Workpiece edge sealing method, device, electronic equipment and storage medium

Abstract

The embodiment of the present invention provides a workpiece edge sealing method, device, electronic device and storage medium, the method comprising: when the workpiece arrives at the first converging port, obtaining a first workpiece video captured by a first image acquisition device; using a preset first deep learning model to process the first workpiece video to identify the current posture of the workpiece; obtaining a digital twin model synchronized with the workpiece transmission, extracting the target posture in the digital twin model, wherein the digital twin model is generated from the basic information of the workpiece. The current posture is compared with the target posture, and if the current posture is consistent with the target posture, the workpiece is allowed to be edge sealed. The preset first deep learning model predicts the current posture of the workpiece through the first workpiece video captured by the first image acquisition device, and compares the target posture in the digital twin model. If the posture is consistent, it means that the workpiece has not undergone posture deviation, and the workpiece is allowed to be edge sealed, thereby avoiding edge banding machine failure or workpiece scrapping.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of edge sealing of panel furniture workpieces, and in particular to a workpiece edge sealing method, device, electronic equipment and storage medium. Background Art

[0002] Edge banding is a common process in panel furniture. Its function is to fit the edge banding tape to the four sides of the workpiece (such as wooden boards) to perform edge banding treatment. The main considerations are: first, the edges are rough around the workpiece after processing by the cutting equipment, and edge banding is used to improve the decorative effect of the workpiece; second, after the workpiece is processed, the moisture in the air enters the workpiece from the edge, causing the edge to expand and cause the workpiece to deform; third, to reduce the volatilization of harmful gases such as formaldehyde, so edge banding of the workpiece is an indispensable process in panel furniture. Among them, edge banding machine is a commonly used edge banding equipment in panel furniture manufacturing.

[0003] In the related technology, there are two main layout schemes for edge banding machines in mainstream automated production lines: 1. Use 4 edge banding machines, with every 2 edge banding machines arranged relatively as a group, and the two groups of edge banding machines are combined front and back to complete the edge banding of the four sides of the workpiece, such as Figure 1 2. Use two edge banding machines and a rotary production line to complete the edge banding of the four sides of the workpiece, such as Figure 2 Among them, the first solution takes up the least factory space, but the equipment cost is high, while the second solution has low equipment cost, and the rotary production line can also serve as a buffer area in the event of a failure, but it takes up a large factory space.

[0004] Since the cost-effectiveness of the second solution is higher than that of the first solution, the second solution is currently adopted. However, the second solution still has the following problems: for a workpiece that has been edge-banded once, for example, the workpiece has four edges, such as A, B, C, and D. Among them, two edge banding machines perform edge banding on two edges, such as A and C, of ​​the workpiece. The workpiece is subsequently transmitted to the rotary production line. During the process of passing through the rotary production line, the workpiece posture will be offset due to deformation of the workpiece or the non-levelness of the rotary production line mechanical mechanism. The workpiece posture offset will cause the edge banding machine to malfunction or the workpiece to be scrapped. Summary of the invention

[0005] In order to solve the technical problem that after the workpiece has been edge banded once, the subsequent workpiece is transmitted to a rotary production line, and in the process of passing through the rotary production line, the workpiece posture deviation may occur due to workpiece deformation or the rotary production line mechanical structure is not level, and the workpiece posture deviation may cause edge banding machine failure or workpiece scrapping, the embodiments of the present invention provide a workpiece edge banding method, device, electronic device and storage medium.

[0006] In a first aspect of an embodiment of the present invention, a workpiece edge sealing method is first provided, wherein a first image acquisition device is arranged at a first converging port of an automated production line and a rotary production line, and the method comprises:

[0007] When the workpiece reaches the first converging port, obtaining a first workpiece video captured by the first image acquisition device;

[0008] Processing the first workpiece video using a preset first deep learning model to identify a current posture of the workpiece;

[0009] Acquire a digital twin model that is synchronized with the workpiece transmission, and extract a target posture in the digital twin model, wherein the digital twin model is generated by basic information of the workpiece;

[0010] The current posture is compared with the target posture, and if the current posture is consistent with the target posture, edge sealing of the workpiece is allowed.

[0011] In an optional implementation, the processing of the first workpiece video by using a preset first deep learning model to identify the current posture of the workpiece includes:

[0012] Processing the first workpiece video using a preset first deep learning model to obtain a boundary of the workpiece, and fitting the boundary of the workpiece to obtain a current polygon;

[0013] The extracting the target posture in the digital twin model includes:

[0014] Extracting a target polygon in the digital twin model;

[0015] The comparing the current posture with the target posture, and allowing the workpiece to be edge-sealed if the current posture is consistent with the target posture, includes:

[0016] The current polygon is compared with the target polygon, and if the current polygon coincides with the target polygon, edge sealing of the workpiece is allowed.

