A processing method and an electronic device

Abstract

The present application provides a processing method and an electronic device. The method includes: obtaining a first image of a target device, and identifying the main ingredient type in the first image; determining the attribute information of each main ingredient under the main ingredient type, and generating description information for the main ingredient, including the main ingredient type and the attribute information; determining vector information corresponding to the description information of the main ingredient; determining first matching information based on the vector information of the main ingredient and a first training model; the first matching information includes a matching manner between the description information of the main ingredient and the description information of at least one auxiliary ingredient; generating a bill of materials for the target device based on the first matching information. Thus, the present application simplifies the user operation and improves the accuracy of generating the bill of materials by automatically generating the bill of materials for the target device after obtaining the first image of the target device.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and more specifically, to a processing method and an electronic device. Background Art

[0002] A bill of materials refers to the detailed list of specific materials for a product. Currently, users can configure it based on a material management system.

[0003] For a device that has been assembled, if a user wants to determine the bill of materials for a certain device belonging to the device, they need to manually compare the materials on the device one by one in the material management system. The operation is very cumbersome and error-prone. Summary of the Invention

[0004] In view of this, this application provides a processing method and an electronic device to solve the above technical problems.

[0005] To achieve the above object, this application provides the following technical solutions:

[0006] A processing method includes:

[0007] Obtain a first image of a target device, and identify the main material type in the first image;

[0008] Determine the attribute information of each main material under the main material type, and generate description information for the main material, including the main material type and the attribute information;

[0009] Determine the vector information corresponding to the description information of the main material;

[0010] Based on the vector information of the main material and a first training model, determine first matching information; the first matching information includes the matching method of the description information of the main material and the description information of at least one auxiliary material;

[0011] Generate a bill of materials for the target device based on the first matching information.

[0012] Optionally, it further includes:

[0013] Based on the vector information of the main material and a second training model, determine second matching information; the second matching information includes the matching method of the description information of the main material and the description information of at least one auxiliary material; where the second training model is different from the first training model;

[0014] Correspondingly, the generating a bill of materials for the target device based on the first matching information includes:

[0015] Determine the intersection of the first matching information and the second matching information, and generate target matching information;

[0016] Generate a bill of materials for the target device based on the target matching information.

[0017] Optionally, it further includes:

[0018] Eliminate incorrect matching methods from the second matching information based on pre-determined matching rules to generate the second matching information after elimination;

[0019] Correspondingly, the determining the intersection of the first matching information and the second matching information to generate a target matching method includes:

[0020] Determine the intersection of the first matching information and the second matching information after elimination to generate target matching information.

[0021] Optionally, the obtaining a first image of a target device and identifying the main material type in the first image includes:

[0022] Control the acquisition unit to take multi-angle photos of the target device equipped with various components to obtain multiple first images of the target device;

[0023] Use a third training model to identify the multiple first images and identify the main material type in the first images.

[0024] Optionally, the determining the first matching information based on the vector information of the main material and the first training model includes:

[0025] Input the vector information of the main material into the first training model, and output first matching vector information through the first training model;

[0026] Generate the first matching information based on the first matching vector information.

[0027] Optionally, it further includes:

[0028] Add the bill of materials to the product bill of materials corresponding to the target device in the rule database.

[0029] Optionally, it further includes:

[0030] Determine a target device; wherein, the target device is assembled from various types of main materials and auxiliary materials;

[0031] Control the acquisition unit to acquire a second image of the interior of the target device;

[0032] Process the second image to obtain the material matching information inside the target device;

[0033] Search for the target product bill of materials corresponding to the target device in the rule database;

[0034] Based on the target product bill of materials, determine whether the material matching information is correct.

[0035] Optionally, it further includes:

[0036] Process the second image to obtain the position information of the main material and / or auxiliary materials;

[0037] Obtain the pre-established three-dimensional model corresponding to the target device;

[0038] Based on the three-dimensional model, detect whether the position information is inserted in place.

