Decoration material delivery and acceptance method and equipment and storage medium

Abstract

The invention provides a decoration material delivery acceptance method and device, and a storage medium. The method comprises the steps of training a style migration network and a classifier; receiving multi-frame original image data actually acquired from the decoration material of the target building; collecting multi-frame model image data describing the same building material as the multi-frame original image data in the target building model; inputting the original image data into a style migration network, and converting the original image data into target image data under a modeling style of the building information model platform; inputting the target image data and the model image data into a classifier to calculate the probability that the content of the target image data is the same as that of the model image data; calculating the coverage rate of the model image data in the building model; and generating a delivery acceptance result for the decoration material of the target building according to the probability and the coverage rate. According to the embodiment, automatic preliminary delivery and acceptance can reduce the subjective influence of auditing personnel and unify the scale of delivery and acceptance, so that the efficiency of delivery and acceptance of decoration materials is improved.

Description

Technical Field

[0001] The embodiments of the present application belong to the technical field of computer vision, and in particular, relate to a method, device and storage medium for delivery and acceptance of decoration materials. Background Art

[0002] When the construction unit completes the decoration of a building, it will take multiple frames of image data of the building, upload the multiple frames of image data to the cloud, and hand them over to the audit unit for acceptance of the decoration materials. The auditors of the audit unit mainly open the design drawings of the building and compare the contents of the image data to determine whether the construction unit has performed the construction in accordance with the requirements.

[0003] This delivery and acceptance model mainly relies on auditors, and has certain requirements on the auditors' experience, which makes delivery and acceptance have a certain threshold. Moreover, auditors have a certain degree of subjectivity in acceptance, and the scale of delivery and acceptance fluctuates, resulting in low efficiency in delivery and acceptance of decoration materials. Summary of the invention

[0004] In view of this, an embodiment of the present application provides a method, device and storage medium for delivery and acceptance of decoration materials, so as to improve the efficiency of delivery and acceptance of decoration materials.

[0005] A first aspect of an embodiment of the present application provides a method for delivery and acceptance of decoration materials, which is applied to a building information modeling platform, wherein the building information modeling platform stores a target building model for modeling a target building; the method comprises:

[0006] Training a style transfer network and a classifier; the style transfer network is used to convert image data into a modeling style of the building information modeling platform; the classifier is used to identify whether the image data content under the same modeling style of the building information modeling platform is the same;

[0007] When performing a delivery and acceptance operation on the decoration materials of the target building, receiving multiple frames of original image data actually collected on the decoration materials of the target building;

[0008] Collecting multiple frames of model image data describing the same building material as the multiple frames of original image data in the target building model;

[0009] Inputting the original image data into the style transfer network to convert it into target image data under the modeling style of the building information model platform;

[0010] If the target image data and the model image data describe the same building material, inputting the target image data and the model image data into the classifier to calculate the probability that the target image data and the model image data have the same content;

[0011] Calculating the coverage of the model image data in the building model;

[0012] A delivery acceptance result is generated for the decoration materials of the target building according to the probability and the coverage rate.

[0013] A second aspect of an embodiment of the present application provides a delivery and acceptance device for decoration materials, which is applied to a building information modeling platform, wherein the building information modeling platform stores a target building model for modeling a target building; the device comprises:

[0014] A model training module, used for training a style transfer network and a classifier; the style transfer network is used for converting image data into a modeling style of the building information modeling platform; the classifier is used for identifying whether the image data content under the same modeling style of the building information modeling platform is the same;

[0015] An image data receiving module, used for receiving multiple frames of original image data actually collected from the decoration materials of the target building when performing a delivery and acceptance operation on the decoration materials of the target building;

[0016] An image data acquisition module, used for acquiring, in the target building model, multiple frames of model image data describing the same building material as the multiple frames of original image data;

[0017] An image data migration module, used for inputting the original image data into the style migration network and converting it into target image data under the modeling style of the building information model platform;

[0018] a probability calculation module, for inputting the target image data and the model image data into the classifier to calculate the probability that the target image data and the model image data have the same content if the target image data and the model image data describe the same building material;

[0019] A coverage calculation module, used for calculating the coverage of the model image data in the building model;

[0020] A delivery and acceptance module is used to generate a delivery and acceptance result for the decoration materials of the target building according to the probability and the coverage rate.

[0021] A third aspect of an embodiment of the present application provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for delivery and acceptance of decoration materials as described in the first aspect above is implemented.

[0022] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for delivery and acceptance of decoration materials as described in the first aspect above is implemented.

[0023] A fifth aspect of the embodiments of the present application provides a computer program product. When the computer program product is run on a computer, the computer is enabled to execute the method for delivery and acceptance of decoration materials described in the first aspect.

[0024] In this embodiment, a style transfer network and a classifier are trained; the style transfer network is used to convert image data into a modeling style of a building information modeling platform; the classifier is used to identify whether the image data content under the modeling style of the same building information modeling platform is the same; when performing a delivery acceptance operation on the decoration materials of a target building, multiple frames of original image data actually collected for the decoration materials of the target building are received; multiple frames of model image data describing the same building materials as the multiple frames of original image data are collected in the target building model; the original image data is input into the style transfer network and converted into target image data under the modeling style of the building information modeling platform; if the target image data and the model image data describe the same building materials, the target image data and the model image data are input into the classifier to calculate the probability that the target image data and the model image data have the same content; the coverage of the model image data in the building model is calculated; and the delivery acceptance result is generated for the decoration materials of the target building based on the probability and the coverage. This embodiment provides a standardized process for preliminary delivery acceptance of the decoration of a building, reduces the experience requirements for auditors, and reduces the threshold for delivery acceptance. Moreover, the automated preliminary delivery acceptance can reduce the subjective influence of auditors, unify the scale of delivery acceptance, and thus improve the efficiency of delivery acceptance of decoration materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or prior art descriptions. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 It is a schematic diagram of a delivery and acceptance method of decoration materials provided in an embodiment of the present application;

[0027] Figure 2 is a schematic diagram of a style transfer network and a classifier provided in an embodiment of the present application;

[0028] Figure 3It is a schematic diagram of a delivery and acceptance device for decoration materials provided in an embodiment of the present application;

[0029] Figure 4 It is a schematic diagram of a terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0030] In the following description, specific details such as specific system structures, technologies, etc. are proposed for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from hindering the description of the present application.

[0031] The technical solution of the present application is described below through specific embodiments.

