CN115797962A - Wall column identification method and device based on assembly type building AI design - Google Patents
Wall column identification method and device based on assembly type building AI design Download PDFInfo
- Publication number
- CN115797962A CN115797962A CN202310056470.9A CN202310056470A CN115797962A CN 115797962 A CN115797962 A CN 115797962A CN 202310056470 A CN202310056470 A CN 202310056470A CN 115797962 A CN115797962 A CN 115797962A
- Authority
- CN
- China
- Prior art keywords
- wall
- target
- wall column
- column
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to the field of artificial intelligence, and discloses a wall column identification method and device based on assembly type building AI design, which are used for improving the accuracy of intelligent wall column identification. The method comprises the following steps: obtaining a sample drawing based on a preset assembly type collaborative design platform, and carrying out line extraction on the sample drawing to obtain sample line data; carrying out wall column marking information identification on the sample line data to obtain a training image with a label; inputting the training image into a preset neural network model for model training to obtain a wall column recognition model; acquiring a target original drawing to be recognized, and inputting the target original drawing into the wall column recognition model for intelligent wall column recognition to obtain target line segment data; and converting the wall column drawing of the target line segment data to obtain a target wall column drawing.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a wall column identification method and device based on an assembly type building AI design.
Background
In the design process of the assembly type building industry, generally, an assembly type designer carries out related design work according to customer requirements, then designed drawings are exported, the drawings are usually CAD graph files, subsequent building designers carry out related design on the basis of the drawings, the functional requirements lead a collaborative design platform to have the editing capacity of CAD graphs in the design process, on the other hand, in the processes of designing, rendering, drawing and the like, the designer needs to carry out complex processing, the images need to carry out a series of image processing operations such as zooming, transparency, gradual change, rotation, shearing, local patching and the like according to the characteristics of the assembly type building design, and related operations and the CAD graphs are integrated.
When the existing scheme is used for recognizing the drawing, if the wall column is on the same layer, the extracted polygon can be a fusion shape of the wall outline and the column outline, and the wall column cannot be separated and cut out simply and directly, so that the restoration degree of the drawing is low, the actual scene of a business is not met, and the accuracy of the existing scheme is low.
Disclosure of Invention
The invention provides a wall column identification method and device based on assembly type building AI design, which are used for improving the accuracy of intelligent wall column identification.
The invention provides a wall column identification method based on an assembly type building AI design, which comprises the following steps: acquiring a sample drawing based on a preset assembly type collaborative design platform, and carrying out line extraction on the sample drawing to obtain sample line data; carrying out wall column marking information identification on the sample line data to obtain a training image with a label; inputting the training image into a preset neural network model for model training to obtain a wall column recognition model; acquiring a target original drawing to be recognized, and inputting the target original drawing into the wall column recognition model for intelligent wall column recognition to obtain target line segment data; and converting the wall column drawing of the target line segment data to obtain a target wall column drawing.
Optionally, in a first implementation manner of the first aspect of the present invention, the performing wall column labeling information recognition on the sample line data to obtain a training image with a label includes: calling a preset depth analysis detection model to perform line feature recognition on the sample line data to obtain feature identification information; carrying out feature classification on the feature identification information to obtain a target feature class; and performing wall column labeling information image conversion on the sample line data according to the target feature class and the feature identification information to obtain a training image with a label.
Optionally, in a second implementation manner of the first aspect of the present invention, the inputting the training image into a preset neural network model for model training to obtain a wall column recognition model includes: inputting the training image into a preset neural network model, and performing image feature extraction on the training image through the neural network model to obtain a training feature image; carrying out wall column recognition classification and feature point selection on the training feature images to obtain a target training result; and performing parameter tuning on the neural network model according to the target training result, and taking the neural network model after parameter tuning as the wall column recognition model.