[0039] Optionally, it further includes:

[0040] Determine the order bill of materials with modifications and the changed material information in the order bill of materials;

[0041] Search for the product bill of materials corresponding to the order bill of materials in the rule database;

[0042] Based on the product bill of materials, determine whether the matching information of the changed material information in the order bill of materials is correct;

[0043] If it is incorrect, modify the order bill of materials based on the product bill of materials.

[0044] An electronic device, comprising:

[0045] A memory for storing programs;

[0046] A processor that runs the program and is used to obtain a first image of a target device, identify the type of main material in the first image; determine the attribute information of each main material under the type of main material, generate description information for the main material that includes the type of main material and the attribute information; determine vector information corresponding to the description information of the main material; determine first matching information based on the vector information of the main material and a first training model; the first matching information includes the matching method of the description information of the main material and the description information of at least one auxiliary material; generate a bill of materials for the target device based on the first matching information.

[0047] As can be seen from the above technical solutions, the present application provides a data processing method. By acquiring a first image of a target device, the main material type in the first image is identified, and then the attribute information of each main material under the main material type is determined. A description information including the main material type and the attribute information is generated for the main material, and vector information corresponding to the description information of the main material is determined. Based on the vector information of the main material and a first training model, first matching information is determined, where the first matching information includes a matching method between the description information of the main material and the description information of at least one auxiliary material. A bill of materials for the target device is generated based on the first matching information. Thus, the present application simplifies the user operation and improves the accuracy of generating the bill of materials by automatically generating the bill of materials for the target device after acquiring the first image of the target device. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only the embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on the provided drawings without creative efforts.

[0049] Figure 1 It is a schematic flowchart of a processing method provided in the first embodiment of the method of the present application;

[0050] Figure 2 It is a schematic diagram of acquiring the first image provided in the first embodiment of the method of the present application;

[0051] Figure 3 It is a schematic flowchart of a processing method provided in the second embodiment of the method of the present application;

[0052] Figure 4 It is a schematic flowchart of a processing method provided in the third embodiment of the method of the present application;

[0053] Figure 5 It is a schematic flowchart of a processing method provided in the fourth embodiment of the method of the present application;

[0054] Figure 6 It is a schematic flowchart of a processing method provided in the fifth embodiment of the method of the present application;

[0055] Figure 7 It is a schematic structural diagram of an electronic device provided in the first embodiment of the device of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0057] Embodiment 1 of the method of the present application provides a processing method. As Figure 1 shown, the method includes the following steps:

[0058] Step 101: Obtain a first image of a target device, and identify the main material type in the first image;

[0059] In the present application, the target device is a device equipped with various components, such as the main board of an electronic device. The first image is an image of the target device. This image can be an image taken of the target device, or an image taken of a paper drawing of the target device, or an electronic drawing of the target device. By identifying the first image, the main material type installed on the target device can be determined. For example, the first image can be obtained through various methods as Figure 2 shown.

[0060] The main material type is the main material type that can be determined by identifying the first image. For example, if a processor is installed on the target device and Intel is marked on the processor, then by identifying the first image, the main material type can be determined to include: Intel CPU. Of course, it can also include the main material types of main materials such as memory and hard disk.

[0061] Optionally, obtaining a first image of a target device and identifying the main material type in the first image may include the following steps:

[0062] 1) Control the acquisition unit to take multi-angle photos of the target device equipped with various components, and obtain multiple first images of the target device;

[0063] 2) Use the third training model to identify the multiple first images, and identify the main material type in the first images.

[0064] Among them, by taking multi-angle photos of the target device, first images of the target device at different angles can be obtained. The number of the first images is not limited in the present application. For example, 1000 first images of the target device can be taken, and then the third training model is used to identify the multiple first images to identify the main material type in the first images.

[0065] Among them, the third training model is a deep learning training model, specifically it can be a Fast R-CNN model, which can be used to identify the main material type in the input image by pre-training a large number of images collected by the acquisition units.

[0066] It should be noted that in addition to being able to identify the first image obtained by the acquisition unit taking multi-angle photos of the target device equipped with various components, the third training model can also identify the first image obtained by the acquisition unit taking a photo of the paper drawing of the target device, or identify the first image of the electronic drawing of the target device. All of these are achievable, and the training images used by the third training model during the training process can be adjusted accordingly.

