Visual inspection method for high-density layout payload product connection correctness

Abstract

The application discloses a kind of high-density layout load cabin product connection correctness verification methods. By establishing single machine connection relationship in advance, complete three-dimensional design model and early design such as identification system, use image acquisition equipment to collect two-dimensional and three-dimensional images of cabin single machine and waveguide, cable and other products and use image library to manage, compare the information contained in the image with connection relationship table and three-dimensional design model by machine vision method, verify single machine installation position correctness and waveguide, cable connection relationship correctness, so as to improve the efficiency, effectiveness and accuracy of high-density layout load cabin product connection correctness verification by visual inspection method.

Description

TECHNICAL FIELD

[0001] The application relates to a visual inspection method for the connection correctness of a high-density layout load cabin product and belongs to the technical field of machine vision. BACKGROUND

[0002] A communication satellite represented by a high-throughput satellite is usually designed with a large platform, multiple beams and high throughput, and the load cabin has the characteristics of high-density layout and multi-level connection. There are thousands of cabin products on the load cabin, mainly single machines, waveguides and cables, and the connection relationship is very complex. The absence, misplacement of single machines and the misconnection and missing connection of waveguides and cables will have a huge impact on the cabin performance and project progress. Therefore, the connection correctness of the load cabin product needs to be inspected, mainly including the correctness of the installation position of the single machine and the correctness of the connection relationship of the waveguide and the cable.

[0003] The load cabin inspection currently adopts a manual visual inspection method, and photographs of the cabin products are taken and stored. The products in the load cabin are mutually shielded, the operation space is small, the assembly inspection and image acquisition have a large workload, high working strength and low implementation efficiency. At the same time, it is easy to miss and misjudge, and some problems are found in the power-on test and the later stage of the project, so that the efficiency and effectiveness of the inspection cannot be guaranteed.

[0004] The machine vision technology can effectively avoid damage to the cabin products on the load cabin caused by collision and the like by receiving and processing the images of the objects by the machine in a non-contact manner to identify, measure and judge the objects. SUMMARY

[0005] The technical problem of the application is to overcome the shortcomings of the prior art and provide a visual inspection method for the connection correctness of a high-density layout load cabin product to realize rapid and effective inspection of the cabin products on the load cabin.

[0006] The technical solution of the application is:

[0007] The application discloses a visual inspection method for the connection correctness of a high-density layout load cabin product, which comprises the following steps:

[0008] A three-dimensional design model of the load cabin is constructed;

[0009] Load cabin product information is obtained according to the three-dimensional design model of the load cabin;

[0010] The cabin products on the load cabin are identified according to the load cabin product information to obtain cabin product identification information;

[0011] Images of the cabin products on the load cabin are collected in sequence to obtain a two-dimensional image containing the cabin product identification information;

[0012] Collect three-dimensional data of the products on the loading cabin and splice them to generate a three-dimensional inspection model under a reference coordinate system;

[0013] Store the collected images in an image library and manage them according to product numbers;

[0014] According to the three-dimensional design model, the identification information, the two-dimensional image and the three-dimensional inspection model, adopt an image intelligent recognition method to perform product connection correctness inspection;

[0015] According to the inspection result, perform product adjustment, re-shooting or re-inspection.

[0016] Further, in the above inspection method, the loading cabin product is identified to obtain loading cabin product identification information, specifically:

[0017] The first shape identification pasted on the single machine represents the single machine code; the interface identification represents the connection relationship at the interface; and the R identification represents the reference hole position and the single machine installation direction;

[0018] The second shape identification pasted on the waveguide represents the waveguide code; the interface identification pasted on the connection between the two ends of the waveguide and the single machine represents the connection relationship; and the inscription on the flange represents the waveguide code and waveguide subassembly information;

[0019] The third shape identification pasted on the cable represents the cable code; and the interface identification pasted on the connection between the two ends of the cable and the single machine represents the connection relationship.

[0020] Further, in the above inspection method, the loading cabin product information includes single machine position coordinates, single machine size and waveguide and cable connection relationship table.

[0021] Further, in the above inspection method, the loading cabin product is sequentially imaged, specifically: the loading cabin single machine, waveguide and cable are sequentially imaged.