Optionally, in a third implementation manner of the first aspect of the present invention, the obtaining a target original drawing to be recognized, and inputting the target original drawing into the wall stud recognition model for intelligent wall stud recognition to obtain target line segment data includes: searching a target original drawing to be identified from the assembly type collaborative design platform; inputting the target original drawing into the wall column identification model for intelligent wall column identification to obtain initial line segment data; and predicting and marking the position data of the initial line segment data to obtain target line segment data.
Optionally, in a fourth implementation manner of the first aspect of the present invention, the converting the wall column drawing of the target line segment data to obtain a target wall column drawing includes: parallel line selection is carried out on the target line segment data according to a preset parallel line selection rule to obtain a parallel pairing line; carrying out wall edge and column edge classification extraction on the parallel alignment line to obtain a wall edge array and a column edge array; and carrying out wall column cutting on the wall side array and the column side array to generate a target wall column drawing.
Optionally, in a fifth implementation manner of the first aspect of the present invention, the performing parallel line selection on the target line segment data according to a preset parallel line selection rule to obtain a parallel paired line includes: generating the maximum thickness of the wall according to the target line segment data; performing parallel line extraction on the target line segment data according to the maximum wall thickness to obtain initial parallel lines; performing distance distribution operation on the initial parallel lines to obtain distance distribution data; and screening the initial parallel lines according to the distance distribution data to obtain parallel paired lines.
Optionally, in a sixth implementation manner of the first aspect of the present invention, the performing wall column cutting on the wall side array and the column side array to generate a target wall column drawing includes: carrying out wall column distinguishing on the wall side array and the column side array to obtain a plurality of outer contours; respectively carrying out attribute analysis on each outer contour to obtain attribute data of each outer contour; and generating a target wall column drawing according to the attribute data of each outer contour.
The invention provides a wall column recognition device based on an assembly type building AI design, which comprises the following components: the acquisition module is used for acquiring a sample drawing based on a preset assembly type collaborative design platform and carrying out line extraction on the sample drawing to obtain sample line data; the identification module is used for carrying out wall column marking information identification on the sample line data to obtain a training image with a label; the training module is used for inputting the training image into a preset neural network model for model training to obtain a wall column recognition model; the processing module is used for acquiring a target original drawing to be recognized, inputting the target original drawing into the wall column recognition model for intelligent wall column recognition, and obtaining target line segment data; and the output module is used for carrying out wall column drawing conversion on the target line segment data to obtain a target wall column drawing.
Optionally, in a first implementation manner of the second aspect of the present invention, the identification module is specifically configured to: calling a preset depth analysis detection model to perform line feature recognition on the sample line data to obtain feature identification information; carrying out feature classification on the feature identification information to obtain a target feature class; and performing wall column labeling information image conversion on the sample line data according to the target feature class and the feature identification information to obtain a training image with a label.
Optionally, in a second implementation manner of the second aspect of the present invention, the training module is specifically configured to: inputting the training image into a preset neural network model, and performing image feature extraction on the training image through the neural network model to obtain a training feature image; carrying out wall column recognition classification and feature point selection on the training feature images to obtain a target training result; and performing parameter tuning on the neural network model according to the target training result, and taking the neural network model after parameter tuning as the wall column recognition model.
Optionally, in a third implementation manner of the second aspect of the present invention, the processing module is specifically configured to: searching a target original drawing to be identified from the assembly type collaborative design platform; inputting the target original drawing into the wall column identification model for intelligent wall column identification to obtain initial line segment data; and predicting and marking the position data of the initial line segment data to obtain target line segment data.
Optionally, in a fourth implementation manner of the second aspect of the present invention, the output module further includes: the selecting unit is used for performing parallel line selection on the target line segment data according to a preset parallel line selection rule to obtain a parallel pairing line; the extraction unit is used for carrying out wall edge and column edge classification extraction on the parallel alignment line to obtain a wall edge array and a column edge array; and the cutting unit is used for carrying out wall column cutting on the wall side array and the column side array to generate a target wall column drawing.