[0067] Optionally, before inputting the first image into the third training model, the effective range in the first image can be marked first, so as to input the first image into the third training model, and the third training model can identify the effective range in the first image.

[0068] It should be noted that if there are multiple identical main material types in the first image, the number of identical main material types can also be determined through this step.

[0069] Step 102: Determine the attribute information of each main material under the main material type, and generate description information for the main material, including the main material type and the attribute information.

[0070] Specifically, the attribute information of each main material under the main material type can be determined from the target requirement document corresponding to the target device. The attribute information is the attribute description of the main material corresponding to this main material type, such as capacity, model number, etc. For example, the attribute information of an Intel CPU is: I5 (I5 is the model number of the Intel CPU).

[0071] Among them, the target requirement document is a user requirement document, which at least includes various main material types for assembling the target device and the attribute information corresponding to the main material types.

[0072] Step 103: Determine the vector information corresponding to the description information of the main material.

[0073] Specifically, k attribute encodings can be formed for k attributes in the description information of the main material on a k-dimensional vector, and then a 1-dimensional vector representing the type is added to form a k + 1-dimensional vector.

[0074] Or, the vector information after encoding various materials is pre-stored in the component library. Specifically, the attributes of the materials can be divided based on the type. For any material with k attributes, k attribute encodings can be formed on a k-dimensional vector, and then a 1-dimensional vector representing the type of the material is added to form a k + 1-dimensional vector.

[0075] Among them, k is a positive integer, and this application does not limit the specific coding method. For example, Huffman coding can be used.

[0076] Step 104: Determine the first matching information based on the vector information of the main ingredient and the first training model;

[0077] Among them, the first matching information includes the matching method of the description information of the main ingredient and the description information of at least one auxiliary ingredient.

[0078] Optionally, determining the first matching information based on the vector information of the main ingredient and the first training model may include the following steps:

[0079] 1) Input the vector information of the main ingredient into the first training model, and output the first matching vector information through the first training model;

[0080] 2) Generate the first matching information based on the first matching vector information.

[0081] Both the input and output of the first training model are in vector form, and it can be used to determine the auxiliary ingredients that match the main ingredient by pre-training a large number of matching information in vector form.

[0082] Specifically, the first matching vector information may include the vector information of the mapping relationship from the main ingredient to the auxiliary ingredient. For the vector information of a main ingredient, at least one auxiliary ingredient matching can be determined through the first training model. That is, the first matching information includes at least one auxiliary ingredient matching that matches the main ingredient type. In the matching method of the description information of the main ingredient and the description information of at least one auxiliary ingredient, the description information of the auxiliary ingredient includes the type of the auxiliary ingredient and the corresponding attribute information.

[0083] Optionally, the first training model is a neural network training model, specifically, it can be a BP neural network training model or a decision tree training model, etc.

[0084] Step 105: Generate a bill of materials for the target device based on the first matching information.

[0085] It can be seen that the present application provides a data processing method. By obtaining a first image of a target device, the main ingredient type in the first image is identified, and then the attribute information of each main ingredient under the main ingredient type is determined. A description information including the main ingredient type and the attribute information is generated for the main ingredient, and vector information corresponding to the description information of the main ingredient is determined. Based on the vector information of the main ingredient and a first training model, first matching information is determined, where the first matching information includes the matching manner between the description information of the main ingredient and the description information of at least one auxiliary ingredient. Based on the first matching information, a material list for the target device is generated. Thus, in the present application, after obtaining the first image of the target device, a material list for the target device is automatically generated, which simplifies the user operation and improves the accuracy of generating the material list.

[0086] Embodiment 2 of the method of the present application provides a processing method. As Figure 3 shown, the method includes the following steps:

[0087] Step 301: Obtain a first image of a target device and identify the main ingredient type in the first image;

[0088] Step 302: Determine the attribute information of each main ingredient under the main ingredient type and generate description information including the main ingredient type and the attribute information for the main ingredient;

[0089] Step 303: Determine vector information corresponding to the description information of the main ingredient;

[0090] Step 304: Determine first matching information based on the vector information of the main ingredient and a first training model;

[0091] The first matching information includes the matching manner between the description information of the main ingredient and the description information of at least one auxiliary ingredient.