[0022] Further, in the above inspection method, according to the three-dimensional design model, the identification information, the two-dimensional image and the three-dimensional inspection model, an image intelligent recognition method is adopted to perform product connection correctness inspection, specifically:

[0023] According to the two-dimensional image and the identification information, the identification of the single machine, the waveguide and the cable on the cabin is recognized through a text recognition technology, a mapping relationship between the waveguide, the cable and the single machine is constructed, and is compared with the connection relationship table to inspect the correctness of the waveguide and cable connection relationship;

[0024] The three-dimensional design model and the three-dimensional inspection model are placed in the same coordinate system for comparison to perform single machine installation position correctness inspection and waveguide connection relationship correctness inspection.

[0025] Further, in the above-mentioned inspection method, the labels of the cable at the two end interfaces are consistent with the labels at the single machine interfaces, and the inspection of the cable connection relationship is correct.

[0026] Further, in the above-mentioned inspection method, the single machine installation position correctness inspection is specifically as follows:

[0027] It is judged whether the following three conditions are met:

[0028] Condition 1: the single machine code is determined through the first graphic mark, and whether the single machine installation direction is correct is determined through the R mark;

[0029] Condition 2: whether the reference hole position coordinates are consistent with the position list is further judged;

[0030] Condition 3: whether the deviation result of the three-dimensional inspection model and the three-dimensional design model is within the required range;

[0031] If all the conditions are met, it is judged that the single machine installation position is correct, and if not, it is judged that the single machine installation position is incorrect.

[0032] Further, in the above-mentioned inspection method, the waveguide connection relationship correctness inspection is specifically as follows: the waveguide information is identified through the waveguide code label on the waveguide or the characters engraved on the flange; whether the waveguide label codes at both ends match the single machine interface codes is identified.

[0033] Further, in the above-mentioned inspection method, according to the layering design result of the three-dimensional design model, after the waveguide at each layer of the cabin plate is completed, an image acquisition device is used to acquire images of each layer of waveguide.

[0034] Further, in the above-mentioned inspection method, for the product that needs to be adjusted or retaken, the product position information is given through the image library, and is highlighted in the three-dimensional design model to guide the retaking; the state of the product on the cabin is recorded and updated through the image library and the connection relationship table, and finally all the products on the cabin reach the state of completed installation and correct connection, and the product images and connection states on the cabin are retrieved through the load cabin product image library.

[0035] The beneficial effects of the present application over the prior art are as follows:

[0036] (1) The present application can greatly improve the efficiency and effectiveness of the load cabin product connection relationship correctness inspection by automatically acquiring images of the load cabin product, identifying the text information of the product on the cabin through machine vision, and comparing with the connection relationship table and the three-dimensional design model, etc., which can overcome the low efficiency and limited reliability of manual image acquisition and visual inspection, and the method proposed in the present application can be applied to the inspection of the connection correctness of the products on the cabin of multiple types of communication satellites.

[0037] (2) The present application integrates the early design materials, the image of the product on the load cabin, the connection correctness inspection state and the result into the image library of the load cabin by fusing the early three-dimensional design model, the connection relation table and the identification system, realizes the effective management of the design materials and the image of the product on the cabin of the load cabin single machine, the waveguide and the cable, and the connection correctness inspection result is displayed intuitively, and the material is inquired and called quickly and conveniently. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 The workflow of the visual inspection method for the connection correctness of the load cabin product in the present application;

[0039] Figure 2 The overall structure of the visual inspection method for the connection correctness of the load cabin product in the present application;

[0040] Figure 3 The connection relation diagram and part of the identification of the load cabin product in the present application;

[0041] Figure 4 The state of the load cabin product in different visual inspection links in the present application;

[0042] Figure 5 The shooting and guiding of the re-shooting process of the load cabin product in the present application;

[0043] Figure 6 The schematic diagram of the installation position correctness inspection of the load cabin single machine in the present application;

[0044] Figure 7 The schematic diagram of the connection relation correctness inspection of the waveguide and the cable of the load cabin in the present application. DETAILED DESCRIPTION

[0045] The present application patent will be further described in detail in combination with the drawings and the specific implementation manners.

[0046] As shown in the drawings, Figure 1 a visual inspection method for the connection correctness of a high-density layout load cabin product, characterized in that it comprises:

[0047] Step 1, constructing a three-dimensional design model of the load cabin;

[0048] Step 2, obtaining product information on the load cabin according to the three-dimensional design model of the load cabin;

[0049] Step 3, identifying the product on the load cabin according to the product information on the load cabin, and obtaining identification information of the product on the load cabin;

[0050] Step 4, sequentially collecting images of the product on the load cabin, and obtaining a two-dimensional image containing the identification information of the product on the load cabin;

[0051] Step 5, collect three-dimensional data of the products on the cabin and splice to generate a three-dimensional inspection model under the reference coordinate system;

[0052] Step 6, store the collected images in an image library and manage them according to product numbers;

[0053] Step 7, according to the three-dimensional design model, the identification information, the two-dimensional image, and the three-dimensional inspection model, use image intelligent recognition method to perform product connection correctness inspection;

[0054] Step 8, according to the inspection results, perform product adjustment, re-shooting, or re-inspection.