[0032] Reference Figure 1 , shows a schematic diagram of a delivery and acceptance method for decoration materials provided in an embodiment of the present application, which is applied to a BIM (Building Information Management, Building Information Model) platform, such as Figure 1 As shown, the following steps may be specifically included:

[0033] Step 101: Train the style transfer network and classifier.

[0034] In an offline environment, a style transfer network and classifier can be constructed based on deep learning, and appropriate samples can be collected to train the style transfer network and classifier separately or jointly. The style transfer network is used to convert image data into the modeling style of the building information modeling platform, and the classifier is used to identify whether the image data content under the modeling style of the same building information modeling platform is the same.

[0035] In one embodiment of the present application, step 101 may include the following steps:

[0036] Step 1011: Collect first sample image data and second sample image data.

[0037] In this embodiment, the first sample image data and the second sample image data may be collected as samples, so as to jointly train the style transfer network and the classifier.

[0038] The building information model platform stores sample building models for three-dimensional modeling of sample buildings. The first sample image data is image data collected from the sample building model under certain visual conditions. The second sample image data is image data actually collected from the decoration materials of the sample building.

[0039] Step 1012: load the first generative adversarial network, the second generative adversarial network, the style transfer network, and the classifier.

[0040] In this embodiment, a first generative adversarial network, a second generative adversarial network, a style transfer network and a classifier can be loaded, wherein the first generative adversarial network and the second generative adversarial network are both generative adversarial networks (GAN), and GAN consists of a generator and a discriminator. The generator receives a random noise as input and generates simulated samples. The goal of the generator is to deceive the discriminator so that it cannot distinguish between generated samples and real samples. The discriminator receives a sample (which may be a real sample or a generated sample) as input and outputs the probability that the sample is a real sample. The goal of the discriminator is to accurately distinguish between real samples and generated samples.

[0041] During the training process, the generator and the discriminator compete with each other: the generator continuously improves the quality of the generated samples in order to deceive the discriminator; the discriminator continuously improves its discrimination ability to better distinguish between real samples and generated samples. This process eventually tends to a balance, the generator can generate high-quality samples, and the discriminator's judgment ability is also optimal.

[0042] In this embodiment, the first generative adversarial network has a first generator and a first discriminator, and the second generative adversarial network has a second generator and a second discriminator; the first generator, the second generator and the style transfer network have the same structure, and the second discriminator has the same structure as the classifier.

[0043] Among them, the first generator, the second generator and the style transfer network can all be CNNs (Convolutional Neural Networks), for example, SANet (Style Attention Network), etc. The first discriminator can be a binary classification model (especially the classification head), and the second discriminator and the classifier can both be binary classification models with certain feature extraction capabilities.

[0044] Step 1013: Under the condition of keeping the first generator unchanged, train the first discriminator according to the first sample image data and the second sample image data, so that the first discriminator is used to classify whether the image data is a modeling style of the building information modeling platform.

[0045] In this embodiment, the second generative adversarial network can be ignored and the first generative adversarial network can be trained alone. The process of training the first generative adversarial network alone can be divided into two alternating training stages.

[0046] The first training stage is to perform supervised training on the first discriminator using the first sample image data and the second sample image data as samples while keeping the first generator unchanged, so that the first discriminator has the ability to classify whether the image data is of the modeling style of the building information modeling platform.

[0047] In a specific implementation, the second sample image data may be input into the first generator, and the first generator converts the second sample image data into third sample image data in a modeling style of a building information modeling platform.

[0048] Construct a first sample pair; the first sample pair includes first sample image data and a first label representing True, and third sample image data and a second label representing False.

[0049] The first sample image data and the third sample image data in the first sample pair are respectively input into the first discriminator by random or other methods to generate first classification information; wherein the first classification information indicates whether it is a modeling style of a building information modeling platform.

[0050] Under the condition of keeping the parameters of the first generator unchanged, the parameters of the first discriminator are updated according to the first classification information and the loss value between the first label and the second label in the binary classification. At this time, the first discriminator will be punished for misclassifying true samples as false samples or misclassifying false samples as true samples. The first discriminator updates its parameters by back-propagating the loss value.

[0051] Step 1014: If the training of the first discriminator is completed, the first generator is trained according to the second sample image data while maintaining the first discriminator unchanged, so that the first generator is used to convert the image data into the modeling style of the building information modeling platform.

[0052] If the first training stage is completed (i.e., the training of the first discriminator is completed), the second training stage will be entered. Under the condition of keeping the first discriminator unchanged, the first generator is supervised trained with the second sample image data as a sample, so that the first generator has the ability to convert the image data into the modeling style of the building information modeling platform.

[0053] In a specific implementation, the second sample image data may be input into the first generator, and the first generator converts the second sample image data into fourth sample image data under the modeling style of the building information modeling platform.

[0054] The fourth sample image data is labeled with a third label (Ture) indicating that it is true.

[0055] The fourth sample image data is input into the first discriminator to generate second classification information; wherein the second classification information indicates whether it is a modeling style of a building information modeling platform.

[0056] Under the condition of keeping the parameters of the first discriminator unchanged, the parameters of the first generator are updated according to the classification loss value between the second classification information and the third label. At this time, the first discriminator and the first generator are back-propagated based on the loss value to obtain the gradient, and the parameters of the first generator are updated based on the gradient without updating the parameters of the first discriminator.

[0057] Generally, the first training stage is performed alternately with the second training stage, that is, the first discriminator is trained for one or more cycles, and the generator is trained for one or more cycles, and this is repeated until the training of the first generative adversarial network is completed.

[0058] Step 1015: If the training of the first generator is completed, the parameters of the first generator are initialized to the style transfer network and the second generator respectively.

[0059] If the training of the first generator is completed, the parameters of the first generator can be initialized to the style transfer network and the second generator respectively, that is, when initializing the style transfer network, the style transfer network inherits the parameters of the first generator, and when initializing the second generator, the second generator inherits the parameters of the first generator.

[0060] Step 1016: Under the condition of keeping the second generator unchanged, train the second discriminator based on the first sample image data and the second sample image data, so that the second discriminator is used to identify whether the image data content under the modeling style of the same building information modeling platform is the same.

[0061] When the first generative adversarial network completes training, the first generative adversarial network can be ignored and the second generative adversarial network can be trained separately. The process of separately training the second generative adversarial network is mainly to train the second discriminator.

