Method and device for recognizing style and appearance design drawings of overseas countries, terminal equipment and storage medium

Abstract

The invention discloses a method and device for recognizing a design drawing of the style and appearance of a countryside, terminal equipment and a storage medium, and the method comprises the steps: predicting a style type and a style score of the design drawing through a style recognition and scoring model, and calculating the style of the design drawing according to the predicted style type by referring to preset reference basic components and reference space design elements; and further calculating a basic component score and a spatial design element score of the design drawing, according to the style score, the basic component score and the spatial design element score, calculating a comprehensive score of the design drawing, according to the comprehensive score, judging whether the design drawing accords with the style design of the countryside overseas countryside or not, and determining a corresponding style type. According to the method, whether the design drawing is the style and appearance design drawing of the overseas countryside or not can be automatically identified, the specific style and type of the overseas countryside can be identified, the technical problems of large errors and time and labor consumption caused by dependence on manual identification of the design drawing in the prior art are solved, manpower is saved, and efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of picture recognition, and in particular to a method, device, terminal equipment and storage medium for recognizing a design drawing of an overseas Chinese hometown. Background Art

[0002] At present, in the construction industry, when studying the culture of hometowns of overseas Chinese, it is necessary to check whether the architectural design drawings conform to the design style of hometowns of overseas Chinese. In this case, it is usually necessary to manually compare the various architectural structures and decorative elements, etc., and judge whether the design drawings conform to the design style of hometowns of overseas Chinese. The manual judgment process has certain subjective errors and consumes time and manpower. Summary of the invention

[0003] The present invention provides a method, device, terminal equipment and storage medium for recognizing design drawings of an overseas Chinese hometown, so as to solve the technical problems of large errors and time-consuming and labor-intensive caused by relying on manual recognition of design drawings in the prior art.

[0004] In order to solve the above technical problems, an embodiment of the present invention provides a method for identifying a design drawing of an overseas Chinese hometown, including:

[0005] Obtaining a design drawing to be identified;

[0006] Input the design drawing into a trained style recognition and scoring model, so that the style recognition and scoring model outputs the style type of the design drawing and the corresponding style score; the style score is used to quantify the similarity between the real style type of the design drawing and the predicted style type;

[0007] According to the design drawing, a number of basic components and space design elements are extracted; wherein the types of the basic components include: doors, windows, columns, patios and roofs; the space design elements include: building form, building orientation and landscape;

[0008] According to the style type, a corresponding reference basic component is selected from a preset basic component library; for each basic component, a basic component similarity between the basic component and the reference basic component is calculated; according to the similarities of all basic components and a preset weight value of each basic component, a basic component score of the design drawing is calculated;

[0009] According to the style type, a corresponding reference space design element is selected from a preset space design element library; for each space design element, a space design element similarity between the space design element and the reference space design element is calculated; according to the similarities of all space design elements and a preset weight value of each space design element, a space design element score of the design drawing is calculated;

[0010] The comprehensive score of the design drawing is calculated based on the style score, the basic component score, the space design element score and the preset weight values ​​of each score; it is determined whether the comprehensive score reaches the preset comprehensive score threshold corresponding to the style type; if so, it is determined that the design drawing is a hometown of overseas Chinese style design drawing, and the design drawing is determined to be a design drawing of the style type; if not, it is determined that the design drawing is not a hometown of overseas Chinese style design drawing.

[0011] As a preferred solution, the training process of the style recognition and scoring model includes:

[0012] Obtaining a number of sample design drawings and the style type and corresponding style score of each sample design drawing;

[0013] According to the style type and style score of each sample design drawing, set the label of the sample design drawing;

[0014] An initial style recognition and scoring model is constructed, the sample design drawing is used as input, the predicted style type of the sample design drawing and the corresponding style score are used as output, and the style recognition and scoring model is trained to obtain a trained style recognition and scoring model.

[0015] As a preferred solution, the style recognition and scoring model is a VGG convolutional neural network; all convolutional layers of the VGG convolutional neural network use a small-size convolution kernel of 3×3.