[0055] Preferably, in step 3, the products on the cabin are labeled to obtain identification information of the products on the cabin, specifically:

[0056] The first shape label pasted on the single machine represents the single machine code; the interface label represents the connection relationship at the interface; and the R label represents the reference hole position and the single machine installation direction;

[0057] The second shape label pasted on the waveguide represents the waveguide code; the interface label pasted on the connection between the two ends of the waveguide and the single machine represents the connection relationship; and the inscription on the flange represents the waveguide code and waveguide subassembly information;

[0058] The third shape label pasted on the cable represents the cable code; and the interface label pasted on the connection between the two ends of the cable and the single machine represents the connection relationship.

[0059] Preferably, in step 3, the information of the products on the cabin includes single machine position coordinates, single machine dimensions, and a connection relationship table of the waveguide and the cable.

[0060] Preferably, in step 4, the products on the cabin are sequentially imaged, specifically: the products on the cabin are sequentially imaged according to the order of the single machine, the waveguide, and the cable.

[0061] Preferably, in step 7, according to the three-dimensional design model, the identification information, the two-dimensional image, and the three-dimensional inspection model, use image intelligent recognition method to perform product connection correctness inspection, specifically:

[0062] Step 71, according to the two-dimensional image and the identification information, identify the labels of the single machine, the waveguide, and the cable on the cabin through text recognition technology, construct the mapping relationship between the waveguide, the cable, and the single machine, and compare with the connection relationship table to inspect the correctness of the connection relationship of the waveguide and the cable;

[0063] Step 72, place the three-dimensional design model and the three-dimensional inspection model in the same coordinate system for comparison to perform single machine installation position correctness inspection and waveguide connection relationship correctness inspection.

[0064] Preferably, the labels of the cable at the two end interfaces are consistent with the labels at the single machine interfaces, and the cable connection relationship is verified to be correct.

[0065] Preferably, in step 72, the single machine installation position correctness verification is specifically as follows:

[0066] whether the following three conditions are met:

[0067] Condition 1: determining the single machine code through the first graphic mark, and determining whether the single machine installation direction is correct through the R mark;

[0068] Condition 2: determining whether the reference hole position coordinates are consistent with the position list;

[0069] Condition 3: whether the deviation result of the three-dimensional verification model and the three-dimensional design model is within the required range;

[0070] If all the conditions are met, it is determined that the single machine installation position is correct; if not, it is determined that the single machine installation position is incorrect.

[0071] Preferably, in step 72, the waveguide connection relationship correctness verification is specifically as follows: identifying the waveguide information through the waveguide code label or the inscription on the flange; and identifying whether the waveguide end label codes match the single machine interface codes.

[0072] Preferably, according to the layering design result of the three-dimensional design model, after the waveguide on each layer of the cabin plate is installed, an image acquisition device is used to acquire the image of each layer of waveguide.

[0073] Preferably, for the product that needs to be adjusted or retaken, the product position information is given through the image library, and is highlighted in the three-dimensional design model to guide the retaking; the state of the product on the cabin is recorded and updated through the image library and the connection relationship table, and the product on the cabin finally all reaches the state of completed installation and correct connection, and the product image and the connection state on the cabin are searched through the load cabin product image library.

[0074] Embodiment

[0075] The embodiment discloses a visual verification method for the connection correctness of a high-density layout load cabin product. Two-dimensional images and three-dimensional images of single machines, waveguides and cables on a load cabin are acquired through an image acquisition device. Position information and connection information contained in the acquired images are identified through machine vision technology, and are compared with a three-dimensional design model and a connection relationship table, so that single machine installation position correctness and waveguide and cable connection relationship correctness identification results of the high-density layout load cabin are obtained, and rapid and effective verification of the product on the load cabin is realized.

[0076] The embodiment is a visual inspection method for correctness of product connection of a high-density layout load cabin based on machine vision. The visual inspection method combines the structural characteristics of the load cabin and the characteristics of the AIT stage, and the technical solution mainly consists of three parts of early design, image acquisition and management, and visual intelligent identification.

