wherein The sorting system further includes: A film placing device for placing a diaphragm; the sorting manipulator is configured to grasp the diaphragm from the film placing device and lay it above the seafood when it places the seafood in the packing box and fills a layer.

5. The sorting system according to claim 4, wherein The sorting manipulator includes a robotic arm and a first adsorption component and a second adsorption component connected to the robotic arm. The first adsorption component is configured to grasp the seafood in an adsorption manner, and the second adsorption component is configured to grasp the diaphragm in an adsorption manner.

6. The sorting system according to claim 5, wherein The first adsorption component includes: A first connecting member connected to the robotic arm; A buffer rod connected to the first connecting member; and A first suction cup connected to the buffer rod, the first suction cup being used for adsorbing the seafood, and the buffer rod being configured to make the first suction cup have a tendency to move away from the robotic arm to closely adhere to the seafood.

7. The sorting system according to claim 5, wherein The second adsorption component includes: A second connecting member connected to the robotic arm; and A second suction cup connected to the second connecting member, the second suction cup being configured to adsorb the diaphragm.

8. The sorting system according to claim 5, wherein The sorting manipulator further includes: A driving cylinder, the robotic arm is connected to the second adsorption component, and the second adsorption component is connected to the first adsorption component through the driving cylinder. The driving cylinder is configured to drive the first adsorption component to move so that the first adsorption component adsorbs the seafood.

9. The sorting system according to claim 1 or 2, wherein The boxing device has a first horizontal direction and a second horizontal direction that are perpendicular to each other, and the boxing device includes: A box conveyor line configured to convey the packing boxes along the first horizontal direction; A first sensor is disposed on the box conveyor line, and the first sensor is configured to sense that a packing box is conveyed to the packing position; A first stopping mechanism is disposed on the box conveyor line, and the first stopping mechanism is configured to stop the packing box in the first horizontal direction when the first sensor senses that the packing box is conveyed to the packing position; and A positioning mechanism is disposed on the box conveyor line, and the positioning mechanism is configured to fix the position of the packing box in the second horizontal direction when the first sensor senses that the packing box is conveyed to the packing position.

10. The sorting system according to claim 9, wherein The packing device further includes: A second sensor is disposed on the box conveyor line, and the second sensor is configured to sense that a packing box is conveyed to the buffer position, and the buffer position is upstream of the packing position; and A second stopping mechanism is disposed on the box conveyor line, and the second stopping mechanism is configured to stop the packing box at the buffer position when the first sensor senses that there is a packing box at the packing position and the second sensor senses that there is a packing box at the buffer position.

11. The sorting system according to claim 1 or 2, wherein The feeding device includes: A feeding conveyor line for conveying the seafood to be sorted; and An identification device is disposed at the feeding end of the feeding conveyor line, and the identification device is configured to identify the quantity and specifications of the seafood at the feeding end to determine the packing strategy for the seafood; wherein, the packing strategy includes the quantity of seafood in the packing box and the stacking layers.

12. The sorting system according to claim 1 or 2, wherein The feeding device includes: A feeding conveyor line for conveying the seafood to be sorted; and An ice glazing machine is located upstream of the feeding conveyor line, and the seafood is conveyed to the feeding conveyor line after being ice-glazed by the ice glazing machine.

13. A sorting method for seafood, characterized in that, The sorting method includes: Determining the head and / or tail of the seafood; Controlling the sorting manipulator to grasp the seafood and arranging the grasped seafood in the packing box with the head and tail distinguished, so that the seafood is arranged with the head and tail crossed in the packing box.

14. The sorting method according to claim 13, wherein The step of determining the head and / or tail of the seafood includes: Using a seafood sorting model to identify the head feature and the tail feature of the seafood; The step of arranging the grasped seafood in the packing box with the head and tail distinguished includes: Based on the head feature and the tail feature, controlling the sorting manipulator to arrange the grasped seafood in the packing box with the head and tail distinguished.

15. The sorting method according to claim 14, wherein The step of controlling the sorting manipulator to grasp the seafood includes: Based on the coordinates of the head feature and the coordinates of the tail feature, determining the grasping point coordinates and guiding the sorting manipulator to grasp the seafood at the position represented by the grasping point coordinates.

16. The sorting method according to claim 14, wherein The step of using a seafood sorting model to identify the head feature and the tail feature of the seafood includes: Obtain the original image containing seafood; Preprocess the original image to obtain a binary image; the contour of the seafood is marked in the binary image; The seafood sorting model identifies the head feature and the tail feature of the seafood based on the contour of the seafood in the binary image.

17. The sorting method according to claim 16, wherein, After the step of obtaining the binary image, the following steps are further included: Determine the size of the seafood based on the contour of the seafood in the binary image, and determine whether the size of the current seafood meets the preset size range, so as to control the sorting manipulator to grab the current seafood and place it in the packing box when the size of the current seafood meets the preset size range.

18. The sorting method according to claim 13, wherein, The step of controlling the sorting manipulator to grab the seafood and place the grabbed seafood in the packing box with the head and tail distinguished includes: Control the sorting manipulator to place the seafood in the packing box and cover a layer; Control the sorting manipulator to grab the diaphragm and lay it above the seafood, and then control the sorting manipulator to grab the seafood and place it with the head and tail distinguished above the diaphragm.

19. A training method for a seafood sorting model, characterized in that, The training method includes: Determine a training set; wherein the training set includes image information of seafood, and the head feature and / or tail feature of the seafood is marked in the image information; Based on the training set, train the seafood sorting model so that the seafood sorting model can identify the head feature and / or tail feature of the seafood from the image information; Wherein, the seafood sorting model is configured to guide the sorting manipulator to place the grabbed seafood in the packing box with the head and tail distinguished based on the head feature and / or tail feature of the seafood, so that the seafood is placed with the head and tail crossed in the packing box.

20. The training method according to claim 19, wherein, After the step that the seafood sorting model identifies the head feature and / or tail feature of the seafood from the image information, the following steps are included: Based on the head feature and tail feature of the seafood, determine the grasping point coordinates, and the grasping point coordinates are configured for the sorting manipulator to grab the seafood at the position represented by the grasping point coordinates.

21. The training method according to claim 19, wherein, The step of determining the training set includes: Obtain the original image containing seafood; Preprocess the original image to obtain a binary image, wherein the head feature and the tail feature of the seafood are marked in the binary image.

Images