Optionally, in a fifth implementation manner of the second aspect of the present invention, the selecting unit is specifically configured to: generating the maximum thickness of the wall according to the target line segment data; performing parallel line extraction on the target line segment data according to the maximum wall thickness to obtain initial parallel lines; performing distance distribution operation on the initial parallel lines to obtain distance distribution data; and screening the initial parallel lines according to the distance distribution data to obtain parallel paired lines.
Optionally, in a sixth implementation manner of the second aspect of the present invention, the cutting unit is specifically configured to: carrying out wall column distinguishing on the wall side array and the column side array to obtain a plurality of outer contours; respectively carrying out attribute analysis on each outer contour to obtain attribute data of each outer contour; and generating a target wall column drawing according to the attribute data of each outer contour.
A third aspect of the present invention provides a wall stud recognition apparatus designed based on an assembly type building AI, including: a memory and at least one processor, the memory having instructions stored therein; the at least one processor invokes the instructions in the memory to cause the assembled building AI design-based wall stud recognition device to perform the assembled building AI design-based wall stud recognition method described above.
A fourth aspect of the present invention provides a computer-readable storage medium having stored therein instructions, which, when run on a computer, cause the computer to execute the above-described wall stud recognition method based on an assembly building AI design.
In the technical scheme provided by the invention, a sample drawing is obtained based on an assembled collaborative design platform, and line extraction is carried out on the sample drawing to obtain sample line data; carrying out wall column marking information identification on the sample line data to obtain a training image with a label; inputting the training image into a preset neural network model for model training to obtain a wall column recognition model; acquiring a target original drawing to be recognized, and inputting the target original drawing into the wall column recognition model for intelligent wall column recognition to obtain target line segment data; and performing wall column drawing conversion on the target line segment data to obtain a target wall column drawing. According to the method, when the target original drawing is recognized, the wall and the column can be very quickly and accurately distinguished in the wall column recognition model, and the drawing outline is completely restored.
Drawings
Fig. 1 is a schematic view of an embodiment of a wall stud identification method based on an assembly building AI design according to an embodiment of the present invention;
fig. 2 is a schematic view of another embodiment of a wall stud identification method based on an assembly building AI design according to an embodiment of the present invention;
fig. 3 is a schematic view of an embodiment of a wall stud recognition device designed based on an assembly building AI according to an embodiment of the present invention;
fig. 4 is a schematic view of another embodiment of a wall stud recognition device designed based on an assembly type building AI according to an embodiment of the present invention;
fig. 5 is a schematic view of an embodiment of a wall stud recognition device designed based on an assembly type building AI according to the embodiment of the present invention;
FIG. 6 is a schematic view of a wall stud structure according to an embodiment of the present invention;
fig. 7 is a drawing of a target wall stud in an embodiment of the invention.
Detailed Description
The embodiment of the invention provides a wall column identification method and device based on an assembly type building AI design, which are used for improving the accuracy of intelligent wall column identification. The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein. Moreover, the terms "comprises," "comprising," or "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
For the sake of understanding, a detailed flow of an embodiment of the present invention is described below, and referring to fig. 1, an embodiment of a wall stud identification method based on an assembly building AI design in an embodiment of the present invention includes:
101. obtaining a sample drawing based on a preset assembly type collaborative design platform, and carrying out line extraction on the sample drawing to obtain sample line data;
it is to be understood that the execution subject of the present invention may be a wall stud recognition device designed based on the assembly building AI, and may also be a terminal or a server, which is not limited herein. The embodiment of the present invention is described by taking a server as an execution subject.
Specifically, firstly, a drawing edited by the assistant platform is obtained, then a plurality of lines extracted from wall column integrated data in the drawing are obtained, wherein the server takes one end of each line as a head point and the other end of each line as a tail point, the head point and the tail point are respectively endowed with a motion range threshold, and when the server extracts the lines of the sample drawing, the positions of the head point and the tail point are scanned to generate sample line data.
