CN-115730364-B - Building model design identification method, device and equipment based on artificial intelligence

CN115730364BCN 115730364 BCN115730364 BCN 115730364BCN-115730364-B

Abstract

The application relates to a building model design identification method, a device, computer equipment and a storage medium based on artificial intelligence. The method comprises the steps of obtaining a building model to be identified, wherein the building model to be identified is a three-dimensional model drawn based on design software, inputting the building model to be identified into a preset identification model, extracting building model component attributes in the building model to be identified and object relations among the building model component attributes, and matching the building model component attributes with building design rules in a building design rule base to obtain an identification result of the building model design to be identified. By adopting the method, the identification accuracy of the building model can be improved.

Inventors

HUA RONGWEI
Request for anonymity

Assignees

久瓴(江苏)数字智能科技有限公司

Dates

Publication Date: 20260505
Application Date: 20210826

Claims (10)

1. An artificial intelligence-based building model design recognition method, which is characterized by comprising the following steps: the method comprises the steps of obtaining a building model to be identified, wherein the building model to be identified is a three-dimensional model drawn based on design software; Inputting the building model to be identified into a preset identification model, and extracting building model component attributes in the building model to be identified and object relations among the building model component attributes; Judging whether the matching attribute data of the building model to be identified meets each building design rule in a building design rule library or not to obtain an identification result, and prompting according to the identification result; the recognition result is used for representing the matching degree of the design rule of the building model to be recognized and the building design rule, the preset recognition model is a neural network comprising the building design rule base, and the building design rule base comprises building design experience rules obtained through learning from a plurality of building sample models and building design standard rules obtained based on industry standards; the process for acquiring the building design experience rule comprises the following steps: Obtaining a plurality of building sample models; Analyzing the building sample model by utilizing a semantic recognition model to obtain characteristic attributes, geometric information and hierarchical relations of different objects in the building sample model; Learning and classifying characteristic attributes, geometric information and hierarchical relations of different objects in the building sample model to obtain the building design experience rule; The standard rule acquisition method comprises the following steps: acquiring an electronic text comprising industry standards; Carrying out semantic recognition on the electronic text to obtain characteristic attributes of different objects in the electronic text and hierarchical relations of the different objects; and obtaining the standard rule according to the characteristic attribute of different objects in the electronic text and the hierarchical relation of the different objects.
2. The method of claim 1, wherein the rule categories in the building design rule base include collision rules, specification rules, connectivity rules, and object relationship rules.
3. The method according to claim 2, wherein the collision rule is used for normalizing whether there is a collision between each building model component in the building model, the specification rule is used for normalizing whether the dimensions of each building model component in the building model meet preset requirements, the communication rule is used for normalizing whether the communication relationship between each building model component meets preset requirements, and the object relationship rule is used for normalizing whether the relationship between the building model component and the building model component attribute meets preset requirements.
4. The method as recited in claim 1, wherein the method further comprises: and updating the building design rule base according to the object obtained by analyzing the building model to be identified, the object attribute, the geometric information or the hierarchical relation.
5. An artificial intelligence-based building model design recognition method, which is characterized by comprising the following steps: acquiring an electronic text comprising industry standards; Carrying out semantic recognition on the electronic text to obtain characteristic attributes of different objects in the electronic text and hierarchical relations of the different objects; Obtaining standard rules according to the characteristic attributes of different objects in the electronic text and the hierarchical relationship of the different objects; acquiring a plurality of building sample models; Analyzing the building sample model by utilizing a semantic recognition model to obtain characteristic attributes, geometric information and hierarchical relations of different objects in the building sample model; Learning and classifying characteristic attributes, geometric information and hierarchical relations of different objects in the building sample model to obtain building design experience rules; the method comprises the steps of obtaining a building model to be identified, wherein the building model to be identified is a three-dimensional model drawn based on design software; Inputting the building model to be identified into a preset identification model; The method comprises the steps of combing building model component attributes and object relations among the building model component attributes to form matching attribute data, judging whether the matching attribute data of a building model to be identified meets each building design rule in a building design rule base or not to obtain an identification result, wherein the preset identification model is a neural network comprising the building design rule base, and the building design rule base comprises building design experience rules learned from a plurality of building sample models and building design standard rules acquired based on industry standards; And prompting according to the identification result.
6. An artificial intelligence based building model design recognition device, the device comprising: The system comprises an acquisition module, a recognition module and a control module, wherein the acquisition module is used for acquiring a building model to be recognized, and the building model to be recognized is a three-dimensional model drawn based on design software; the identification module is used for inputting the building model to be identified into a preset identification model and extracting building model component attributes in the building model to be identified and object relations among the building model component attributes; Judging whether the matching attribute data of the building model to be identified meets each building design rule in a building design rule library or not to obtain an identification result, and prompting according to the identification result; The recognition result is used for representing whether the design rule of the building model to be recognized accords with a design specification, the preset recognition model is a neural network comprising the building design rule base, and the building design rule base comprises building design experience rules obtained through learning from a plurality of building sample models and standard rules of building design obtained based on industry standards; The identification module is further used for acquiring a plurality of building sample models; Analyzing the building sample model by utilizing a semantic recognition model to obtain characteristic attributes, geometric information and hierarchical relations of different objects in the building sample model; Learning and classifying characteristic attributes, geometric information and hierarchical relations of different objects in the building sample model to obtain the building design experience rule; the identification module is further configured to: acquiring an electronic text comprising industry standards; Carrying out semantic recognition on the electronic text to obtain characteristic attributes of different objects in the electronic text and hierarchical relations of the different objects; and obtaining the standard rule according to the characteristic attribute of different objects in the electronic text and the hierarchical relation of the different objects.
7. The apparatus of claim 6, wherein the rule categories in the building design rule base include collision rules, specification rules, connectivity rules, and object relationship rules.
8. An artificial intelligence based building model design recognition device, the device comprising: The system comprises a rule acquisition module, a plurality of building sample models, a rule classification module and a building design rule acquisition module, wherein the rule acquisition module is used for acquiring an electronic text comprising industry standards, carrying out semantic recognition on the electronic text to obtain characteristic attributes of different objects in the electronic text and hierarchical relations of different objects, obtaining standard rules according to the characteristic attributes of different objects in the electronic text and the hierarchical relations of different objects; The processing module is used for acquiring a building model to be identified, inputting the building model to be identified into a preset identification model, and judging whether matching attribute data of the building model to be identified meet each building design rule in a building design rule base to obtain an identification result, wherein the building model to be identified is a three-dimensional model drawn based on design software, the preset identification model is a neural network comprising the building design rule base, and the building design rule base comprises building design experience rules obtained through learning from a plurality of building sample models and standard rules of building design obtained based on industry standards; and the prompt module is used for prompting according to the identification result.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 5 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 5.