[0016] As a preferred solution, when training the style recognition and scoring model, the sample design drawing is input into the style recognition and scoring model to obtain a predicted style type and a corresponding style score as a prediction result;

[0017] According to the preset loss function, calculate the loss function value of the prediction result and the corresponding label;

[0018] The style scoring model is optimized according to the loss function value so that the loss function value converges.

[0019] As a preferred solution, the loss function is a cross entropy loss function.

[0020] As a preferred solution, the calculating the basic component similarity between the basic component and the reference basic component includes:

[0021] Extracting the pattern, decoration type and shape of the basic component;

[0022] calculating a pattern similarity between the pattern and the pattern of the reference base component;

[0023] calculating a decoration type similarity between the decoration type and the decoration type of the reference base component;

[0024] Calculating the shape similarity between the shape and the shape of the reference basic component;

[0025] The basic component similarity is calculated according to the pattern similarity, the decoration type similarity and the shape similarity, a preset pattern similarity weight value, a decoration type similarity weight value and a shape similarity weight value.

[0026] As a preferred solution, the calculating the spatial design element similarity between the spatial design element and the reference spatial design element includes:

[0027] Calculating the architectural form similarity between the architectural form of the spatial design element and the architectural form of the reference spatial design element;

[0028] Calculating the architectural orientation similarity between the architectural orientation of the spatial design element and the architectural orientation of the reference spatial design element;

[0029] Calculating the landscape similarity between the landscape of the spatial design element and the landscape of the reference spatial design element;

[0030] The spatial design element similarity is calculated according to the building form similarity, the building orientation similarity, the landscape similarity, a preset building form similarity weight value, a preset building orientation similarity weight value and a preset landscape similarity weight value.

[0031] On the basis of the above-mentioned embodiment, another embodiment of the present invention provides a device for identifying design drawings of hometown style, including: a design drawing acquisition module, a style recognition and scoring module, a preprocessing module, a basic component scoring module, a space design element scoring module and a recognition result output module;

[0032] The design drawing acquisition module is used to acquire the design drawing to be identified;

[0033] The style recognition and scoring module is used to input the design drawing into a trained style recognition and scoring model so that the style recognition and scoring model outputs the style type of the design drawing and the corresponding style score; the style score is used to quantify the similarity between the real style type of the design drawing and the predicted style type;

[0034] The pre-processing module is used to extract a number of basic components and space design elements according to the design drawing; wherein the types of the basic components include: doors, windows, columns, patios and roofs; the space design elements include: building form, building orientation and landscape;

[0035] The basic component scoring module is used to select a corresponding reference basic component from a preset basic component library according to the style type; for each basic component, calculate the basic component similarity between the basic component and the reference basic component; and calculate the basic component score of the design drawing according to the similarities of all basic components and the preset weight value of each basic component;

[0036] The spatial design element scoring module is used to select corresponding reference spatial design elements from a preset spatial design element library according to the style type; for each spatial design element, calculate the spatial design element similarity between the spatial design element and the reference spatial design element; and calculate the spatial design element score of the design drawing according to the similarities of all spatial design elements and the preset weight value of each spatial design element;

[0037] The recognition result output module is used to calculate the comprehensive score of the design drawing according to the style score, the basic component score, the space design element score and the preset weight values ​​of each score; determine whether the comprehensive score reaches the preset comprehensive score threshold corresponding to the style type, and if so, determine that the design drawing is a hometown of overseas Chinese style design drawing, and determine that the design drawing is a design drawing of the style type; if not, determine that the design drawing is not a hometown of overseas Chinese style design drawing.

[0038] Based on the above embodiments, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the method for identifying the design drawings of the overseas Chinese hometown style described in the above invention embodiments.

[0039] Based on the above embodiments, another embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the method for identifying the design drawings of the overseas Chinese hometown landscape described in the above invention embodiment.