[0062] In this process, the second generator is kept unchanged, and the second discriminator is supervisedly trained with the first sample image data and the second sample image data as samples, so that the second discriminator has the ability to identify whether the image data content under the modeling style of the same building information modeling platform is the same.

[0063] In a specific implementation, the second sample image data is input into the first generator, and the first generator converts the second sample image data into fifth sample image data under the modeling style of the building information modeling platform.

[0064] Construct a second sample pair; the second sample pair includes the first image pair and a fourth label indicating true (Ture), the second image pair and a fifth label indicating false (False); wherein the first image pair includes the first sample image data and the fifth sample image data with the same content; the second image pair includes the first sample image data and the fifth sample image data with different contents.

[0065] The first image pair and the second image pair in the second sample pair are respectively input into the second discriminator by random or other methods to generate third classification information; wherein the third classification information indicates the probability of the same content.

[0066] Under the condition of keeping the parameters of the second generator unchanged, the parameters of the second discriminator are updated according to the third classification information and the loss value between the fourth label and the fifth label in the binary classification. At this time, the second discriminator will be punished for misclassifying true samples as false samples or misclassifying false samples as true samples. The second discriminator updates its parameters through back propagation of the loss value.

[0067] Step 1017: If the training of the second discriminator is completed, the parameters of the second discriminator are initialized into the classifier.

[0068] If the second discriminator is completed, the parameters of the second discriminator may be initialized into the classifier, that is, when the classifier is initialized, the classifier inherits the parameters of the second discriminator.

[0069] In this embodiment, the style transfer network and the classifier are trained based on the first generative adversarial network and the second generative adversarial network. Not only is the training scheme mature, samples are reused, the amount of annotation is reduced, and the development cost is effectively reduced, but also the joint training makes the style transfer network and the classifier correlated. The cascaded use of the style transfer network and the classifier can improve the accuracy of the review of the delivery and acceptance of decoration materials.

[0070] In this embodiment, since the first generator, the second generator and the style transfer network have the same structure, and the second discriminator has the same structure as the classifier, the application of the first generator, the second generator and the style transfer network is basically similar, and the application of the second discriminator and the classifier is basically similar, so the description is relatively simple. For relevant details, please refer to the description of the style transfer network and classifier in step 104 and step 105. This embodiment will not be described in detail here.

[0071] Step 102: When performing a delivery and acceptance operation on the decoration materials of the target building, receiving multiple frames of original image data actually collected on the decoration materials of the target building.

[0072] When the construction unit completes the decoration of the target building, the construction personnel can use mobile terminals such as mobile phones and tablets in actual indoor or outdoor environments to collect multiple frames of original image data of the target building (mainly decoration materials). The construction personnel operate on the BIM platform to trigger the delivery and acceptance operation for the decoration materials of the target building. At this time, the multiple frames of original image data can be uploaded to the BIM platform for review.

[0073] Step 103: Collect multiple frames of model image data describing the same building material as the multiple frames of original image data in the target building model.

[0074] In the building information model platform, a target building model of three-dimensional modeling of the target building is stored. In this embodiment, corresponding multiple frames of model image data are collected in the target building model with reference to multiple frames of original image data. The so-called corresponding means that each frame of original image data and each frame of model image data describe the same building material.

[0075] In one embodiment of the present application, step 103 may include the following steps:

[0076] Step 1031: For each frame of original image data, perform a semantic segmentation operation on the original image data to obtain first semantic information representing the building material.

[0077] In this embodiment, a semantic segmentation model can be pre-built, such as U-Net (U-type network), FCN (Fully Convolutional Network), DeepLab, etc., and image data (including real image data and modeling and rendering image data) containing various building materials (types) can be used as samples to perform supervised training on the semantic segmentation model.

[0078] The original image data of each frame is input into the semantic segmentation model, and the semantic segmentation model performs a semantic segmentation operation on the original image data of each frame, and generates first semantic information representing the building material (type) for each pixel point on the original image data of each frame.

[0079] Step 1032: Convert the first semantic information into a structured first semantic sequence.

[0080] In this embodiment, the first semantic information on each frame of the original image data may be converted into a structured first semantic sequence according to a preset encoding method, so as to facilitate analysis and comparison.

[0081] Exemplarily, the original image data is segmented into a plurality of first image blocks according to a preset segmentation method, for example, each of the first image blocks is a rectangular area with a specified width and height.

[0082] In each first image block, a first proportion of the quantity of each type of first semantic information in the first image block to the quantity of all first semantic information is counted.

[0083] The first proportions of various first semantic information are compared in the same first image block, and the first semantic information with the highest first proportion is configured for the first image block.

[0084] The first semantic information configured for each first image block is arranged in a preset arrangement (such as from left to right, from top to bottom, etc.) to obtain a structured first semantic sequence.

[0085] Step 1033: Control the camera to collect model image data at multiple positions and multiple angles in the target building model.

[0086] In this embodiment, a walkable area can be defined in the target building model, and the height in the walkable area can be set according to the habit of photographing the human body. Multiple shooting points (i.e., horizontal positions) are collected on average in the walkable area. At the shooting points, the lens is positioned at a corresponding height (i.e., vertical position), and the lens is controlled to capture model image data at multiple angles.

[0087] Step 1034: Identify second semantic information representing building materials in each frame of model image data.

[0088] In this embodiment, each frame model image data is input into the semantic segmentation model, and the semantic segmentation model performs a semantic segmentation operation on each frame model image data, generating second semantic information representing the building material (type) for each pixel point on each frame model image data.

[0089] Step 1035: Convert the second semantic information into a structured second semantic sequence.

[0090] In this embodiment, the second semantic information on each frame model image number can be converted into a structured second semantic sequence in the same encoding manner to facilitate analysis and comparison.

[0091] Exemplarily, the model image data is segmented into a plurality of second image blocks according to a preset segmentation method, for example, each of the second image blocks is a rectangular area of ​​a specified width and height.

[0092] In each second image block, a second proportion of the quantity of each type of second semantic information in the second image block to the quantity of all second semantic information is counted.

[0093] The second proportions of various second semantic information in the same second image block are compared, and the second image block is configured with the second semantic information having the highest proportion.

[0094] The second semantic information configured for each second image block is arranged in a preset arrangement manner (such as from left to right, from top to bottom, etc.) to obtain a structured second semantic sequence.