[0017] In an optional embodiment, the method further comprises:

[0018] If the current polygon does not overlap with the target polygon, determining the posture offset angle of the workpiece;

[0019] If the posture deviation angle is greater than a preset angle threshold, the workpiece is determined to be abnormal and the workpiece is ejected.

[0020] In an optional embodiment, the method further comprises:

[0021] If the posture deviation angle is not greater than the preset angle threshold, adjusting the posture of the workpiece according to the posture deviation angle;

[0022] After the workpiece is adjusted in posture, the workpiece is transferred to the edge banding equipment through the automated production line for edge banding.

[0023] In an optional embodiment, a second image acquisition device is provided at a second converging port of the automated production line and the rotary production line, and the method further comprises:

[0024] When the workpiece reaches the second converging port, acquiring a second workpiece video acquired by the second image acquisition device;

[0025] Processing the second workpiece video using a preset second deep learning model to extract N processing areas of the workpiece;

[0026] Determine the processing areas with sealed edges among the N processing areas, and count the number of the processing areas with sealed edges;

[0027] If the number is greater than a preset threshold, it is determined that the workpiece has completed edge sealing, and the workpiece is discharged.

[0028] In an optional embodiment, the determining the edge-sealed processing area among the N processing areas includes:

[0029] For any of the processing areas, extract material information in the digital twin model, and input the processing area and the material information into the preset second deep learning model;

[0030] An output result of the preset second deep learning model is obtained, and if the output result meets a preset condition, it is determined that the processing area is a processing area with sealed edges.

[0031] In an optional implementation, the step of inputting the processing area and the material information into the preset second deep learning model includes:

[0032] Obtain a machine vision training set, and determine whether the material information exists in the machine vision training set;

[0033] If the material information exists in the machine vision training set, the processing area and the material information are input into the preset second deep learning model.

[0034] In an optional embodiment, the method further comprises:

[0035] If the material information does not exist in the machine vision training set, extracting M frames of images from the second workpiece video and adding them to the machine vision training set;

[0036] Based on the machine vision training set with images added, the preset second deep learning model is iteratively optimized.

[0037] In an optional embodiment, the method further comprises:

[0038] If the number is not greater than the preset threshold, it is determined that the workpiece has not completed edge banding, and the workpiece is transferred to the rotary production line.

[0039] In a second aspect of an embodiment of the present invention, a workpiece edge banding device is provided, wherein a first image acquisition device is arranged at a first converging port of an automated production line and a rotary production line, and the device comprises:

[0040] A video acquisition module, configured to acquire a first workpiece video acquired by the first image acquisition device when the workpiece arrives at the first converging port;

[0041] A posture recognition module, used to process the first workpiece video using a preset first deep learning model to recognize the current posture of the workpiece;

[0042] a posture extraction module, used to obtain a digital twin model synchronized with the workpiece transmission, and extract a target posture in the digital twin model, wherein the digital twin model is generated by basic information of the workpiece;

[0043] The workpiece edge sealing module is used to compare the current posture with the target posture, and if the current posture is consistent with the target posture, it is allowed to edge seal the workpiece.

[0044] In a third aspect of an embodiment of the present invention, there is further provided an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0045] Memory, used to store computer programs;

[0046] The processor is used to implement the workpiece edge sealing method described in the first aspect when executing the program stored in the memory.

[0047] In a fourth aspect of the embodiments of the present invention, a storage medium is further provided, wherein instructions are stored in the storage medium, and when the instructions are executed on a computer, the computer executes the workpiece edge sealing method described in the first aspect.

[0048] In a fifth aspect of the embodiments of the present invention, a computer program product comprising instructions is further provided. When the computer program product is run on a computer, the computer is enabled to execute the workpiece edge sealing method described in the first aspect.

[0049] The technical solution provided by the embodiment of the present invention is to set a first image acquisition device at the first converging port of the automated production line and the rotary production line, and when the workpiece arrives at the first converging port, obtain a first workpiece video acquired by the first image acquisition device, and use a preset first deep learning model to process the first workpiece video to identify the current posture of the workpiece, obtain a digital twin model synchronized with the workpiece transmission, and extract the target posture in the digital twin model, wherein the digital twin model is generated from the basic information of the workpiece, and the current posture is compared with the target posture. If the current posture is consistent with the target posture, the workpiece is allowed to be edge-sealed. By setting a first image acquisition device at the first converging port of the automated production line and the rotary production line, and using a preset first deep learning model to process the first workpiece video acquired by the first image acquisition device to identify the current posture of the workpiece, and compare it with the target posture in the digital twin model, if the postures of the two are consistent, it means that the workpiece has not undergone posture deviation, and the workpiece can be allowed to be edge-sealed, thereby avoiding edge banding machine failure or workpiece scrapping. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0052] Figure 1 A schematic diagram of a device layout shown in an embodiment of the present invention;