[0092] Step 305: Determine second matching information based on the vector information of the main ingredient and a second training model;

[0093] The second matching information includes the matching manner between the description information of the main ingredient and the description information of at least one auxiliary ingredient.

[0094] It should be noted that in the present application, the second training model is different from the first training model, but the functions of both are to determine the matching information. In order to improve the accuracy of the matching information, the present application uses two different training models. For example, if the first training model is a BP neural network training model, then the second training model can be a decision tree training model; or if the first training model is a decision tree training model, then the second training model can be a BP neural network training model. Or, the first training model and the second training model are other different neural network training models, which can all be realized.

[0095] Among them, determining the second collocation information based on the vector information of the main ingredient and the second training model may include the following steps:

[0096] 1) Input the vector information of the main ingredient into the second training model, and output the second collocation vector information through the second training model;

[0097] 2) Generate the second collocation information based on the second collocation vector information.

[0098] Among them, both the input and output of the second training model are in vector form, and it can be used to determine the auxiliary ingredients paired with the main ingredient by pre-training a large number of collocation information in vector form.

[0099] Specifically, the second collocation vector information may include the vector information of the mapping relationship from the main ingredient to the auxiliary ingredient. For the vector information of one main ingredient, at least one auxiliary ingredient collocation can be determined through the second training model. That is, the second collocation information includes at least one auxiliary ingredient collocation paired with the main ingredient type. In the collocation method of the description information of the main ingredient and the description information of at least one auxiliary ingredient in the second collocation information, the description information of the auxiliary ingredient includes the type of the auxiliary ingredient and the corresponding attribute information.

[0100] Step 306: Determine the intersection of the first collocation information and the second collocation information to generate the target collocation information;

[0101] In this embodiment, take the intersection of the first collocation information determined based on the first training model and the second collocation information determined based on the second training model as the final collocation information.

[0102] To ensure the accuracy of the generated second collocation information, before generating the target collocation information, it is also possible to first check whether there is incorrect collocation information in the second collocation information. That is, in another method embodiment, it may also include: eliminating the incorrect collocation methods in the second collocation information based on the pre-determined collocation rules to generate the second collocation information after elimination. Correspondingly, the determining the intersection of the first collocation information and the second collocation information to generate the target collocation method includes: determining the intersection of the first collocation information and the second collocation information after elimination to generate the target collocation information.

[0103] Among them, the collocation rules are configuration rules set by the user and can be used to check whether there are errors in the collocation methods in the second collocation information.

[0104] Step 307: Generate a bill of materials for the target device based on the target collocation information.

[0105] It can be seen that the present application provides a data processing method. By obtaining a first image of a target device, the main ingredient type in the first image is identified. Then, the attribute information of each main ingredient under the main ingredient type is determined, and description information including the main ingredient type and the attribute information is generated for the main ingredient. Vector information corresponding to the description information of the main ingredient is determined, and first matching information is determined based on the vector information of the main ingredient and a first training model; second matching information is determined based on the vector information of the main ingredient and a second training model; the intersection of the first matching information and the second matching information is determined to generate target matching information; and a bill of materials for the target device is generated based on the target matching information. It can be seen that the present application determines the final target matching information based on two different training models, and thus generates a bill of materials for the target device based on the target matching information, further improving the accuracy of bill of materials generation.

[0106] Embodiment 3 of the method of the present application provides a processing method, as Figure 4 shown, the method includes the following steps:

[0107] Step 401: Obtain a first image of a target device and identify the main ingredient type in the first image;

[0108] Step 402: Determine the attribute information of each main ingredient under the main ingredient type, and generate description information including the main ingredient type and the attribute information for the main ingredient;

[0109] Step 403: Determine vector information corresponding to the description information of the main ingredient;

[0110] Step 404: Determine first matching information based on the vector information of the main ingredient and a first training model;

[0111] Among them, the first matching information includes the matching manner between the description information of the main ingredient and the description information of at least one auxiliary ingredient.