[0095] In this example, the first semantic information is converted into a structured first semantic sequence, and the second semantic information is converted into a structured second semantic sequence using a block voting mode, which can reflect the general content of the original image data and the model image data, and the operation is simple and the resource consumption is low.

[0096] Step 1036: Calculate the distance between the first semantic sequence and the second semantic sequence.

[0097] In this embodiment, the distance between the first semantic sequence and the second semantic sequence, such as the Hamming distance, can be calculated to measure the similarity between the first semantic sequence and the second semantic sequence, thereby representing the similarity between the original image data and the model image data.

[0098] Step 1037: If the distance is less than or equal to a preset similarity threshold, it is determined that the original image data and the model image data describe the same building material.

[0099] The distance between the first semantic sequence and the second semantic sequence is compared with a preset similarity threshold. If the distance between the first semantic sequence and the second semantic sequence is less than or equal to the preset similarity threshold, it indicates that the degree of similarity between the original image data and the model image data is high. It can be determined that the original image data and the model image data describe the same building material.

[0100] The original image data and the model image data belong to different styles, and there are certain differences in content and texture between the two. If operators such as SIFT (Scale Invariant Feature Transform) are used to directly calculate the similarity between the original image data and the model image data, the similarity has high volatility and it is difficult to select a reasonable threshold.

[0101] In this embodiment, the target building belongs to a specific semantic environment, and the probability of semantic collision between decoration materials in different positions is low. Therefore, the similarity between the original image data and the model image data is measured based on the semantics of the decoration materials, which has good generalization, ensures the accuracy of the similarity, and reasonably selects the similarity threshold, thereby better adapting to the comparison task of original image data and model image data of different styles.

[0102] Step 104: Input the original image data into the style transfer network and convert it into target image data in the modeling style of the building information modeling platform.

[0103] In this embodiment, the original image data is input into the style transfer network, and the style transfer network converts the original image data into target image data under the modeling style of the building information modeling platform, so as to compare whether the content is the same as the model image data under the same style.

[0104] Since the modeling capability of the building information modeling platform is predictable, the style transfer network has good transfer capabilities and can transfer the original image data under the real style to the virtual style (i.e., the modeling style of the building information modeling platform). This can effectively reduce the information bias during transfer, thereby ensuring the accuracy of comparing the content with the model image data under the same style.

[0105] On the contrary, the decoration capabilities of construction units in the real world are not predictable. If the model image data in the virtual style (i.e., the modeling style of the building information modeling platform) is migrated to the real style, the information deviation during the migration will be large.

[0106] In a style transfer network structure, such as Figure 2 As shown, the style transfer network includes a first convolution block ConvBlock_1, a second convolution block ConvBlock_2, a third convolution block ConvBlock_3 and a fully connected block FCBlock; wherein, the first convolution block ConvBlock_1 has two convolutional layers, the second convolution block ConvBlock_2 has two convolutional layers, the third convolution block ConvBlock_3 has four convolutional layers, and the fully connected block FCBlock has two fully connected layers.

[0107] Furthermore, the convolution kernels of the convolution layers in the first convolution block ConvBlock_1, the second convolution block ConvBlock_2, and the third convolution block ConvBlock_3 are all 3× × 3. Use ReLU (Rectified Linear Unit) as the activation function.

[0108] In this structure, the original image data is input into the first convolution block ConvBlock_1, and the encoding operation is performed through two convolution layers to obtain the first image style feature. When the size of the original image data is 224××224××3, the size of the first image style feature is 224××224××64.

[0109] Perform a maximum pooling operation on the first image style feature to obtain the second image style feature. The filter of the maximum pooling operation is 2× ×2, the step size is 2, and the size of the second image style feature is 112××112××64.

[0110] The second image style feature is input into the second convolution block ConvBlock_2, and the encoding operation is performed through two convolution layers to obtain the third image style feature. The size of the third image style feature is 112××112××128.

[0111] The third image style feature is subjected to a maximum pooling operation (Max Pooling) to obtain a fourth image style feature. The size of the fourth image style feature is 56× × 56× × 128.

[0112] The fourth image style feature is input into the third convolution block ConvBlock_3, and the decoding operation is performed through four convolution layers to obtain the fifth image style feature. The size of the fifth image style feature is 56× × 56× × 256.

[0113] The fifth image style feature is subjected to a maximum pooling operation (Max Pooling) to obtain the sixth image style feature. The size of the sixth image style feature is 28× × 28× × 256.

[0114] The sixth image style feature is input into the fully connected block FCBlock, and a decoding operation is performed through two fully connected layers to obtain the seventh image style feature.

[0115] The seventh image style feature is activated using functions such as Softmax to obtain target image data under the modeling style of the building information modeling platform.

[0116] Step 105: If the target image data and the model image data describe the same building material, the target image data and the model image data are input into a classifier to calculate the probability that the target image data and the model image data have the same content.

[0117] In this embodiment, the corresponding target image data and model image data are aligned based on the original image data. That is, for the original image data and model image data describing the same building material, the target image data and model image data after the style transfer of the original image data are combined into an image pair, so that the target image data and the model image data in the image pair describe the same building material.

[0118] At this time, the target image data and the model image data are input into the classifier together, and the classifier calculates the probability that the target image data and the model image data have the same content.

[0119] The so-called target image data and the model image data having the same content may refer to the target image data and the model image data having the same texture of the decoration material, the target image data and the model image data having the same color of the rusted material, and so on.

[0120] In a classifier structure, such as Figure 2 As shown, the classifier includes a first branch structure Branch_1, a second branch structure Branch_2, a backbone structure Backbone and a classification head structure Head; wherein, the first branch structure Branch_1 has four convolutional layers, the second branch structure Branch_2 has four convolutional layers, the backbone structure Backbone has three convolutional layers, and the classification head structure Head has three fully connected layers.

[0121] Furthermore, the convolution kernels of the convolution layers in the first branch structure Branch_1, the second branch structure Branch_2, and the backbone structure Backbone are all 3× × 3. Use ReLU as the activation function.

[0122] In this structure, the target image data is input into the first branch structure Branch_1, and features are extracted through four convolutional layers, which are recorded as the first image content features.

[0123] The model image data is input into the second branch structure Branch_2, and features are extracted through four convolutional layers, which is recorded as extracting the second image content features;

[0124] The first image content feature and the second image content feature are concatenated into a new feature, which is recorded as the third image content feature.

[0125] The third image content feature is input into the backbone structure Backbone, and features are further extracted through three convolutional layers, which are recorded as the fourth image content feature.