[0053] Figure 2 A schematic diagram of another device layout shown in an embodiment of the present invention;

[0054] Figure 3 A schematic diagram of another device layout shown in an embodiment of the present invention;

[0055] Figure 4 It is a schematic diagram of an implementation process of a workpiece edge sealing method shown in an embodiment of the present invention;

[0056] Figure 5 A schematic diagram of a digital twin model shown in an embodiment of the present invention;

[0057] Figure 6 It is a schematic diagram of synchronization of transmission between a wooden board and a digital twin model shown in an embodiment of the present invention;

[0058] Figure 7 A schematic diagram of another device layout shown in an embodiment of the present invention;

[0059] Figure 8 It is a schematic diagram of an implementation process of another workpiece edge sealing method shown in an embodiment of the present invention;

[0060] Fig. 9 It is a structural schematic diagram of a workpiece edge sealing device shown in an embodiment of the present invention;

[0061] Fig.10 It is a schematic diagram of the structure of an electronic device shown in an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0064] There are two main layout schemes for edge banding machines in mainstream automated production lines: 1. Use 4 edge banding machines, with every 2 edge banding machines arranged relatively as a group, and the two groups of edge banding machines are combined front and back to complete the edge banding of the four sides of the workpiece, such as Figure 1 As shown. Figure 1It can be seen that edge banding machine 1# and edge banding machine 2# only perform edge banding on the B and D surfaces of the workpiece each time, and edge banding machine 3# and edge banding machine 4# only perform edge banding on the A and C surfaces of the workpiece each time. Edge banding machine 1#, edge banding machine 2#, edge banding machine 3#, and edge banding machine 4# are connected in series along the feeding direction, and the overall layout is compact, saving plant space, but the equipment cost is relatively high.

[0065] 2. Use two edge banding machines and a rotary production line to complete the edge banding of the four sides of the workpiece, such as Figure 2 As shown. Figure 2 It can be seen that the workpiece coming from the feeding direction is edge banded once by edge banding machine 1# and edge banding machine 2#, and then diverted to the rotary production line at the converging port 2, and rotated 90 degrees at the same time. In the converging port 1, it is mixed with new workpieces for edge banding for the second time, and then diverted and discharged through the converging port 2 again. This solution has low equipment cost, and the rotary production line can also serve as a buffer area in the event of a failure, but it occupies a large factory space.

[0066] Since the cost-effectiveness of the second solution is higher than that of the first solution, the second solution is currently adopted. However, the second solution still has the following problems: for a workpiece that has been edge-banded once, for example, the workpiece has four edges, such as A, B, C, and D. Among them, two edge banding machines perform edge banding on two edges, such as A and C, of ​​the workpiece. The workpiece is subsequently transmitted to the rotary production line. During the process of passing through the rotary production line, the workpiece posture will be offset due to deformation of the workpiece or the non-levelness of the rotary production line mechanical mechanism. The workpiece posture offset will cause the edge banding machine to malfunction or the workpiece to be scrapped.

[0067] In order to solve such problems, the embodiment of the present invention sets a first image acquisition device at the first converging port of the automated production line and the rotary production line, for example, Figure 3 As shown, a camera 1 is set at the intersection 1 of the automated production line and the rotary production line. Based on this, Figure 4 FIG. 1 is a schematic diagram of an implementation flow of a workpiece edge sealing method provided by an embodiment of the present invention. The method can be applied to a processor and may specifically include the following steps:

[0068] S401, when the workpiece arrives at the first converging port, obtaining a first workpiece video captured by the first image acquisition device.

[0069] In an embodiment of the present invention, for the workpiece coming in from the feeding direction, it is edge banded once by edge banding machine 1# and edge banding machine 2#, and then diverted to the rotary production line at the second converging port. At the same time, it rotates 90 degrees. When the workpiece reaches the first converging port, machine vision is used to identify whether the workpiece has any posture deviation during the rotation process, and the first workpiece video captured by the first image acquisition device is obtained based on this.

[0070] For example, in an embodiment of the present invention, a wooden board coming in from the feeding direction is edge banded once by edge banding machine 1# and edge banding machine 2#, and then diverted to the rotary production line at the converging port 2. At the same time, it rotates 90 degrees. When the wooden board reaches the converging port 1, machine vision is used to identify whether the wooden board has any posture deviation during the rotation process, and the first wooden board video captured by camera 1 is obtained based on this.