[0112] Step 405: Generate a bill of materials for the target device based on the first matching information;

[0113] Step 406: Add the bill of materials to the product bill of materials corresponding to the target device in the rule database.

[0114] Among them, the rule database is used to store various product bills of materials. After generating a bill of materials for a target device, for subsequent use convenience, the bill of materials of the target device can be added to the product bill of materials corresponding to the target device in the rule database. The product bill of materials can be a bill of materials for a device, and the device can include various devices, such as the target device can be a part of the device.

[0115] Embodiment 4 of the method of this application provides a processing method, as Figure 5 shown, the method includes the following steps:

[0116] Step 501: Obtain a first image of the target device, and identify the main material type in the first image;

[0117] Step 502: Determine the attribute information of each main material under the main material type, and generate description information for the main material, including the main material type and the attribute information;

[0118] Step 503: Determine the vector information corresponding to the description information of the main material;

[0119] Step 504: Determine the first matching information based on the vector information of the main material and the first training model;

[0120] Among them, the first matching information includes the matching method of the description information of the main material and the description information of at least one auxiliary material.

[0121] Step 505: Generate a bill of materials for the target device based on the first matching information;

[0122] Step 506: Add the bill of materials to the product bill of materials corresponding to the target device in the rule database;

[0123] Among them, the rule database is used to store various product bills of materials. After generating the bill of materials for the target device, for the convenience of subsequent use, the bill of materials of the target device can be added to the product bill of materials corresponding to the target device in the rule database. Among them, the product bill of materials can be the bill of materials for the equipment, and the equipment can include various devices, such as the target device can be a part of the equipment.

[0124] Step 507: Determine the target device;

[0125] Among them, the target device is assembled from various types of main materials and auxiliary materials, such as the target device can be a computer.

[0126] Step 508: Control the acquisition unit to acquire a second image of the interior of the target device;

[0127] The second image of the interior of the target device can be acquired by scanning the interior of the target device through the acquisition unit.

[0128] Step 509: Process the second image to obtain the material matching information inside the target device;

[0129] By recognizing the second image, the material collocation information inside the target device can be obtained, and the material collocation information may include the collocation method of the main material and the auxiliary material.

[0130] Step 510: Search for the target product material list corresponding to the target device in the rule database;

[0131] Various product material lists are stored in the rule database. Specifically, the target product material list corresponding to the identifier of the target device can be determined in the rule database based on the identifier of the target device.

[0132] Step 511: Judge whether the material collocation information is correct based on the target product material list.

[0133] Whether the recognized material collocation information is correct can be judged through the target product material list. If there is incorrect collocation information, optionally, reminders or markings can also be made.

[0134] In order to further improve the accuracy of detecting the target device, in another method embodiment of the present application, the following steps may further be included:

[0135] 1) Process the second image to obtain the position information of the main material and / or the auxiliary material;

[0136] 2) Obtain the pre-established three-dimensional model corresponding to the target device;

[0137] 3) Detect whether the position information is inserted in place based on the three-dimensional model.

[0138] Wherein, the three-dimensional model is a 3D model established based on the positions of the main materials and auxiliary materials of the target device, and can reflect the positions where each material is located. Then, through the three-dimensional model, it can be detected whether the position information of the main material and / or the auxiliary material is inserted in place. For example, whether the memory module is inserted in place, whether the wire port is okay, etc.

[0139] Through this embodiment, the accuracy of detecting the target device can be ensured, the abnormality of the target device can be reduced, and the quality and efficiency of intelligent manufacturing can be improved.

[0140] Embodiment 5 of the method of the present application provides a processing method, as Figure 6 shown, and the method includes the following steps:

[0141] Step 601: Obtain the first image of the target device and recognize the main material type in the first image;

[0142] Step 602: Determine the attribute information of each main material under the main material type, and generate description information for the main material, including the main material type and the attribute information;

[0143] Step 603: Determine the vector information corresponding to the description information of the main material;

[0144] Step 604: Determine the first matching information based on the vector information of the main material and the first training model;

[0145] Wherein, the first matching information includes the matching method of the description information of the main material and the description information of at least one auxiliary material.