[0126] The fourth image content feature is input into the classification head structure Head to perform a binary classification operation, and the fourth image content feature is mapped to the probability that the target image data and the model image data have the same content through three fully connected layers.

[0127] Step 106: Calculate the coverage of the model image data in the building model.

[0128] In this embodiment, since there are differences in angles when construction workers collect model image data, there are differences in the decoration materials of the model image data. Therefore, multiple frames of model image data can be traversed to calculate the coverage of each frame of model image data in the building model by projection, expanded area, etc., to facilitate the review of decoration materials.

[0129] In a specific implementation, by traversing the building model, the building model can be vertically projected onto a horizontal plane to obtain an overall projection area, and, for each frame of model image data, the structure representing the model image data in the target building model can be vertically projected onto a horizontal plane to obtain a local projection area.

[0130] For each frame of model image data, the ratio between the area of ​​the local projection region and the area of ​​the overall projection region is calculated to obtain the area ratio.

[0131] Calculate the degree of dispersion (DOD) of multiple local projection areas in the overall projection area, such as log-likelihood ratio.

[0132] For each frame of model image data, the area ratio is multiplied by the sum of a preset dispersion coefficient (such as 0.5) and the dispersion degree to obtain the coverage of the model image data in the building model.

[0133] Then, coverage can be expressed as Coverage = SP × × (DP+DOD), where Coverage is coverage, SP is area ratio, DP is dispersion coefficient, and DOD is dispersion degree.

[0134] Step 107: Generate a delivery acceptance result for the decoration materials of the target building based on the probability and coverage rate.

[0135] In this embodiment, a preliminary review of the decoration materials of the target building is conducted based on the probability of identical content between the multi-frame target image data and the multi-frame model image data and their corresponding coverage rate, and a delivery acceptance result is generated and provided to the review unit as a reference for delivery acceptance.

[0136] In the specific implementation, for multiple frames of model image data, the probability and coverage are substituted into the following formula to calculate the confidence that the building's decoration materials meet the design requirements:

[0137]

[0138] Among them, Score is the confidence, n is the number of model image data, C C overage i is the coverage rate corresponding to the i-th frame model image data, ΔC C Overage is the preset compensation coefficient (when the vertical projection area is the main area, the compensation coefficient can be used to compensate for the unfolded area), P i is the probability corresponding to the model image data of the i-th frame.

[0139] The confidence level is compared to a preset confidence threshold.

[0140] If the confidence level is greater than or equal to a preset confidence threshold, the delivery and acceptance result of the building's decoration materials is determined to be a successful delivery and acceptance.

[0141] If the confidence level is less than a preset confidence threshold, the delivery acceptance result of the building's decoration materials is determined to be a delivery acceptance failure.

[0142] In addition, in addition to acceptance success or acceptance failure, other information may also be included in the delivery acceptance result, such as confidence, marking the structure involved in the original image data in the target building model, using LLM (Large Language Model) to describe the situation of decoration materials in the original image data, etc. This embodiment does not limit this.

[0143] In this embodiment, a style transfer network and a classifier are trained; the style transfer network is used to convert image data into a modeling style of a building information modeling platform; the classifier is used to identify whether the image data content under the modeling style of the same building information modeling platform is the same; when performing a delivery acceptance operation on the decoration materials of a target building, multiple frames of original image data actually collected for the decoration materials of the target building are received; multiple frames of model image data describing the same building materials as the multiple frames of original image data are collected in the target building model; the original image data is input into the style transfer network and converted into target image data under the modeling style of the building information modeling platform; if the target image data and the model image data describe the same building materials, the target image data and the model image data are input into the classifier to calculate the probability that the target image data and the model image data have the same content; the coverage of the model image data in the building model is calculated; and the delivery acceptance result is generated for the decoration materials of the target building based on the probability and the coverage. This embodiment provides a standardized process for preliminary delivery acceptance of the decoration of a building, reduces the experience requirements for auditors, and reduces the threshold for delivery acceptance. Moreover, the automated preliminary delivery acceptance can reduce the subjective influence of auditors, unify the scale of delivery acceptance, and thus improve the efficiency of delivery acceptance of decoration materials.

[0144] It should be noted that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0145] Reference Figure 3 , showing a schematic diagram of a delivery and acceptance device for decoration materials provided in an embodiment of the present application, which is applied to a building information modeling platform, wherein the building information modeling platform stores a target building model for modeling a target building; Figure 3 As shown, the device may specifically include the following modules:

[0146] The model training module 301 is used to train a style transfer network and a classifier; the style transfer network is used to convert image data into a modeling style of the building information modeling platform; the classifier is used to identify whether the image data content under the same modeling style of the building information modeling platform is the same;

[0147] The image data receiving module 302 is used to receive multiple frames of original image data actually collected from the decoration materials of the target building when performing a delivery and acceptance operation on the decoration materials of the target building;

[0148] An image data acquisition module 303 is used to acquire, in the target building model, multiple frames of model image data describing the same building material as the multiple frames of original image data;

[0149] An image data migration module 304 is used to input the original image data into the style migration network to convert it into target image data under the modeling style of the building information model platform;

[0150] A probability calculation module 305 is used for inputting the target image data and the model image data into the classifier to calculate the probability that the target image data and the model image data have the same content if the target image data and the model image data describe the same building material;

[0151] A coverage calculation module 306, used to calculate the coverage of the model image data in the building model;

[0152] The delivery and acceptance module 307 is used to generate a delivery and acceptance result for the decoration materials of the target building according to the probability and the coverage rate.

[0153] Optionally, the model training module 301 includes:

[0154] A sample collection module, used to collect first sample image data and second sample image data; the first sample image data is image data collected from the sample building model, and the second sample image data is image data actually collected from the decoration materials of the sample building;

[0155] A model loading module, used to load a first generative adversarial network, a second generative adversarial network, a style transfer network and a classifier; the first generative adversarial network has a first generator and a first discriminator, and the second generative adversarial network has a second generator and a second discriminator; the first generator, the second generator and the style transfer network have the same structure, and the second discriminator has the same structure as the classifier;

[0156] a first discriminator training module, configured to train the first discriminator according to the first sample image data and the second sample image data while maintaining the first generator unchanged, so that the first discriminator is used to classify whether the image data is a modeling style of the building information modeling platform;

[0157] A generator training module, configured to, if the training of the first discriminator is completed, train the first generator according to the second sample image data while maintaining the first discriminator unchanged, so that the first generator is used to convert the image data into a modeling style of the building information modeling platform;

[0158] A first initialization module, configured to initialize the parameters of the first generator to the style transfer network and the second generator respectively if the training of the first generator is completed;

[0159] a second discriminator training module, used for training the second discriminator according to the first sample image data and the second sample image data under the condition of keeping the second generator unchanged, so that the second discriminator is used to identify whether the image data contents under the modeling style of the same building information modeling platform are the same;

[0160] The second initialization module is used to initialize the parameters of the second discriminator into the classifier if the training of the second discriminator is completed.