[0077] In the embodiment, the early design part of the load cabin is first described in detail. The early design of the embodiment mainly includes three-dimensional design model design, connection relationship table design, and identification system. In the design stage, the three-dimensional model design of the load cabin is completed, and the waveguide part is designed in layers. After the three-dimensional model design is completed, the single machine position coordinates and the single machine size are quickly exported in batches, as well as the connection relationship table of the waveguide and the cable. The single machine, the waveguide, and the cable are pasted with identification, reflecting the product information on the cabin. The corresponding relationship of the product identification on the cabin constitutes the product connection network of the load cabin, which is an important basis for checking the correctness of the product connection of the load cabin. The identification of the product on the cabin in the embodiment is as follows:

[0078] (1) The diamond-shaped identification pasted on the single machine reflects the single machine code; the interface identification reflects the connection relationship at the interface; and the R identification reflects the reference hole position and the single machine installation direction.

[0079] (2) The square identification pasted on the waveguide reflects the waveguide code; the interface identification pasted at the two ends connected with the single machine reflects the connection relationship; in addition, the characters engraved on the flange reflect the waveguide code and the waveguide subassembly information.

[0080] (3) The identification pasted on the cable reflects the cable code; the interface identification pasted at the two ends connected with the single machine reflects the connection relationship.

[0081] The image acquisition and management part of the load cabin in the embodiment is described in detail. The assembly sequence of the products on the cabin is usually: single machine -> waveguide -> cable, and the initial image acquisition of the products on the cabin is also performed in this order. The acquired content is the shape image of the products on the cabin and the identification information such as the diamond-shaped identification and the interface identification, mainly in the form of two-dimensional images. The single machine and the waveguide are rigid materials, and in addition, the point cloud and other three-dimensional data of the single machine and the waveguide need to be acquired and spliced to generate a three-dimensional inspection model under the reference coordinate system. The cable is flexible, and the actual path has certain differences with the design model, so it is not necessary to separately acquire the three-dimensional data of the cable.

[0082] The image acquisition of the payload cabin product is mainly automatic shooting, supplemented by automatic or manual shooting of a small amount of products, and the acquired images are stored in an image library for management according to the product number. The products on the payload cabin are densely arranged, and there may be occlusion between products such as waveguides. In addition, due to the large amount of image acquisition data, there may be missed shooting and unclear shooting. For the products on the cabin for which no image is acquired or the acquired image does not meet the identification conditions, two ways are used to improve the image acquisition of the payload cabin and improve the inspection efficiency. In this embodiment, for the image acquisition of the waveguide assembly, the three-dimensional model is divided into layers according to the design, and after the waveguide on each layer of the cabin plate is installed, the image acquisition device is used to acquire the image, so as to avoid the occlusion of the lower waveguide caused by the installation of the upper waveguide, and to obtain complete acquisition information. Secondly, after the image acquisition of the products on the payload cabin is completed, the position information is displayed in the image library in combination with the visual intelligent recognition result, and the three-dimensional model is highlighted to guide the re-shooting and re-inspection, so as to further improve the correctness of the visual inspection and the inspection efficiency. In this embodiment, for the products for which the connection error is identified by visual intelligent recognition, the model is highlighted to provide guidance, the shooting is performed again after adjustment, and the image library is updated, and the version is updated to reflect the latest actual state of the payload cabin.

[0083] Further, the visual intelligent recognition part of the payload cabin in this embodiment is described in detail. First, the identification of the single machine, waveguide and cable on the cabin is recognized by the text recognition technology, the mapping relationship between the waveguide, cable and single machine is constructed, the real connection on the payload cabin is restored, and the connection relationship table is compared to verify the correctness of the waveguide and cable connection relationship. The middle section of the cable needs to be tied, so in the identification process, the label of the cable at the two end interfaces is consistent with the label at the single machine interface, so the cable connection relationship is considered correct. In addition, by placing the three-dimensional design model and the three-dimensional inspection model in the same coordinate system for comparison, it is verified whether the single machine is missing, whether the installation position is offset or rotated, and whether the waveguide installation has a large deviation. According to the visual inspection stage of the payload cabin in this embodiment, the payload cabin products are in different image acquisition and connection correctness verification states, and after all the payload cabin products are verified to be in the correct connection state, the visual inspection process of the connection correctness of the payload cabin products is completed.