[0040] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0041] The present invention obtains a design drawing to be identified; inputs the design drawing into a trained style recognition and scoring model, so that the style recognition and scoring model outputs the style type and corresponding style score of the design drawing; extracts a number of basic components and space design elements according to the design drawing; selects corresponding reference basic components from a preset basic component library according to the style type; calculates the basic component similarity between the basic component and the reference basic component for each basic component; calculates the basic component score of the design drawing according to the similarities of all basic components and the preset weight value of each basic component; selects corresponding reference space design elements from a preset space design element library according to the style type; For each spatial design element, calculate the spatial design element similarity between the spatial design element and the reference spatial design element; calculate the spatial design element score of the design drawing according to the similarities of all spatial design elements and the preset weight value of each spatial design element; calculate the comprehensive score of the design drawing according to the style score, the basic component score, the spatial design element score and the preset weight values ​​of each score; judge whether the comprehensive score reaches the preset comprehensive score threshold corresponding to the style type, if so, determine that the design drawing is a hometown style design drawing, and determine that the design drawing is a design drawing of the style type, if not, determine that the design drawing is not a hometown style design drawing. The present invention can automatically identify whether the design drawing is a hometown style design drawing and can identify the specific hometown style type, which solves the technical problems of large errors and time-consuming and labor-intensive caused by the existing technology relying on manual identification of design drawings, saves manpower, and improves efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flow chart of a method for identifying a design drawing of an overseas Chinese hometown provided by an embodiment of the present invention;

[0043] Figure 2 It is a structural schematic diagram of a device for identifying a design drawing of an overseas Chinese hometown provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned figure descriptions and any variations thereof are intended to cover non-exclusive inclusions.

[0046] In the description of the embodiments of the present application, “several” means two or more than two, unless otherwise clearly and specifically defined.

[0047] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] Embodiment 1

[0050] Please refer to Figure 1 , which is a flow chart of a method for identifying a design drawing of an overseas Chinese hometown provided by an embodiment of the present invention, comprising:

[0051] S1. Obtain a design drawing to be identified.

[0052] S2. Input the design drawing into a trained style recognition and scoring model so that the style recognition and scoring model outputs the style type of the design drawing and the corresponding style score; the style score is used to quantify the similarity between the actual style type of the design drawing and the predicted style type.

[0053] It should be noted that the method for identifying the design drawings of hometowns with overseas Chinese style provided by the present invention is mainly used to identify the design drawings of hometowns with overseas Chinese style in Guangdong. The hometowns with overseas Chinese style buildings in Guangdong are mainly divided into three categories: Cantonese-style buildings, Chaoshan-style buildings and Hakka-style buildings. Therefore, the style types that may be output by the style recognition and scoring model include: Cantonese-style, Chaoshan-style and Hakka-style.

[0054] In this embodiment, the design drawing is input into the trained style recognition and scoring model so that the style recognition and scoring model can complete two prediction tasks: one is to predict the style type of the design drawing; the other is to score the similarity between the real style type of the design drawing and the predicted style type. The model outputs the predicted style type and style score.

[0055] In a preferred embodiment, the training process of the style recognition and scoring model includes:

[0056] Obtaining a number of sample design drawings and the style type and corresponding style score of each sample design drawing;

[0057] According to the style type and style score of each sample design drawing, set the label of the sample design drawing;

[0058] An initial style recognition and scoring model is constructed, the sample design drawing is used as input, the predicted style type of the sample design drawing and the corresponding style score are used as output, and the style recognition and scoring model is trained to obtain a trained style recognition and scoring model.

[0059] It should be noted that the value range of the style score is [0,1].

[0060] In this embodiment, the sample design drawings include several Cantonese-type architectural sample design drawings, several Chaoshan-type architectural sample design drawings, several Hakka-type architectural sample design drawings, and several other style-type architectural sample design drawings. Among them, the style type of the Cantonese-type architectural sample design drawings is Cantonese-type, and the style score is 1; the style type of the Chaoshan-type architectural sample design drawings is Chaoshan-type, and the style score is 1; the style type of the Hakka-type architectural sample design drawings is Hakka-type, and the style score is 1; the style type of the other style-type architectural sample design drawings is Cantonese-type, Chaoshan-type, or Hakka-type, and the style score is 0.

[0061] In a preferred embodiment, the style recognition and scoring model is a VGG convolutional neural network; all convolutional layers of the VGG convolutional neural network use a small-size convolution kernel of 3×3.