[0161] Optionally, the first discriminator training module is further used for:

[0162] Inputting the second sample image data into the first generator to convert it into third sample image data in the modeling style of the building information model platform;

[0163] Constructing a first sample pair; the first sample pair includes the first sample image data and a first label indicating true, and the third sample image data and a second label indicating false;

[0164] Inputting the first sample image data and the third sample image data in the first sample pair into the first discriminator to generate first classification information; the first classification information indicates whether it is the modeling style of the building information modeling platform;

[0165] Under the condition of keeping the parameters of the first generator unchanged, updating the parameters of the first discriminator according to the first classification information and the loss value between the first label and the second label;

[0166] The generator training module is also used to:

[0167] Inputting the second sample image data into the first generator to convert it into fourth sample image data in the modeling style of the building information modeling platform;

[0168] Annotating the fourth sample image data with a third label indicating true;

[0169] Inputting the fourth sample image data into the first discriminator to generate second classification information; the second classification information indicates whether it is the modeling style of the building information modeling platform;

[0170] Under the condition of keeping the parameters of the first discriminator unchanged, updating the parameters of the first generator according to the loss value between the second classification information and the third label;

[0171] The second discriminator training module is also used for:

[0172] Inputting the second sample image data into the first generator to convert it into fifth sample image data in the modeling style of the building information modeling platform;

[0173] Constructing a second sample pair; the second sample pair includes the first image pair and a fourth label indicating true, and the second image pair and a fifth label indicating false; the first image pair includes the first sample image data and the fifth sample image data having the same content; the second image pair includes the first sample image data and the fifth sample image data having different contents;

[0174] Inputting the first image pair and the second image pair in the second sample pair into the second discriminator to generate third classification information; the third classification information indicates the probability of having the same content;

[0175] Under the condition of keeping the parameters of the second generator unchanged, the parameters of the second discriminator are updated according to the third classification information and the loss value between the fourth label and the fifth label.

[0176] Optionally, the image data acquisition module 303 includes:

[0177] A first semantic segmentation module is used to perform a semantic segmentation operation on the original image data of each frame to obtain first semantic information representing the building material;

[0178] A first semantic sequence conversion module, used for converting the first semantic information into a structured first semantic sequence;

[0179] A model acquisition module, used to control the camera to acquire model image data at multiple positions and multiple angles in the target building model;

[0180] A second semantic segmentation module, used for identifying second semantic information representing building materials in each frame of the model image data;

[0181] A second semantic sequence conversion module, used for converting the second semantic information into a structured second semantic sequence;

[0182] A semantic distance calculation module, used for calculating the distance between the first semantic sequence and the second semantic sequence;

[0183] The description determination module is used to determine that the original image data and the model image data describe the same building material if the distance is less than or equal to a preset similarity threshold.

[0184] Optionally, the first semantic sequence conversion module is further used for:

[0185] Dividing the original image data into a plurality of first image blocks according to a preset dividing method;

[0186] Counting first proportions of various types of the first semantic information in the first image block;

[0187] configuring the first image block with the first semantic information having the highest first proportion;

[0188] Arranging the first semantic information configured for each of the first image blocks according to a preset arrangement method to obtain a structured first semantic sequence;

[0189] The second semantic sequence conversion module is also used for:

[0190] Dividing the model image data into a plurality of second image blocks according to a preset dividing method;

[0191] Counting second proportions of various types of the second semantic information in the second image block;

[0192] configuring the second image block with the second semantic information having the second highest proportion;

[0193] The second semantic information configured for each of the second image blocks is arranged in a preset arrangement manner to obtain a structured second semantic sequence.

[0194] Optionally, the style transfer network includes a first convolution block, a second convolution block, a third convolution block and a fully connected block; the first convolution block has two convolution layers, the second convolution block has two convolution layers, the third convolution block has four convolution layers, and the fully connected block has two fully connected layers;

[0195] The image data migration module 304 is also used for:

[0196] Inputting the original image data into the first convolution block to perform an encoding operation to obtain a first image style feature;

[0197] Performing a maximum pooling operation on the first image style feature to obtain a second image style feature;

[0198] Inputting the second image style feature into the second convolution block to perform an encoding operation to obtain a third image style feature;

[0199] Performing a maximum pooling operation on the third image style feature to obtain a fourth image style feature;

[0200] Inputting the fourth image style feature into the third convolution block to perform a decoding operation to obtain a fifth image style feature;

[0201] Performing a maximum pooling operation on the fifth image style feature to obtain a sixth image style feature;

[0202] Inputting the sixth image style feature into the fully connected block to perform a decoding operation to obtain a seventh image style feature;

[0203] An activation operation is performed on the seventh image style feature to obtain target image data under the modeling style of the building information modeling platform.

[0204] Optionally, the classifier includes a first branch structure, a second branch structure, a trunk structure and a classification head structure; the first branch structure has four convolutional layers, the second branch structure has four convolutional layers, the trunk structure has three convolutional layers, and the classification head structure has three fully connected layers;

[0205] The probability calculation module 305 is also used for:

[0206] Inputting the target image data into the first branch structure to extract first image content features;

[0207] Inputting the model image data into the second branch structure to extract second image content features;

[0208] splicing the first image content feature and the second image content feature into a third image content feature;

[0209] Inputting the third image content feature into the backbone structure to extract a fourth image content feature;

[0210] The fourth image content feature is input into the classification head structure to be mapped into the probability that the target image data has the same content as the model image data.