[0146] Step 605: Generate a bill of materials for the target device based on the first matching information;

[0147] Step 606: Add the bill of materials to the product bill of materials corresponding to the target device in the rule database;

[0148] Step 607: Determine the modified order bill of materials and the changed material information in the order bill of materials;

[0149] Wherein, the order bill of materials is a bill of materials generated according to a customer order. When a customer changes the production order or there is a shortage of materials, etc., the order bill of materials needs to be changed, and the change of materials in the order bill of materials may affect the corresponding matching information. Therefore, this embodiment can detect the modified order bill of materials.

[0150] Step 608: Search in the rule database for the product bill of materials corresponding to the order bill of materials;

[0151] The order bill of materials and the product bill of materials have the same identifier for the same device, and the product bill of materials corresponding to the order bill of materials can be determined in the rule database.

[0152] Step 609: Determine whether the matching information of the changed material information in the order bill of materials is correct based on the product bill of materials;

[0153] Step 610: If it is incorrect, modify the order bill of materials based on the product bill of materials.

[0154] When changing the material information in the order bill of materials, the other material information matched with the changed material information can be determined through the product bill of materials, so as to correspondingly adjust the matching information in the order bill of materials, achieving the accuracy of the order material request change.

[0155] Corresponding to the above-mentioned processing method, an embodiment of the device of the present application provides an electronic device, as Figure 7 shown. The electronic device includes: a memory 710 and a processor 720; wherein:

[0156] A memory 710 for storing programs;

[0157] A processor 720 that runs the program, is used to obtain a first image of a target device, and identify the main ingredient type in the first image; determine the attribute information of each main ingredient under the main ingredient type, generate description information for the main ingredient that includes the main ingredient type and the attribute information; determine vector information corresponding to the description information of the main ingredient; determine first matching information based on the vector information of the main ingredient and a first training model; generate a bill of materials for the target device based on the first matching information.

[0158] Wherein, the first matching information includes the matching manner between the description information of the main ingredient and the description information of at least one auxiliary ingredient.

[0159] The first image is an image of a target device. This image can be an image taken of the target device, or an image taken of a paper drawing of the target device, or an electronic drawing of the target device. By identifying the first image, the main ingredient type installed on the target device can be determined.

[0160] Optionally, when the processor obtains a first image of a target device and identifies the main ingredient type in the first image, it includes: controlling an acquisition unit to take multi-angle photos of a target device installed with various components, obtaining multiple first images of the target device; using a third training model to identify the first images, and identifying the main ingredient type in the first images.

[0161] Wherein, the third training model is a deep learning training model, specifically, it can be a Fast R-CNN model, and can be used to identify the main ingredient type in an input image by pre-training a large number of images collected by the acquisition unit.

[0162] It should be noted that the third training model can not only identify the first images obtained by the acquisition unit taking multi-angle photos of a target device installed with various components, but also identify the first images obtained by the acquisition unit taking photos of the paper drawing of the target device, or identify the first images of the electronic drawing of the target device. This can all be achieved, and the training images used by the third training model during the training process can be adjusted accordingly.

[0163] Optionally, before the processor inputs the first image into the third training model, it can first mark the valid range in the first image, so as to input the first image into the third training model, and the third training model can identify the valid range in the first image.

[0164] Optionally, the processor determines first matching information based on the vector information of the main ingredient and a first training model, including: inputting the vector information of the main ingredient into the first training model, and outputting first matching vector information through the first training model; generating first matching information based on the first matching vector information.

[0165] As can be seen, the present application provides a data processing method. By acquiring a first image of a target device, the type of the main ingredient in the first image is identified, and then the attribute information of each main ingredient under the main ingredient type is determined. A description information including the main ingredient type and the attribute information is generated for the main ingredient, the vector information corresponding to the description information of the main ingredient is determined, and first matching information is determined based on the vector information of the main ingredient and a first training model. The first matching information includes the matching manner between the description information of the main ingredient and the description information of at least one auxiliary ingredient. A bill of materials for the target device is generated based on the first matching information. Thus, after acquiring the first image of the target device, the present application automatically generates a bill of materials for the target device, simplifies the user operation, and improves the accuracy of generating the bill of materials.