[0062] In this embodiment, the VGG convolutional neural network used in the present invention is composed of multiple convolutional layers and pooling layers stacked alternately, and finally connected to several fully connected layers and output layers. All convolutional layers use a small-size convolution kernel of 3×3, which makes the model easier to understand and implement, and can effectively extract feature information from the image.

[0063] In a preferred embodiment, when training the style recognition and scoring model, the sample design drawing is input into the style recognition and scoring model to obtain a predicted style type and a corresponding style score as a prediction result;

[0064] According to the preset loss function, calculate the loss function value of the prediction result and the corresponding label;

[0065] The style scoring model is optimized according to the loss function value so that the loss function value converges.

[0066] In a preferred embodiment, the loss function is a cross entropy loss function.

[0067] In this embodiment, the expression of the cross entropy loss function is:

[0068]

[0069] In the formula, Loss represents the loss function value; N represents the number of sample design drawings; c represents the number of style types; y ij represents the true label of the i-th sample design drawing. If the sample design drawing belongs to the j-th style type, it is 1, otherwise it is 0; P ij It indicates the probability that the model predicts the i-th sample design image to be of the j-th style type. If the model does not predict the probability of a certain style type, the default value is 0.

[0070] It should be noted that the goal of the cross entropy loss function is to minimize the difference between the model's predicted probability and the true label, that is, to make the model's prediction result as close to the true label as possible. According to the loss function, the optimization algorithm is selected to adjust the model to ensure that the model can converge stably during the training process.

[0071] S3. According to the design drawing, extract a number of basic components and space design elements; wherein the types of the basic components include: doors, windows, columns, patios and roofs; the space design elements include: building form, building orientation and landscape.

[0072] In step S3, basic components are extracted from the design drawing through semantic segmentation technology. The types of basic components include doors, windows, columns, patios and roofs. Spatial design elements are extracted from the design drawing through image recognition technology. Spatial design elements include: architectural form, architectural orientation and landscape. Architectural form refers to the spatial form of a building. For example, the spatial form of Hakka architecture includes forms such as Weilongwu, Siheyuan, Zoumalou and Diaojiaolou. Landscape includes natural landscape and man-made landscape. Natural landscape depends on the specific local environment, while cultural landscape needs to be considered based on local customs, history, religious beliefs and other humanistic factors.

[0073] S4. According to the style type, select a corresponding reference basic component from a preset basic component library; for each basic component, calculate the basic component similarity between the basic component and the reference basic component; and calculate the basic component score of the design drawing according to the similarities of all basic components and the preset weight value of each basic component.

[0074] It should be noted that the basic component library contains reference basic components of various styles, including door reference basic components, window reference basic components, column reference basic components, patio reference basic components and roof reference basic components of various styles. For example, in Hakka architecture, the door reference basic components include arches, rectangular doors, moon doors and double doors, etc., the window reference basic components include wooden windows, stone windows, flower windows, paper windows, lattice windows, sliding windows, screen windows, moon doors and windows, and lattice windows, etc., the column reference basic components include load-bearing columns, decorative columns, corridor columns, arch columns, corner columns and god columns, etc., the patio reference basic components include courtyard patio, wall patio, corner patio, corridor patio, atrium patio, well-shaped patio, skylight patio, corridor patio and open patio, etc., the roof reference basic components include four-slope roof, five-phoenix roof, hard roof, hip roof, hanging roof, round roof, angle roof and surrounding roof, etc. The infrastructure score is used to quantify the similarity between the infrastructure of the design and the infrastructure of the predicted style type.

[0075] In step S4, the basic component similarity between each basic component and the corresponding reference basic component of the same style type is calculated, and then the basic component score is calculated according to the weight value of each basic component.

[0076] In a preferred embodiment, the calculating the basic component similarity between the basic component and the reference basic component comprises:

[0077] Extracting the pattern, decoration type and shape of the basic component;

[0078] calculating a pattern similarity between the pattern and the pattern of the reference base component;

[0079] calculating a decoration type similarity between the decoration type and the decoration type of the reference base component;

[0080] Calculating the shape similarity between the shape and the shape of the reference basic component;

[0081] The basic component similarity is calculated according to the pattern similarity, the decoration type similarity and the shape similarity, a preset pattern similarity weight value, a decoration type similarity weight value and a shape similarity weight value.