[0211] Optionally, the coverage calculation module 306 is further used for:

[0212] Projecting the building model vertically onto a horizontal plane to obtain an overall projection area;

[0213] For each frame of the model image data, vertically projecting the structure representing the model image data in the target building model onto a horizontal plane to obtain a local projection area;

[0214] For each frame of the model image data, calculating the ratio between the area of ​​the local projection area and the area of ​​the overall projection area to obtain an area ratio;

[0215] Calculating the dispersion of the plurality of local projection areas in the overall projection area;

[0216] For each frame of the model image data, multiply the area ratio by the sum of a preset dispersion coefficient and the dispersion degree to obtain a coverage rate of the model image data in the building model;

[0217] The delivery acceptance module 307 is also used for:

[0218] For multiple frames of the model image data, the probability and the coverage rate are substituted into the following formula to calculate the confidence that the decoration materials of the building meet the design requirements:

[0219]

[0220] Wherein, Score is the confidence, n is the number of model image data, C C overage i is the coverage rate corresponding to the model image data of the i-th frame, ΔC C Overage is the preset compensation coefficient, P i is the probability corresponding to the model image data of the i-th frame;

[0221] If the confidence level is greater than or equal to a preset confidence threshold, determining that the delivery and acceptance result of the decoration materials of the building is a successful delivery and acceptance;

[0222] If the confidence level is less than a preset confidence threshold, it is determined that the delivery and acceptance result of the decoration materials of the building is a delivery and acceptance failure.

[0223] An embodiment of the present application provides a delivery and acceptance device for decoration materials. By using this device, each step in the aforementioned method embodiments can be implemented.

[0224] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment part.

[0225] Reference Figure 4 , shows a schematic diagram of a terminal device provided in an embodiment of the present application. Figure 4 As shown, the terminal device 400 in the embodiment of the present application includes: a processor 410, a memory 420, and a computer program 421 stored in the memory 420 and executable on the processor 410. When the processor 410 executes the computer program 421, the steps in each embodiment of the above-mentioned method for delivery and acceptance of decoration materials are implemented. Alternatively, when the processor 410 executes the computer program 421, the functions of each module / unit in the above-mentioned device embodiments are implemented.

[0226] Exemplarily, the computer program 421 may be divided into one or more modules / units, which are stored in the memory 420 and executed by the processor 410 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which may be used to describe the execution process of the computer program 421 in the terminal device 400.

[0227] The terminal device 400 may include, but is not limited to, a processor 410 and a memory 420. Those skilled in the art will appreciate that Figure 4 It is only an example of the terminal device 400 and does not constitute a limitation on the terminal device 400. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device 400 may also include input and output devices, network access devices, buses, etc.

[0228] The processor 410 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0229] The memory 420 may be an internal storage unit of the terminal device 400, such as a hard disk or memory of the terminal device 400. The memory 420 may also be an external storage device of the terminal device 400, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device 400. Further, the memory 420 may also include both an internal storage unit of the terminal device 400 and an external storage device. The memory 420 is used to store the computer program 421 and other programs and data required by the terminal device 400. The memory 420 may also be used to temporarily store data that has been output or is to be output.

[0230] An embodiment of the present application also discloses a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for delivery and acceptance of decoration materials as described in the aforementioned embodiments is implemented.

[0231] The embodiment of the present application further discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the delivery and acceptance method of decoration materials as described in the above-mentioned embodiments is implemented.

[0232] The embodiment of the present application further discloses a computer program product. When the computer program product is run on a computer, the computer is enabled to execute the delivery and acceptance method of decoration materials described in the aforementioned embodiments.

[0233] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application is described in detail with reference to the above-mentioned embodiments, a person skilled in the art should understand that the technical solutions described in the above-mentioned embodiments can still be modified, or some of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for delivery and acceptance of decoration materials, characterized in that: Applied to a building information model platform, the building information model platform stores a target building model for modeling a target building; the method comprises: Training a style transfer network and a classifier; the style transfer network is used to convert image data into a modeling style of the building information modeling platform; the classifier is used to identify whether the image data content under the same modeling style of the building information modeling platform is the same; When performing a delivery and acceptance operation on the decoration materials of the target building, receiving multiple frames of original image data actually collected on the decoration materials of the target building; Collecting multiple frames of model image data describing the same building material as the multiple frames of original image data in the target building model; Inputting the original image data into the style transfer network to convert it into target image data under the modeling style of the building information model platform; If the target image data and the model image data describe the same building material, inputting the target image data and the model image data into the classifier to calculate the probability that the target image data and the model image data have the same content; Calculating the coverage of the model image data in the building model; A delivery acceptance result is generated for the decoration materials of the target building according to the probability and the coverage rate.

2. The method according to claim 1, characterized in that: The building information model platform stores a sample building model for modeling a sample building; The training style transfer network and classifier include: Collecting first sample image data and second sample image data; the first sample image data is image data collected from the sample building model, and the second sample image data is image data actually collected from the decoration materials of the sample building; Loading a first generative adversarial network, a second generative adversarial network, a style transfer network and a classifier; the first generative adversarial network has a first generator and a first discriminator, and the second generative adversarial network has a second generator and a second discriminator; the first generator, the second generator and the style transfer network have the same structure, and the second discriminator has the same structure as the classifier; Under the condition of keeping the first generator unchanged, training the first discriminator according to the first sample image data and the second sample image data, so that the first discriminator is used to classify whether the image data is a modeling style of the building information modeling platform; If the training of the first discriminator is completed, the first generator is trained according to the second sample image data while maintaining the first discriminator unchanged, so that the first generator is used to convert the image data into the modeling style of the building information modeling platform; If the training of the first generator is completed, the parameters of the first generator are initialized to the style transfer network and the second generator respectively; Under the condition of keeping the second generator unchanged, training the second discriminator according to the first sample image data and the second sample image data, so that the second discriminator is used to identify whether the image data contents under the modeling style of the same building information modeling platform are the same; If the training of the second discriminator is completed, the parameters of the second discriminator are initialized into the classifier.