[0166] In the second embodiment of the device of the present application, the processor is further configured to determine second matching information based on the vector information of the main ingredient and a second training model; the second matching information includes the matching manner between the description information of the main ingredient and the description information of at least one auxiliary ingredient. Correspondingly, the processor generates a bill of materials for the target device based on the first matching information, including: determining the intersection of the first matching information and the second matching information to generate target matching information; generating a bill of materials for the target device based on the target matching information.

[0167] It should be noted that in the present application, the second training model is different from the first training model, but the functions of both are to determine matching information. In order to improve the accuracy of the matching information, the present application uses two different training models. For example, if the first training model is a BP neural network training model, then the second training model can be a decision tree training model; or if the first training model is a decision tree training model, then the second training model can be a BP neural network training model. Or, the first training model and the second training model are other different neural network training models, which can all be realized.

[0168] Optionally, the processor determines second matching information based on the vector information of the main ingredient and a second training model, including: inputting the vector information of the main ingredient into the second training model, and outputting second matching vector information through the second training model; generating second matching information based on the second matching vector information.

[0169] In the third embodiment of the device of the present application, the processor is further configured to eliminate incorrect matching methods from the second matching information based on a pre-determined matching rule, and generate the second matching information after elimination; correspondingly, the processor determines the intersection of the first matching information and the second matching information after elimination to generate a target matching method, including: determining the intersection of the first matching information and the second matching information after elimination to generate target matching information.

[0170] It can be seen that the present application provides a data processing method. By acquiring a first image of a target device, the main material type in the first image is identified, and then the attribute information of each main material under the main material type is determined, and a description information including the main material type and the attribute information for the main material is generated. The vector information corresponding to the description information of the main material is determined, and the first matching information is determined based on the vector information of the main material and a first training model; the second matching information is determined based on the vector information of the main material and a second training model; the intersection of the first matching information and the second matching information is determined to generate target matching information; and a bill of materials for the target device is generated based on the target matching information. It can be seen that the present application determines the final target matching information based on two different training models, and thus generates a bill of materials for the target device based on the target matching information, further improving the accuracy of generating the bill of materials.

[0171] In the fourth embodiment of the device of the present application, the processor is further configured to add the bill of materials to the product bill of materials corresponding to the target device in the rule database.

[0172] The rule database is used to store various product bills of materials. After generating the bill of materials for the target device, for the convenience of subsequent use, the bill of materials of the target device can be added to the product bill of materials corresponding to the target device in the rule database. The product bill of materials can be a bill of materials for a device, and the device can include various devices, such as the target device can be a part of the device.

[0173] In the fifth embodiment of the device of the present application, the processor is further configured to determine a target device; control the acquisition unit to acquire a second image inside the target device; process the second image to obtain the material matching information inside the target device; search in the rule database for the target product bill of materials corresponding to the target device; and determine whether the material matching information is correct based on the target product bill of materials.

[0174] The target device is assembled from various types of main materials and auxiliary materials.

[0175] In the sixth embodiment of the device of the present application, the processor is further configured to process the second image to obtain the position information of the main material and / or auxiliary materials; obtain a pre-established three-dimensional model corresponding to the target device; and detect whether the position information is inserted in place based on the three-dimensional model.

[0176] Wherein, the three-dimensional model is a 3D model established based on the positions of the main materials and auxiliary materials of the target device, which can reflect the positions of each material. Then, through the three-dimensional model, it can be detected whether the position information of the main material and / or auxiliary materials is inserted in place. For example, whether the memory module is inserted in place, whether the wire port is good, etc.

[0177] Through this embodiment, the accuracy of detecting the target device can be ensured, the abnormality of the target device can be reduced, and the quality and efficiency of intelligent manufacturing can be improved.