[0071] It should be noted that the workpiece may be, for example, a wooden board or a board of other materials that needs edge sealing, and the embodiment of the present invention does not limit this. In addition, the first image acquisition device may be, for example, a conventional camera or other image sensors, and the embodiment of the present invention does not limit this.

[0072] S402: Process the first workpiece video using a preset first deep learning model to identify a current posture of the workpiece.

[0073] In an embodiment of the present invention, when the workpiece arrives at the first converging port, a first workpiece video is captured by a first image acquisition device, and the first workpiece video can be processed using a preset first deep learning model to identify the current posture of the workpiece.

[0074] It should be noted that the first deep learning model preset here can specifically be a Lapnet network, and of course it can also be a CNN algorithm, a YOLO algorithm, etc. The embodiment of the present invention is not limited to this and can be selected according to the specific situation.

[0075] For example, in an embodiment of the present invention, when the wooden board reaches the converging port 1, a first wooden board video is captured by camera 1, and the first wooden board video can be processed using a preset first deep learning model to identify the current posture of the wooden board.

[0076] In an embodiment of the present invention, a preset first deep learning model can be used to process the first workpiece video to obtain the boundary of the workpiece, and the boundary of the workpiece is fitted to obtain a current polygon, so as to represent the current posture of the workpiece with the current polygon.

[0077] For example, in an embodiment of the present invention, the first workpiece video can be processed using a preset first deep learning model to obtain the boundary of the wooden board, and the boundary of the wooden board can be fitted to obtain the current quadrilateral, and the current quadrilateral can be used to represent the current posture of the wooden board.

[0078] S403, acquiring a digital twin model that is synchronized with the workpiece transmission, and extracting a target posture in the digital twin model, wherein the digital twin model is generated based on basic information of the workpiece.

[0079] In the embodiment of the present invention, when a workpiece enters the automated production line from the feed port, basic information of the workpiece is obtained by scanning the QR code on the workpiece, for example, the basic information includes length, width, height, color, material, etc.

[0080] The embodiment of the present invention obtains the basic information of the workpiece, thanks to the high performance and high computing power of industrial computers to support the real-time generation of the digital twin model of the current workpiece in the corresponding UI interface, such as Figure 5 As shown, key parameter information is displayed, as shown in Table 1 below.

[0081]

[0082] Table 1

[0083] This digital twin model only displays the basic length, width and height in the user interface, but contains all the processing information (length, width and height, edge banding type, drilling position, drill bit size, etc.) within the control system, and the digital twin model is synchronized with the actual workpiece transmission. When the workpiece needs to enter the equipment for processing, the control system extracts the processing information corresponding to the digital twin model and sends it to the equipment for processing.

[0084] Based on this, when the workpiece arrives at the first convergence port, the digital twin model synchronized with the workpiece transmission is obtained, and the target posture in the digital twin model is extracted. Here, the digital twin model is synchronized with the workpiece transmission, which means that in the actual factory, when the workpiece reaches the first convergence port, the corresponding digital twin model is synchronously transmitted to the first convergence port on the UI interface.

[0085] Among them, in an embodiment of the present invention, when the workpiece arrives at the first converging port, a digital twin model synchronized with the workpiece transmission is obtained, the target polygon in the digital twin model is extracted, and the target posture in the digital twin model is represented by the target polygon.

[0086] For example, in an embodiment of the present invention, when the wooden board reaches the convergence port 1, the digital twin model of the wooden board is synchronized with the transmission of the wooden board. At this time, on the UI interface, the digital twin model of the wooden board is displayed as having also reached the convergence port 1, thereby obtaining the digital twin model synchronized with the transmission of the wooden board 1, extracting the target quadrilateral in the digital twin model, and using the target quadrilateral to represent the target posture in the digital twin model.

[0087] S404, comparing the current posture with the target posture, and if the current posture is consistent with the target posture, edge sealing of the workpiece is allowed.

[0088] In an embodiment of the present invention, the current posture of the workpiece and the target posture in the digital twin model are compared. If the current posture is consistent with the target posture, it means that the workpiece has not undergone posture deviation during the rotation of the rotary production line. At this time, edge banding of the workpiece can be allowed. Specifically, the workpiece is transmitted to edge banding machine 1# and edge banding machine 2# through an automated production line for edge banding, thereby avoiding edge banding machine failure or workpiece scrapping.

[0089] Among them, for the current posture of the workpiece, the current posture of the workpiece is represented by the current polygon, and for the target posture in the digital twin model, the target posture in the digital twin model is represented by the target polygon. The current polygon is compared with the target polygon. If the current polygon coincides with the target polygon, edge banding of the workpiece is allowed. Specifically, the workpiece is transmitted to edge banding machine 1# and edge banding machine 2# through an automated production line for (secondary) edge banding.