[0082] It should be noted that the basic component library also stores details such as the patterns, decoration types and shapes of each reference basic component. Decoration types can be classified according to the material, craftsmanship, decoration themes and application parts. For example, according to the material, the decoration types include: wood, masonry, plaster, ceramics, glass, metal and other types; according to the application parts, the decoration types include: roof decoration, wall decoration, door and window decoration, base decoration and other types. Shape refers to the main outline of the basic component.

[0083] In this embodiment, the pattern, decoration type and shape are extracted from the basic component through image recognition technology, and the similarity is calculated with the pattern, decoration type and shape of the reference basic component to obtain the pattern similarity, decoration type similarity and shape similarity, and then the basic component similarity is calculated according to the preset weight value.

[0084] S5. According to the style type, select corresponding reference spatial design elements from a preset spatial design element library; for each spatial design element, calculate the spatial design element similarity between the spatial design element and the reference spatial design element; calculate the spatial design element score of the design drawing according to the similarities of all spatial design elements and the preset weight value of each of the spatial design elements.

[0085] It should be noted that the spatial design element library contains reference spatial design elements of various styles, including architectural forms, building orientations and landscapes of various styles. The spatial design element score is used to quantify the similarity between the spatial design elements of the design drawing and the spatial design elements of the predicted style type.

[0086] In step S5, the spatial design element similarity between each spatial design element and the corresponding spatial design element of the same style type is calculated, and then the spatial design element score is calculated according to each spatial design element similarity.

[0087] In a preferred embodiment, the calculating the spatial design element similarity between the spatial design element and the reference spatial design element comprises:

[0088] Calculating the architectural form similarity between the architectural form of the spatial design element and the architectural form of the reference spatial design element;

[0089] Calculating the architectural orientation similarity between the architectural orientation of the spatial design element and the architectural orientation of the reference spatial design element;

[0090] Calculating the landscape similarity between the landscape of the spatial design element and the landscape of the reference spatial design element;

[0091] The spatial design element similarity is calculated according to the building form similarity, the building orientation similarity, the landscape similarity, a preset building form similarity weight value, a preset building orientation similarity weight value and a preset landscape similarity weight value.

[0092] In this embodiment, the architectural form similarity, architectural orientation similarity and landscape similarity between the spatial design elements of the design drawing and the reference spatial design elements are calculated respectively, and then the spatial design element similarity is calculated according to a preset weight value.

[0093] S6. Calculate the comprehensive score of the design drawing according to the style score, the basic component score, the space design element score and the preset weight values ​​of each score; determine whether the comprehensive score reaches the preset comprehensive score threshold corresponding to the style type. If so, determine that the design drawing is a hometown of overseas Chinese style design drawing, and determine that the design drawing is a design drawing of the style type; if not, determine that the design drawing is not a hometown of overseas Chinese style design drawing.

[0094] In step S6, the comprehensive score of the design drawing is calculated according to the style score, the basic component score, the space design element score, the weight value of the style score, the weight value of the basic component score and the weight value of the space design element score. According to the predicted style type, the comprehensive score threshold corresponding to the style type is determined, and it is judged whether the calculated comprehensive score reaches the comprehensive score threshold. If so, it is determined that the design drawing is a design drawing of the style of the hometown of overseas Chinese, specifically a design drawing of the predicted style type. If not, it is determined that the design drawing is not a design drawing of the style of the hometown of overseas Chinese.

[0095] Embodiment 2

[0096] Please refer to Figure 2 , is a schematic diagram of the structure of a device for identifying a design drawing of an overseas Chinese hometown style provided by an embodiment of the present invention, comprising: a design drawing acquisition module, a style recognition and scoring module, a preprocessing module, a basic component scoring module, a space design element scoring module and a recognition result output module;

[0097] The design drawing acquisition module is used to acquire the design drawing to be identified;

[0098] The style recognition and scoring module is used to input the design drawing into a trained style recognition and scoring model so that the style recognition and scoring model outputs the style type of the design drawing and the corresponding style score; the style score is used to quantify the similarity between the real style type of the design drawing and the predicted style type;