[0178] In the seventh embodiment of the device of the present application, the processor is further configured to determine the order bill of materials with modifications and the changed material information in the order bill of materials; search for the product bill of materials corresponding to the order bill of materials in the rule database; judge whether the matching information of the changed material information in the order bill of materials is correct based on the product bill of materials; if not, modify the order bill of materials based on the product bill of materials.

[0179] When changing the material information in the order bill of materials, the other material information matched with the changed material information can be determined through the product bill of materials, so as to adjust the matching information accordingly in the order bill of materials, realizing the accuracy of the order material request change.

[0180] The features described in each embodiment of this specification can be replaced or combined with each other. Each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0181] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A processing method, comprising: Obtaining a first image of a target device and identifying the main ingredient type in the first image; Determining the attribute information of each main ingredient under the main ingredient type, and generating description information for the main ingredient, including the main ingredient type and the attribute information; Determining vector information corresponding to the description information of the main ingredient; Inputting the vector information of the main ingredient into a first training model, outputting first matching vector information through the first training model, and generating first matching information based on the first matching vector information; the first matching information includes the matching method of the description information of the main ingredient and the description information of at least one auxiliary ingredient; the first training model is obtained by pre-training matching information in vector form; Generating a bill of materials for the target device based on the first matching information.

2. The method according to claim 1, further comprising: Determining second matching information based on the vector information of the main ingredient and a second training model; The second matching information includes the matching method of the description information of the main ingredient and the description information of at least one auxiliary ingredient; wherein, the second training model is different from the first training model; Correspondingly, the generating a bill of materials for the target device based on the first matching information includes: Determining the intersection of the first matching information and the second matching information to generate target matching information; Generating a bill of materials for the target device based on the target matching information.

3. The method according to claim 2, further comprising: Eliminating incorrect matching methods from the second matching information based on a pre-determined matching rule to generate the second matching information after elimination; Correspondingly, the determining the intersection of the first matching information and the second matching information to generate a target matching method includes: Determining the intersection of the first matching information and the second matching information after elimination to generate target matching information.

4. The method according to claim 1, wherein the obtaining a first image of a target device and identifying the main ingredient type in the first image includes: Controlling an acquisition unit to take multi-angle photos of a target device equipped with various components to obtain multiple first images of the target device; Using a third training model to identify the first images to identify the main ingredient type in the first images.

5. The method according to claim 1, further comprising: Adding the bill of materials to the product bill of materials corresponding to the target device in a rule database.

6. The method according to claim 5, further comprising: Determining a target device; wherein, the target device is assembled from various types of main ingredients and auxiliary ingredients; Controlling an acquisition unit to acquire a second image of the interior of the target device; Processing the second image to obtain the material matching information inside the target device; Searching in the rule database for a target product bill of materials corresponding to the target device; Judging whether the material matching information is correct based on the target product bill of materials.

7. The method according to claim 6, further comprising: Processing the second image to obtain the position information of the main ingredient and / or the auxiliary ingredient; Obtain the pre-established three-dimensional model corresponding to the target device; Based on the three-dimensional model, detect whether the position information is inserted in place.

8. The method according to claim 5, further comprising: Determine the order bill of materials that has been modified, and the changed material information in the order bill of materials; Search in the rule database for the product bill of materials corresponding to the order bill of materials; Based on the product bill of materials, judge whether the matching information of the changed material information in the order bill of materials is correct; If it is incorrect, modify the order bill of materials based on the product bill of materials.

9. An electronic device, comprising: A memory for storing programs; A processor that runs the program and is used to obtain a first image of a target device and identify the main material type in the first image; Determine the attribute information of each main material under the main material type, generate description information for the main material that includes the main material type and the attribute information; determine vector information corresponding to the description information of the main material; input the vector information of the main material into a first training model, output first matching vector information through the first training model, generate first matching information based on the first matching vector information, and the first training model is obtained by pre-training matching information in vector form; The first matching information includes the matching method of the description information of the main material and the description information of at least one auxiliary material; generate a bill of materials for the target device based on the first matching information.

Images