[0090] For example, in the embodiment of the present invention, for the current posture of the wooden board, the current posture of the wooden board is represented by the current quadrilateral, and for the target posture in the digital twin model, the target quadrilateral is used to represent the target posture in the digital twin model, such as Figure 6 As shown, the current quadrilateral is compared with the target quadrilateral, by Figure 6 It can be seen that the current quadrilateral coincides with the target quadrilateral, which means that the wooden board has not undergone posture deviation during the rotation of the rotary production line. At this time, the wooden board can be transmitted to edge banding machine 1# and edge banding machine 2# through the automated production line for secondary edge banding.

[0091] In addition, if the current polygon does not overlap with the target polygon, it means that the workpiece has undergone posture deviation during the rotation of the rotary production line. Specifically, it may be because the workpiece is deformed or the rotary production line mechanical structure is not level, causing the workpiece posture deviation. At this time, the posture deviation angle of the workpiece can be determined. Among them, the current polygon can be compared with the target polygon to obtain the posture deviation angle of the workpiece. If the posture deviation angle is greater than the preset angle threshold, the workpiece is determined to be abnormal and the workpiece is discharged. Here, discharge means discharging the workpiece as waste.

[0092] For example, if the current quadrilateral of the wooden board does not overlap with the target quadrilateral in the digital twin model, it means that the wooden board has undergone posture deviation during the rotation of the rotary production line. Specifically, it may be because the wooden board is deformed or the rotary production line mechanical mechanism is not level, causing the wooden board to have posture deviation. At this time, the posture deviation angle of the wooden board can be determined, and it can be judged whether the posture deviation angle is greater than the preset angle threshold. If the posture deviation angle is greater than the preset angle threshold, it means that the posture of the wooden board cannot be adjusted. At this time, it can be determined that the wooden board is abnormal and the wooden board is discharged as waste.

[0093] For the posture offset angle of the workpiece, if the posture offset angle is not greater than the preset angle threshold, it means that the posture of the workpiece can be adjusted. At this time, the posture of the workpiece is adjusted according to the posture offset angle. Specifically, the posture offset angle can be updated to the digital twin model, and the control system adjusts the posture of the workpiece according to the posture offset angle in the digital twin model. After the posture of the workpiece is adjusted, the workpiece is transferred to the edge banding equipment through the automated production line for edge banding, which can also avoid edge banding machine failure or workpiece scrapping.

[0094] For example, for the posture offset angle of the wooden board, if the posture offset angle is not greater than the preset angle threshold, it means that the posture of the wooden board can be adjusted. At this time, the posture offset angle can be updated to the digital twin model, and the control system adjusts the posture of the wooden board according to the posture offset angle in the digital twin model. After the posture of the wooden board is adjusted, the wooden board is transmitted to edge banding machine 1# and edge banding machine 2# through the automated production line for secondary edge banding.

[0095] It should be noted that the posture of the workpiece can be adjusted by using a posture correction mechanical structure, which has a simple principle and a wide variety of types. For example, mechanical grippers, pneumatic stop rods and other structures can be used to achieve similar effects, and the embodiments of the present invention are not limited to this.

[0096] Through the above description of the technical solution provided by the embodiment of the present invention, a first image acquisition device is set at the first converging port of the automated production line and the rotary production line, and when the workpiece arrives at the first converging port, a first workpiece video captured by the first image acquisition device is obtained, and the first workpiece video is processed using a preset first deep learning model to identify the current posture of the workpiece, obtain a digital twin model synchronized with the workpiece transmission, and extract the target posture in the digital twin model, wherein the digital twin model is generated from the basic information of the workpiece, and the current posture is compared with the target posture. If the current posture is consistent with the target posture, edge sealing of the workpiece is allowed.

[0097] By setting a first image acquisition device at the first converging port of the automated production line and the rotary production line, the first workpiece video captured by the first image acquisition device is processed using a preset first deep learning model to identify the current posture of the workpiece and compare it with the target posture in the digital twin model. If the postures of the two are consistent, it means that there is no posture deviation in the workpiece, and the workpiece can be allowed to be edge banded, thereby avoiding edge banding machine failure or workpiece scrapping.

[0098] In addition, in the embodiment of the present invention, for the second local solution, the following problem still exists: it is difficult to distinguish and identify and divert the workpieces that have been edge-banded once and the new workpieces when mixed through the edge banding machine. If the diversion is wrong, the semi-finished products will flow into the next process. For example, after a wooden board has been edge-banded once, it is difficult to distinguish and identify and divert it from the new wooden board through the edge banding machine subsequently. It is possible that the wooden board that has been edge-banded once (i.e., the semi-finished product) will be diverted to the next process.