[0099] The pre-processing module is used to extract a number of basic components and space design elements according to the design drawing; wherein the types of the basic components include: doors, windows, columns, patios and roofs; the space design elements include: building form, building orientation and landscape;

[0100] The basic component scoring module is used to select a corresponding reference basic component from a preset basic component library according to the style type; for each basic component, calculate the basic component similarity between the basic component and the reference basic component; and calculate the basic component score of the design drawing according to the similarities of all basic components and the preset weight value of each basic component;

[0101] The spatial design element scoring module is used to select corresponding reference spatial design elements from a preset spatial design element library according to the style type; for each spatial design element, calculate the spatial design element similarity between the spatial design element and the reference spatial design element; and calculate the spatial design element score of the design drawing according to the similarities of all spatial design elements and the preset weight value of each spatial design element;

[0102] The recognition result output module is used to calculate the comprehensive score of the design drawing according to the style score, the basic component score, the space design element score and the preset weight values ​​of each score; determine whether the comprehensive score reaches the preset comprehensive score threshold corresponding to the style type, and if so, determine that the design drawing is a hometown of overseas Chinese style design drawing, and determine that the design drawing is a design drawing of the style type; if not, determine that the design drawing is not a hometown of overseas Chinese style design drawing.

[0103] Embodiment 3

[0104] Accordingly, an embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the method for identifying the design drawings of the overseas Chinese hometown style described in the above-mentioned embodiment of the invention.

[0105] Embodiment 4

[0106] Accordingly, an embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the method for identifying the design drawings of the overseas Chinese hometown landscape described in the above-mentioned embodiment of the invention.

[0107] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0108] Those skilled in the art can clearly understand that for the sake of convenience and brevity, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0109] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0110] The processor 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. The processor is the control center of the device, and various interfaces and lines are used to connect various parts of the entire device.

[0111] The memory can be used to store the computer program, and the processor realizes various functions of the device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0112] The storage medium is a storage medium, and the computer program is stored in the storage medium. When the computer program is executed by the processor, the steps of each method embodiment described above can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer readable media do not include electric carrier signals and telecommunication signals.

[0113] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for identifying a design drawing of an overseas Chinese hometown, characterized in that: include: Obtaining a design drawing to be identified; Inputting the design drawing into a trained style recognition and scoring model so that the style recognition and scoring model outputs the style type of the design drawing and a corresponding style score; the style score is used to quantify the similarity between the real style type of the design drawing and the predicted style type; According to the design drawing, a number of basic components and space design elements are extracted; wherein the types of the basic components include: doors, windows, columns, patios and roofs; the space design elements include: building form, building orientation and landscape; According to the style type, a corresponding reference basic component is selected from a preset basic component library; for each basic component, a basic component similarity between the basic component and the reference basic component is calculated; according to the similarities of all basic components and a preset weight value of each basic component, a basic component score of the design drawing is calculated; According to the style type, a corresponding reference space design element is selected from a preset space design element library; for each space design element, a space design element similarity between the space design element and the reference space design element is calculated; according to the similarities of all space design elements and a preset weight value of each space design element, a space design element score of the design drawing is calculated; The comprehensive score of the design drawing is calculated based on the style score, the basic component score, the space design element score and the preset weight values ​​of each score; it is determined whether the comprehensive score reaches the preset comprehensive score threshold corresponding to the style type; if so, it is determined that the design drawing is a hometown of overseas Chinese style design drawing, and the design drawing is determined to be a design drawing of the style type; if not, it is determined that the design drawing is not a hometown of overseas Chinese style design drawing.

2. The method for recognizing the design drawing of the hometown of overseas Chinese as claimed in claim 1, characterized in that: The training process of the style recognition and scoring model includes: Obtaining a number of sample design drawings and the style type and corresponding style score of each sample design drawing; According to the style type and style score of each sample design drawing, set the label of the sample design drawing; An initial style recognition and scoring model is constructed, with the sample design drawing as input, and the predicted style type and corresponding style score of the sample design drawing as output, and the style recognition and scoring model is trained to obtain a trained style recognition and scoring model.