Description

Building model design identification method, device and equipment based on artificial intelligence Technical Field The application relates to the technical field of building design, in particular to a building model design identification method, device, equipment and storage medium based on artificial intelligence. Background With the development of scientific technology, the field of architectural design is more automated. Conventional building models utilize tools that are self-contained in the design software to inspect the design model after the design is complete. However, existing tools can only perform simple collision-like rule checks, and other more complex and deep indexes need to be performed manually. However, the designed building model is inspected manually, the experience and the careful degree of personnel are needed, and when the large model is faced by the manual inspection mode, the inspection efficiency and the accuracy are greatly reduced, and due to factors such as the experience of the designer, the missing inspection is easy to exist, so that the error rate of the design model is higher. Disclosure of Invention In view of the foregoing, it is desirable to provide an artificial intelligence-based building model design recognition method, apparatus, computer device, and storage medium that can improve model accuracy. In a first aspect, an embodiment of the present application provides a building model design recognition method based on artificial intelligence, where the method includes: the method comprises the steps of obtaining a building model to be identified, wherein the building model to be identified is a three-dimensional model drawn based on design software; Inputting the building model to be identified into a preset identification model, extracting building model component attributes in the building model to be identified and object relations among the building model component attributes, and matching the building model component attributes with building design rules in a building design rule library to obtain an identification result of the building model design to be identified; The recognition result is used for representing the matching degree of the design rules of the building model to be recognized and the building design rules, the recognition model is a neural network comprising a building design rule base, and the building design rule base comprises building design experience rules obtained through learning from a plurality of building sample models and building design standard rules obtained based on industry standards. In one embodiment, the matching with the building design rules in the building design rule library to obtain the recognition result of the building model design to be recognized includes: Carding object relations among building model component attributes to form matching attribute data; And judging whether the matching attribute data of the building model to be identified meets each building design rule in the building design rule library or not, and obtaining the identification result. In one embodiment, the process for obtaining the building design experience rule includes: Obtaining a plurality of building sample models; Analyzing the building sample model by utilizing a semantic recognition model to obtain characteristic attributes, geometric information and hierarchical relations of different objects in the building sample model; And learning and classifying the characteristic attributes, the geometric information and the hierarchical relation of different objects in the building sample model to obtain the building design experience rule. In one embodiment, the method for obtaining the standard rule includes: acquiring an electronic text comprising industry standards; Carrying out semantic recognition on the electronic text to obtain characteristic attributes of different objects in the electronic text and hierarchical relations of the different objects; and obtaining the standard rule according to the characteristic attribute of different objects in the electronic text and the hierarchical relation of the different objects. In one embodiment, the rule categories in the building design rule base include collision rules, specification rules, connectivity rules and object relationship rules. In a second aspect, an embodiment of the present application provides a building model design recognition method based on artificial intelligence, the method including: acquiring an electronic text comprising industry standards; Carrying out semantic recognition on the electronic text to obtain characteristic attributes of different objects in the electronic text and hierarchical relations of the different objects; Obtaining the standard rule according to the characteristic attribute of different objects in the electronic text and the hierarchical relation of the different objects; Obtaining a plurality of building sample models; Analyzing the building sample