[0099] In order to solve such problems, the embodiment of the present invention sets a second image acquisition device at the second converging port of the automated production line and the rotary production line, for example, Figure 7 As shown, a camera 2 is set at the intersection 2 of the automated production line and the rotary production line. Based on this, Figure 8 FIG. 1 is a schematic diagram of an implementation flow of another workpiece edge sealing method provided by an embodiment of the present invention. The method can be applied to a processor and may specifically include the following steps:

[0100] S801, when the workpiece arrives at the second converging port, obtaining a second workpiece video captured by the second image acquisition device.

[0101] In an embodiment of the present invention, for the workpiece coming in from the feeding direction, it is edge banded once by edge banding machine 1# and edge banding machine 2#. When the workpiece reaches the second converging port, the workpiece is diverted by machine vision to determine whether the workpiece needs to be transferred to the rotary production line, and the second workpiece video captured by the second image acquisition device is obtained accordingly.

[0102] For example, in an embodiment of the present invention, for a wooden board coming in from the feeding direction, it is edge banded once by edge banding machine 1# and edge banding machine 2#. When the wooden board reaches the convergence port 2, the wooden board is classified by machine vision to determine whether the wooden board needs to be transferred to the rotary production line, and the second wooden board video captured by camera 2 is obtained accordingly.

[0103] It should be noted that the second image acquisition device may be, for example, a conventional camera or other image sensors, which is not limited in the embodiment of the present invention.

[0104] S802: Process the second workpiece video using a preset second deep learning model to extract N processing areas of the workpiece.

[0105] In an embodiment of the present invention, when the workpiece arrives at the second converging port, a second workpiece video is captured by a second image acquisition device, and the second workpiece video can be processed by a preset second deep learning model to extract N processing areas of the workpiece, where the processing area is the edge of the workpiece that needs to be sealed, such as Figure 1 The four sides A, B, C, and D are shown.

[0106] It should be noted that the second deep learning model preset here can specifically be a Lapnet network, and of course it can also be a CNN algorithm, a YOLO algorithm, etc. The embodiment of the present invention is not limited to this and can be selected according to the specific situation.

[0107] For example, in an embodiment of the present invention, when the wooden board reaches the convergence port 2, a second wooden board video is captured by camera 2, and the second wooden board video can be processed using a preset second deep learning model to extract four processing areas of the wooden board, i.e., four edges that need to be edge sealed.

[0108] S803, determining the processing areas with sealed edges among the N processing areas, and counting the number of the processing areas with sealed edges.

[0109] In the embodiment of the present invention, for N processing areas of a workpiece, the processing areas with edge sealing among the N processing areas are determined, and the number of the processing areas with edge sealing is counted.

[0110] Among them, for any processing area, the material information in the digital twin model is extracted, the processing area and the material information are input into the preset second deep learning model, and the output result of the preset second deep learning model is obtained. If the output result meets the preset conditions, the processing area is determined to be a sealed processing area.

[0111] For example, for any of the four processing areas, extract the material information in the digital twin model of the wood board, assuming that the particle board is yellow catalpa, input the processing area and material information into the preset second deep learning model, obtain the output result of the preset second deep learning model, and if the output result is 1, determine that the processing area is an edge-sealed processing area. Different output results represent different meanings, as shown in Table 2 below.

[0112] Output meaning 1 Edge banded 0 Unsealed

[0113] Table 2

[0114] It should be noted that in the embodiment of the present invention, the features of the sealed edges and the unsealed edges are different. The processing area and material information are input into the preset second deep learning model in order to combine the material information in the digital twin model, which can further improve the prediction accuracy.

[0115] Among them, a machine vision training set is obtained, and it is determined whether the material information exists in the machine vision training set. If the material information exists in the machine vision training set, the processing area and the material information are input into a preset second deep learning model.

[0116] In addition, if the material information does not exist in the machine vision training set, M frames of images are extracted from the second workpiece video and added to the machine vision training set. Based on the machine vision training set with the images added, the preset second deep learning model is iteratively optimized.

[0117] For example, if the material information, assuming that the particle board yellow catalpa wood does not exist in the machine vision training set, M frames of images (that is, many characteristic photos of the current wood board) are extracted from the second wood board video and added to the machine vision training set, based on the machine vision training set with added images, through unsupervised learning (such as "restricted Boltzmann machine") or supervised learning, the preset second deep learning model is iteratively optimized to improve the recognition accuracy.

[0118] S804: If the number is greater than a preset threshold, it is determined that the workpiece has completed edge sealing, and the workpiece is discharged.

[0119] In the embodiment of the present invention, for the number of edge-sealed processing areas, if the number is greater than a preset threshold, it is determined that the workpiece has completed edge sealing, and the workpiece is discharged. If the number is not greater than the preset threshold, it is determined that the workpiece has not completed edge sealing, and the workpiece is transferred to the rotary production line. In this way, the workpiece can be diverted to ensure that the semi-finished product will not be diverted to the next process.