3. The method for recognizing the design drawing of the hometown of overseas Chinese as claimed in claim 2, characterized in that: The style recognition and scoring model is a VGG convolutional neural network; all convolutional layers of the VGG convolutional neural network use a small-size convolution kernel of 3×3.

4. The method for recognizing the design drawing of the hometown of overseas Chinese as claimed in claim 3, characterized in that: When training the style recognition and scoring model, the sample design drawing is input into the style recognition and scoring model to obtain a predicted style type and a corresponding style score as a prediction result; According to the preset loss function, calculate the loss function value of the prediction result and the corresponding label; The style scoring model is optimized according to the loss function value so that the loss function value converges.

5. The method for recognizing the design drawing of the hometown of overseas Chinese as claimed in claim 4, characterized in that: The loss function is a cross entropy loss function.

6. The method for recognizing the design drawing of the hometown of overseas Chinese as claimed in claim 1, characterized in that: The calculating the basic component similarity between the basic component and the reference basic component comprises: Extracting the pattern, decoration type and shape of the basic component; calculating a pattern similarity between the pattern and the pattern of the reference base component; calculating a decoration type similarity between the decoration type and the decoration type of the reference base component; Calculating the shape similarity between the shape and the shape of the reference basic component; The basic component similarity is calculated according to the pattern similarity, the decoration type similarity and the shape similarity, a preset pattern similarity weight value, a decoration type similarity weight value and a shape similarity weight value.

7. The method for recognizing the design drawing of the hometown of overseas Chinese as claimed in claim 1, characterized in that: The calculating the spatial design element similarity between the spatial design element and the reference spatial design element comprises: Calculating the architectural form similarity between the architectural form of the spatial design element and the architectural form of the reference spatial design element; Calculating the architectural orientation similarity between the architectural orientation of the spatial design element and the architectural orientation of the reference spatial design element; Calculating the landscape similarity between the landscape of the spatial design element and the landscape of the reference spatial design element; The spatial design element similarity is calculated according to the building form similarity, the building orientation similarity, the landscape similarity, a preset building form similarity weight value, a preset building orientation similarity weight value and a preset landscape similarity weight value.

8. A device for recognizing the design drawings of hometowns of overseas Chinese, characterized in that: include: Design drawing acquisition module, style recognition and scoring module, preprocessing module, basic component scoring module, space design element scoring module and recognition result output module; The design drawing acquisition module is used to acquire the design drawing to be identified; The style recognition and scoring module is used to input the design drawing into a trained style recognition and scoring model so that the style recognition and scoring model outputs the style type of the design drawing and the corresponding style score; the style score is used to quantify the similarity between the real style type of the design drawing and the predicted style type; The pre-processing module is used to extract a number of basic components and space design elements according to the design drawing; wherein the types of the basic components include: doors, windows, columns, patios and roofs; the space design elements include: building form, building orientation and landscape; The basic component scoring module is used to select a corresponding reference basic component from a preset basic component library according to the style type; for each basic component, calculate the basic component similarity between the basic component and the reference basic component; and calculate the basic component score of the design drawing according to the similarities of all basic components and the preset weight value of each basic component; The spatial design element scoring module is used to select corresponding reference spatial design elements from a preset spatial design element library according to the style type; for each spatial design element, calculate the spatial design element similarity between the spatial design element and the reference spatial design element; and calculate the spatial design element score of the design drawing according to the similarities of all spatial design elements and the preset weight value of each spatial design element; The recognition result output module is used to calculate the comprehensive score of the design drawing according to the style score, the basic component score, the space design element score and the preset weight values ​​of each score; determine whether the comprehensive score reaches the preset comprehensive score threshold corresponding to the style type, and if so, determine that the design drawing is a hometown of overseas Chinese style design drawing, and determine that the design drawing is a design drawing of the style type; if not, determine that the design drawing is not a hometown of overseas Chinese style design drawing.

9. A terminal device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the method for identifying the design drawings of the overseas Chinese hometown landscape as described in any one of claims 1 to 7 when executing the computer program.

10. A storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the method for identifying the design drawings of the hometown of overseas Chinese as described in any one of claims 1 to 7.

Images