3. The method according to claim 2, characterized in that The step of training the first discriminator according to the first sample image data and the second sample image data under the condition of maintaining the first generator unchanged, so that the first discriminator is used to classify whether the image data is a modeling style of the building information modeling platform, includes: Inputting the second sample image data into the first generator to convert it into third sample image data in the modeling style of the building information model platform; Constructing a first sample pair; the first sample pair includes the first sample image data and a first label indicating true, and the third sample image data and a second label indicating false; Inputting the first sample image data and the third sample image data in the first sample pair into the first discriminator to generate first classification information; the first classification information indicates whether it is the modeling style of the building information modeling platform; Under the condition of keeping the parameters of the first generator unchanged, updating the parameters of the first discriminator according to the first classification information, the loss value between the first label and the second label; The step of training the first generator according to the second sample image data while maintaining the first discriminator unchanged, so that the first generator is used to convert the image data into the modeling style of the building information modeling platform, comprises: Inputting the second sample image data into the first generator to convert it into fourth sample image data in the modeling style of the building information modeling platform; Annotating the fourth sample image data with a third label indicating true; Inputting the fourth sample image data into the first discriminator to generate second classification information; the second classification information indicates whether it is the modeling style of the building information modeling platform; Under the condition of keeping the parameters of the first discriminator unchanged, updating the parameters of the first generator according to the loss value between the second classification information and the third label; The method of training the second discriminator according to the first sample image data and the second sample image data under the condition of keeping the second generator unchanged, so that the second discriminator is used to identify whether the image data contents under the modeling style of the same building information modeling platform are the same, comprises: Inputting the second sample image data into the first generator to convert it into fifth sample image data in the modeling style of the building information modeling platform; Constructing a second sample pair; the second sample pair includes the first image pair and a fourth label indicating true, and the second image pair and a fifth label indicating false; the first image pair includes the first sample image data and the fifth sample image data having the same content; the second image pair includes the first sample image data and the fifth sample image data having different contents; Inputting the first image pair and the second image pair in the second sample pair into the second discriminator to generate third classification information; the third classification information indicates the probability of having the same content; Under the condition of keeping the parameters of the second generator unchanged, the parameters of the second discriminator are updated according to the third classification information, the loss value between the fourth label and the fifth label.

4. The method according to claim 1, characterized in that The step of collecting multiple frames of model image data describing the same building material as the multiple frames of original image data in the target building model comprises: For each frame of the original image data, a semantic segmentation operation is performed on the original image data to obtain first semantic information representing the building material; converting the first semantic information into a structured first semantic sequence; Controlling the camera to collect model image data at multiple positions and multiple angles in the target building model; identifying second semantic information representing building materials in each frame of the model image data; converting the second semantic information into a structured second semantic sequence; Calculating a distance between the first semantic sequence and the second semantic sequence; If the distance is less than or equal to a preset similarity threshold, it is determined that the original image data and the model image data describe the same building material.

5. The method according to claim 4, characterized in that The converting the first semantic information into a structured first semantic sequence includes: Dividing the original image data into a plurality of first image blocks according to a preset dividing method; Counting first proportions of various types of the first semantic information in the first image block; configuring the first image block with the first semantic information having the highest first proportion; Arranging the first semantic information configured for each of the first image blocks according to a preset arrangement method to obtain a structured first semantic sequence; The converting the second semantic information into a structured second semantic sequence includes: Dividing the model image data into a plurality of second image blocks according to a preset dividing method; Counting second proportions of various types of the second semantic information in the second image block; configuring the second image block with the second semantic information having the second highest proportion; The second semantic information configured for each of the second image blocks is arranged in a preset arrangement manner to obtain a structured second semantic sequence.

6. The method according to any one of claims 1 to 5, characterized in that The style transfer network includes a first convolution block, a second convolution block, a third convolution block and a fully connected block; the first convolution block has two convolution layers, the second convolution block has two convolution layers, the third convolution block has four convolution layers, and the fully connected block has two fully connected layers; The step of inputting the original image data into the style transfer network to convert the original image data into target image data in the modeling style of the building information model platform includes: Inputting the original image data into the first convolution block to perform an encoding operation to obtain a first image style feature; Performing a maximum pooling operation on the first image style feature to obtain a second image style feature; Inputting the second image style feature into the second convolution block to perform an encoding operation to obtain a third image style feature; Performing a maximum pooling operation on the third image style feature to obtain a fourth image style feature; Inputting the fourth image style feature into the third convolution block to perform a decoding operation to obtain a fifth image style feature; Performing a maximum pooling operation on the fifth image style feature to obtain a sixth image style feature; Inputting the sixth image style feature into the fully connected block to perform a decoding operation to obtain a seventh image style feature; An activation operation is performed on the seventh image style feature to obtain target image data under the modeling style of the building information modeling platform.

7. The method according to claim 6, characterized in that The classifier includes a first branch structure, a second branch structure, a trunk structure and a classification head structure; the first branch structure has four convolutional layers, the second branch structure has four convolutional layers, the trunk structure has three convolutional layers, and the classification head structure has three fully connected layers; The step of inputting the target image data and the model image data into the classifier to calculate the probability that the target image data and the model image data have the same content comprises: Inputting the target image data into the first branch structure to extract first image content features; Inputting the model image data into the second branch structure to extract second image content features; splicing the first image content feature and the second image content feature into a third image content feature; Inputting the third image content feature into the backbone structure to extract a fourth image content feature; The fourth image content feature is input into the classification head structure to be mapped into the probability that the target image data has the same content as the model image data.

8. The method according to any one of claims 1 to 5, characterized in that The calculating the coverage rate of the model image data in the building model comprises: Projecting the building model vertically onto a horizontal plane to obtain an overall projection area; For each frame of the model image data, vertically projecting the structure representing the model image data in the target building model onto a horizontal plane to obtain a local projection area; For each frame of the model image data, calculating the ratio between the area of ​​the local projection area and the area of ​​the overall projection area to obtain an area ratio; Calculating the dispersion of the plurality of local projection areas in the overall projection area; For each frame of the model image data, multiply the area ratio by the sum of a preset dispersion coefficient and the dispersion degree to obtain a coverage rate of the model image data in the building model; Generating a delivery acceptance result for the decoration materials of the target building according to the probability and the coverage rate includes: For multiple frames of the model image data, the probability and the coverage rate are substituted into the following formula to calculate the confidence that the decoration materials of the building meet the design requirements: Among them, Score is the confidence, n is the number of model image data, CCoverage i is the coverage rate corresponding to the model image data of the i-th frame, ΔCCoverage is the preset compensation coefficient, P i is the probability corresponding to the model image data of the i-th frame; If the confidence level is greater than or equal to a preset confidence threshold, determining that the delivery and acceptance result of the decoration materials of the building is a successful delivery and acceptance; If the confidence level is less than a preset confidence threshold, it is determined that the delivery and acceptance result of the decoration materials of the building is a delivery and acceptance failure.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for delivery and acceptance of decoration materials as described in any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for delivery and acceptance of decoration materials as described in any one of claims 1 to 8 is implemented.

Images