[0120] For example, in an embodiment of the present invention, for the number of processing areas that have been edge-sealed, if the number is greater than 2, it is determined that the wooden material has been edge-sealed twice and updated to the digital twin model. At this time, the wooden board is a finished product, so it can be determined that the wooden board has completed edge sealing, and the subsequent control system discharges the wooden board.

[0121] For example, in an embodiment of the present invention, for the number of processing areas that have been edge-sealed, if the number is not greater than 2, it is determined that the wooden board has been edge-sealed once and updated to the digital twin model. At this time, the wooden board is a semi-finished product, so it can be determined that the wooden board has not completed edge sealing. The subsequent control system transfers the wooden board to the rotary production line.

[0122] Through the above description of the technical solution provided by the embodiment of the present invention, a second image acquisition device is set at the second converging port of the automated production line and the rotary production line. When the workpiece arrives at the second converging port, a second workpiece video acquired by the second image acquisition device is obtained, and the second workpiece video is processed using a preset second deep learning model to extract N processing areas of the workpiece, determine the processing areas with edge sealing among the N processing areas, and count the corresponding number of processing areas with edge sealing. If the number is greater than a preset threshold, it is determined that the workpiece has completed edge sealing, and the workpiece is discharged. If the number is not greater than the preset threshold, it is determined that the workpiece has not completed edge sealing, and the workpiece is transferred to the rotary production line.

[0123] A second image acquisition device is set at the second converging port of the automated production line and the rotary production line, and a preset second deep learning model is used to process the second workpiece video acquired by the second image acquisition device to extract N processing areas of the workpiece, determine the edge-sealed processing areas among the N processing areas, and count the corresponding number of edge-sealed processing areas. If the number is greater than a preset threshold, it is determined that the workpiece has completed edge sealing, and the workpiece is discharged. If the number is not greater than the preset threshold, it is determined that the workpiece has not completed edge sealing, and the workpiece is transferred to the rotary production line. In this way, the diversion of the workpiece can be achieved, ensuring that the semi-finished products will not be diverted to the next process.

[0124] Corresponding to the above method embodiment, the embodiment of the present invention also provides a workpiece edge sealing device, such as Fig. 9 As shown, the device may include: a video acquisition module 910 , a posture recognition module 920 , a posture extraction module 930 , and a workpiece edge sealing module 940 .

[0125] A video acquisition module 910 is used to acquire a first workpiece video acquired by the first image acquisition device when the workpiece arrives at the first converging port;

[0126] A posture recognition module 920 is used to process the first workpiece video using a preset first deep learning model to recognize the current posture of the workpiece;

[0127] a posture extraction module 930, for acquiring a digital twin model synchronized with the workpiece transmission, and extracting a target posture in the digital twin model, wherein the digital twin model is generated from basic information of the workpiece;

[0128] The workpiece edge sealing module 940 is used to compare the current posture with the target posture, and if the current posture is consistent with the target posture, it is allowed to perform edge sealing on the workpiece.

[0129] The embodiment of the present invention further provides an electronic device, such as Fig.10 As shown, it includes a processor 101, a communication interface 102, a memory 103 and a communication bus 104, wherein the processor 101, the communication interface 102, and the memory 103 communicate with each other through the communication bus 104.

[0130] Memory 103, used for storing computer programs;

[0131] The processor 101 is used to execute the program stored in the memory 103, and implements the following steps:

[0132] When the workpiece arrives at the first converging port, a first workpiece video captured by the first image acquisition device is obtained; the first workpiece video is processed using a preset first deep learning model to identify the current posture of the workpiece; a digital twin model synchronized with the workpiece transmission is obtained, and a target posture in the digital twin model is extracted, wherein the digital twin model is generated from basic information of the workpiece; the current posture is compared with the target posture, and if the current posture is consistent with the target posture, edge sealing of the workpiece is allowed.

[0133] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0134] The communication interface is used for communication between the above electronic device and other devices.

[0135] The memory may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0136] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0137] In another embodiment provided by the present invention, a storage medium is provided. The storage medium stores instructions. When the instructions are executed on a computer, the computer executes the workpiece edge sealing method described in any one of the above embodiments.

[0138] In another embodiment provided by the present invention, a computer program product including instructions is also provided, and when the computer is run on the computer, the computer is enabled to execute the workpiece edge sealing method described in any one of the above embodiments.

[0139] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a storage medium, or transmitted from one storage medium to another storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk Solid State Disk (SSD)), etc.