[0084] In this embodiment, the image acquisition and recognition of the payload cabin products are performed simultaneously to improve the speed of the visual inspection of the connection correctness of the payload cabin products. For a small amount of products for which no effective image is acquired, the image acquisition point is shot again or manually supplemented to improve the re-inspection. When the single machine installation position or the waveguide and cable connection relationship is identified to be incorrect, adjustment needs to be made on the cabin. After the adjustment is completed, the image information of the corresponding acquisition point at the adjustment position is acquired again, and is updated to the image library for re-identification, so as to obtain the complete visual inspection result of the connection correctness of the payload cabin products.

[0085] As Figure 1 shown, the embodiment provides a visual inspection method for connection correctness of high-density layout payload cabin product, which comprises the following steps: first, the pre-design of the payload cabin and the cabin product is carried out, after completion, the image acquisition process and the image intelligent recognition and judgment process are carried out at the same time, the image library and the connection correctness of the payload cabin product are constantly updated, and the visual inspection efficiency of the payload cabin is improved. At the same time, the product images that are missed and do not meet the identification requirements are guided to be retaken and rechecked, and the collected images and the identification results are updated, and the process ends after the connection correctness of the payload cabin is checked without error.

[0086] As Figure 2 shown, the overall structure of the visual inspection method for connection correctness of the payload cabin product proposed in the embodiment comprises pre-design, image acquisition and management, and visual intelligent recognition. The judgment standard and object of the payload cabin product are established through pre-design. The information of the payload cabin product identification, installation position and structure is obtained through image acquisition and management, and the image of the cabin product is effectively managed through the image library. The image contained in the image of the cabin product is intelligently recognized through visual intelligent recognition, and the installation position correctness of the waveguide and the connection relationship correctness of the waveguide and the cable are checked through text recognition, model matching and coordinate calculation technologies.

[0087] As Figure 3 shown is a system diagram of the connection relationship and identification of the single machine, waveguide and cable on the payload cabin. Through the diamond-shaped identification and the interface identification, the connection relationship network of the cabin product is established, which together with the three-dimensional design model serves as an important basis for judging the connection correctness of the cabin product.

[0088] As Figure 4 shown is the visual inspection state of the payload cabin product. The payload cabin product can be divided into (payload cabin product has been installed; not installed), (image of installed product has been collected and entered; not collected and entered), (entered image has been recognized and judged for connection correctness; not judged for connection correctness), (cabin product connection is correct; cabin product connection is incorrect), (connected product has been adjusted and re-collected image and recognized; not adjusted), (all payload cabin products have completed image collection and connection correctness inspection; not completed). Through the state of the payload cabin product, the visual inspection stage and effect can be intuitively reflected. After all the payload cabin products complete image collection and connection correctness judgment, the visual inspection process of the payload cabin product is completed.

[0089] As Figure 5As shown is a load cabin product shooting and retake process diagram, image acquisition points are designed in advance through three-dimensional design model layering design, and product images on the cabin are collected through collection equipment.For products that need to be retaken, product position information is given through an image library, and is highlighted in the model to guide fast retaking, and improve the integrity of the load cabin product images and the efficiency of visual inspection.

[0090] As shown in the figure, Figure 6 As shown is a single machine installation position correctness judgment method in the embodiment, first, the single machine code is determined through the diamond mark, then whether the single machine installation direction is correct is determined through the R mark, then whether the reference hole position coordinates are consistent with the position list is judged, and whether the deviation result of the inspection model and the design model is within the required range is verified.If the above requirements are met, it can be judged that the single machine installation position is correct.

[0091] As shown in the figure, Figure 7 As shown is a waveguide and cable connection relationship correctness judgment method in the embodiment, the identification information of the waveguide is recognized, the single machine information connected at both ends of the waveguide is obtained, and the connection relationship table is compared without error, so that the cable connection relationship is correct.In addition, the waveguide layering design model and the inspection model are compared, and the deviation is small, which indicates that the waveguide installation condition is good.In this embodiment, whether the identification information at the interface of both ends of the cable matches the single machine or is matched and queried through the connection relationship table, the correctness of the cable connection relationship is verified.When the single machine installation position or the waveguide and cable connection relationship is incorrect, adjustment needs to be made on the cabin.After the adjustment is completed, the image information of the adjustment place is collected again and is updated to the image library, and finally the load cabin whole-cabin product connection correctness visual inspection is completed.

[0092] Although the content of the present application has been described in detail through the above preferred embodiments, it should be recognized that the above description should not be considered as a limitation of the present application.After reading the above content by those skilled in the art, various modifications and alternatives of the present application will be obvious.The scope of protection of the present application should be defined by the appended claims.

