Drawing architectural element identification method and system based on machine learning

Abstract

The invention discloses a drawing architectural element identification method and system based on machine learning. The drawing architectural element identification method is based on the method of combining a rule and machine learning. Rule matching and mathematical modeling are conducted on vector diagram metainformation obtained from a drawing, the feature distribution law of an architectural element in the drawing is strictly and accurately analyzed through the training process, the data sets of a primitive database, an architectural element rule base and an architectural element feature base are formed, the architectural element identification method in the probability sense is obtained, and the recognition rate of an architectural element and a property parameter of the architectural element are further improved. The drawing architectural element identification system based on machine learning tries to break through the limitation of an existing drawing identification method, provides an efficient drawing analysis and identification tool for a metro design and construction department, and has the three advantages of being high in precision ratio, identification efficiency and intelligent degree.

Description

Technical field [0001] The invention belongs to the technical field of machine learning, and specifically relates to a method and system for recognizing drawing architectural elements based on machine learning. The method and system are based on traditional machine learning technology, and propose a recognition method that is different from the recognition of architectural elements based on image processing, which can effectively analyze and extract architectural drawings and judge the category of unknown architectural elements to be tested. Background technique [0002] With the continuous improvement of the national economic development level, in order to improve people's work efficiency and quality of life, more and more cities have already or are preparing to construct subway projects. Usually, the design unit of subway engineering projects uses mature CAD software to design corresponding engineering drawings according to the actual conditions of local geology, hydrology, etc...

AI Technical Summary

Problems solved by technology

However, a lot of practice shows that this kind of method can't satisfy users when analyzing more complex design drawings.

For example, when a design drawing contains multiple architectural elements of the same type with different attribute parameters, some architectural elements that do not have general attribute parameters are often ignored, which confuses the readers of the drawings; in addition, pure The recognition algorithm based on rules and written for specific drawings has poor versatility due to the diversity of drawing content, or has a single function and can only recognize individual architectural elements

These problems will affect the efficiency of users' reading and analysis of drawings, and affect the overall progress of construction.

Images