[0140] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0141] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0142] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A workpiece edge sealing method, characterized in that: A first image acquisition device is arranged at a first converging port of the automated production line and the rotary production line, and the method comprises: When the workpiece reaches the first converging port, obtaining a first workpiece video captured by the first image acquisition device; Processing the first workpiece video using a preset first deep learning model to identify a current posture of the workpiece; Acquire a digital twin model that is synchronized with the workpiece transmission, and extract a target posture in the digital twin model, wherein the digital twin model is generated by basic information of the workpiece; Comparing the current posture with the target posture, and allowing the workpiece to be edge-sealed if the current posture is consistent with the target posture; The step of processing the first workpiece video by using a preset first deep learning model to identify the current posture of the workpiece includes: Processing the first workpiece video using a preset first deep learning model to obtain a boundary of the workpiece, and fitting the boundary of the workpiece to obtain a current polygon; The extracting the target posture in the digital twin model includes: Extracting a target polygon in the digital twin model; The comparing the current posture with the target posture, and allowing the workpiece to be edge-sealed if the current posture is consistent with the target posture, includes: The current polygon is compared with the target polygon, and if the current polygon coincides with the target polygon, edge sealing of the workpiece is allowed.

2. The method according to claim 1, characterized in that The method further comprises: If the current polygon does not overlap with the target polygon, determining the posture offset angle of the workpiece; If the posture deviation angle is greater than a preset angle threshold, the workpiece is determined to be abnormal and the workpiece is ejected.

3. The method according to claim 2, characterized in that The method further comprises: If the posture deviation angle is not greater than the preset angle threshold, adjusting the posture of the workpiece according to the posture deviation angle; After the posture of the workpiece is adjusted, the workpiece is transferred to the edge banding equipment through the automated production line for edge banding.

4. The method according to claim 1, characterized in that: A second image acquisition device is arranged at a second converging port of the automated production line and the rotary production line, and the method further comprises: When the workpiece reaches the second converging port, acquiring a second workpiece video acquired by the second image acquisition device; Processing the second workpiece video using a preset second deep learning model to extract N processing areas of the workpiece; Determine the processing areas with sealed edges among the N processing areas, and count the number of the processing areas with sealed edges; If the number is greater than a preset threshold, it is determined that the workpiece has completed edge sealing, and the workpiece is discharged.

5. The method according to claim 4, characterized in that The step of determining the edge-sealed processing area among the N processing areas comprises: For any of the processing areas, extract material information in the digital twin model, and input the processing area and the material information into the preset second deep learning model; An output result of the preset second deep learning model is obtained, and if the output result meets a preset condition, it is determined that the processing area is a processing area with sealed edges.

6. The method according to claim 5, characterized in that The step of inputting the processing area and the material information into the preset second deep learning model includes: Obtain a machine vision training set, and determine whether the material information exists in the machine vision training set; If the material information exists in the machine vision training set, the processing area and the material information are input into the preset second deep learning model.

7. The method according to claim 6, characterized in that The method further comprises: If the material information does not exist in the machine vision training set, extracting M frames of images from the second workpiece video and adding them to the machine vision training set; Based on the machine vision training set with images added, the preset second deep learning model is iteratively optimized.

8. The method according to claim 4, characterized in that The method further comprises: If the number is not greater than the preset threshold, it is determined that the workpiece has not completed edge banding, and the workpiece is transferred to the rotary production line.

9. A workpiece edge sealing device, characterized in that: A first image acquisition device is arranged at the first converging port of the automated production line and the rotary production line, and the device comprises: A video acquisition module, configured to acquire a first workpiece video acquired by the first image acquisition device when the workpiece arrives at the first converging port; A posture recognition module, used to process the first workpiece video using a preset first deep learning model to recognize the current posture of the workpiece; a posture extraction module, used to obtain a digital twin model synchronized with the workpiece transmission, and extract a target posture in the digital twin model, wherein the digital twin model is generated by basic information of the workpiece; A workpiece edge sealing module, used for comparing the current posture with the target posture, and allowing the workpiece to be edge sealed if the current posture is consistent with the target posture; The step of processing the first workpiece video by using a preset first deep learning model to identify the current posture of the workpiece includes: Processing the first workpiece video using a preset first deep learning model to obtain a boundary of the workpiece, and fitting the boundary of the workpiece to obtain a current polygon; The extracting the target posture in the digital twin model includes: Extracting a target polygon in the digital twin model; The comparing the current posture with the target posture, and allowing the workpiece to be edge-sealed if the current posture is consistent with the target posture, includes: The current polygon is compared with the target polygon, and if the current polygon coincides with the target polygon, edge sealing of the workpiece is allowed.

10. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, for implementing the method steps described in any one of claims 1 to 8 when executing a program stored in a memory.

11. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images