[0093] The content not described in detail in the specification of the present application belongs to the known technology of those skilled in the art.

Claims

1. A visual inspection method for the correct connection of a high-density layout load cell product, characterized in that, include: Construct a three-dimensional design model of the payload compartment; Based on the three-dimensional design model of the payload compartment, the product information on the payload compartment is obtained; Based on the product information on the load compartment, the products on the load compartment are identified to obtain product identification information on the load compartment; The products on the load chamber are sequentially imaged to obtain two-dimensional images containing product identification information on the load chamber. Collect and stitch together the 3D data of the products on the cabin to generate a 3D inspection model in the reference coordinate system. The collected images are stored in an image library and managed according to product number; Based on the aforementioned 3D design model, identification information, 2D image, and 3D inspection model, an image intelligent recognition method is used to verify the correctness of the product connection. Adjustments, re-photographing, or re-inspection of the product are made based on the inspection results. The step of verifying the correctness of product connections using an image intelligent recognition method based on the three-dimensional design model, identification information, two-dimensional image, and three-dimensional inspection model is as follows: Based on the two-dimensional images and identification information, the identification of the single units, waveguides and cables on the cabin is identified by text recognition technology, the mapping relationship between waveguides, cables and single units is constructed, and the relationship is compared with the connection relationship table to verify the correctness of the connection relationship between waveguides and cables; The 3D design model and the 3D verification model are compared in the same coordinate system to verify the correctness of the single-unit installation position and the correctness of the waveguide connection relationship.

2. The visual inspection method for the correct connection of a high-density layout load cell product according to claim 1, characterized in that: The process of identifying the products on the payload compartment to obtain product identification information specifically involves: The first shape mark affixed to the unit represents the unit's code; the mark at the interface represents the connection relationship at the interface; the affixed R mark represents the location of the reference hole and the unit's installation direction. A second-shape label is affixed to the waveguide to represent the waveguide designation; interface labels are affixed at both ends where it connects to the unit to represent the connection relationship; and lettering on the flange represents the waveguide designation and waveguide sub-assembly information. The third shape mark affixed to the cable represents the cable code; the interface mark affixed to both ends of the cable where it connects to the unit represents the connection relationship.

3. The visual inspection method for the correct connection of a high-density layout load cell product according to claim 1, characterized in that: The product information on the payload compartment includes the unit's position coordinates, unit dimensions, and a table showing the connection relationships of waveguides and cables.

4. The visual inspection method for the correct connection of a high-density layout load cell product according to claim 1, characterized in that: Images are acquired sequentially of the products on the payload compartment, specifically in the order of individual units, waveguides, and cables on the payload compartment.

5. The visual inspection method for the correct connection of a high-density layout load cell product according to claim 1, characterized in that: The labels at both ends of the cable are consistent with the labels at the stand-alone interface, confirming that the cable connection is correct.

6. The visual inspection method for the correct connection of a high-density layout load cell product according to claim 1, characterized in that: The specific method for verifying the correctness of the single-unit installation location is as follows: Determine whether the following three conditions are met: Condition 1: Determine the unit code through the first graphic identifier, and then determine whether the unit's installation direction is correct through the R identifier; Condition 2: Then check whether the coordinates of the reference hole position are consistent with the position list; Condition 3: Whether the deviation between the 3D verification model and the 3D design model is within the required range; If all conditions are met, the installation location of the unit is considered correct; otherwise, the installation location of the unit is considered incorrect.

7. The visual inspection method for the correct connection of a high-density layout load cell product according to claim 1, characterized in that: The correctness verification of waveguide connection relationship is specifically as follows: identify waveguide information by waveguide code labels on the waveguide or engravings on the flange; identify whether the label codes at both ends of the waveguide match the single-machine interface code.

8. The visual inspection method for the correct connection of a high-density layout load cell product according to claim 1, characterized in that: Based on the layered design results of the 3D design model, after each waveguide layer of the cabin is installed, an acquisition device is used to acquire images of each waveguide layer.

9. The visual inspection method for the correct connection of a high-density layout load cell product according to claim 1, characterized in that: For products that require adjustment or reshooting, the product location information is provided through the image library and highlighted in the 3D design model to guide the reshooting; the status of products on the payload bay is recorded and updated through the image library and connection relationship table, and all products on the payload bay eventually reach the state of complete installation and correct connection, and the images and connection status of products on the payload bay are retrieved through the product image library.

Images