102. Carrying out wall column marking information identification on the sample line data to obtain a training image with a label;
specifically, the server analyzes drawing contents by using a Mask-RCNN neural network, further performs wall column labeling information identification on sample line data to obtain a labeled training image, wherein the server intelligently identifies and extracts the sample line data into a labeling information distribution table, and after the sample line data is loaded, the server identifies the layer information of the sample line data primitives, combines the dimension data labeled by the sample line data and the layer height data information on the sample line data to obtain the labeled training image, so that the wall column labeling information in the sample line data can be accurately and rapidly identified, and the data processing efficiency is improved.
103. Inputting the training image into a preset neural network model for model training to obtain a wall column recognition model;
specifically, the server trains the neural network model by using the training images, so that the model can fully learn feature points of wall stud recognition classification in drawings of each model, wherein the server acquires the training images as training data of the neural network model, the training images comprise clean segment data without noise and noise segment data containing noise, labeling information parameters of the training data are calculated, noise feature vectors are constructed according to the labeling information parameters, and the noise feature vectors are input into the neural network model for training to obtain the wall stud recognition model.
104. Acquiring a target original drawing to be recognized, inputting the target original drawing into a wall column recognition model for intelligent wall column recognition, and obtaining target line segment data;
specifically, the server acquires a plurality of structural drawing frames, fig. 6 is a schematic diagram of a wall column structure, the structural drawing frames are analyzed to acquire a plurality of primitives, all the acquired primitives are input into corresponding target original drawings, and intelligent wall column identification is performed on the drawing frames on an initial template image to acquire target line segment data, wherein when the target original drawings are input into a wall column identification model for identification, the server firstly determines the maximum thickness of a wall body according to the data, counts the distance distribution of all parallel alignment lines to acquire the target line segment data, wherein the parallel alignment lines meet the requirements in opposite directions, the ratio of the length to the distance is greater than four, and then the parallel alignment lines meeting the requirements are acquired to acquire the target line segment data.
105. And converting the wall column drawing of the target line segment data to obtain a target wall column drawing.
Specifically, the server performs position analysis on the target line segment data, determines position information corresponding to the target line segment data, further generates a wall column primitive according to the position information to obtain a corresponding wall column primitive, and finally performs wall column drawing conversion according to the wall column primitive to obtain a target wall column drawing.
In the embodiment of the invention, a sample drawing is obtained based on an assembled collaborative design platform, and line extraction is carried out on the sample drawing to obtain sample line data; carrying out wall column marking information identification on the sample line data to obtain a training image with a label; inputting the training image into a preset neural network model for model training to obtain a wall column recognition model; acquiring a target original drawing to be recognized, inputting the target original drawing into a wall column recognition model for intelligent wall column recognition, and obtaining target line segment data; and carrying out wall column drawing conversion on the target line segment data to obtain a target wall column drawing. According to the method, when the target original drawing is recognized, the wall and the column can be very quickly and accurately distinguished in the wall column recognition model, and the drawing outline is completely restored.
Referring to fig. 2, another embodiment of a wall stud recognition method based on an assembly type building AI design according to an embodiment of the present invention includes:
201. acquiring a sample drawing based on a preset assembly type collaborative design platform, and carrying out line extraction on the sample drawing to obtain sample line data;
202. carrying out wall column marking information identification on the sample line data to obtain a training image with a label;
specifically, a preset depth analysis detection model is called to perform line feature recognition on the sample line data to obtain feature identification information; carrying out feature classification on the feature identification information to obtain a target feature class; and performing wall column labeling information image conversion on the sample line data according to the target feature class and the feature identification information to obtain a training image with a label.
The method comprises the steps that a server calls a preset depth analysis detection model to conduct line tracing on sample line data and serially connects the sample line data, the server compares the serially connected lines with a feature image and a feature graph to obtain feature identification information, the server divides the feature identification information into a plurality of characters which are easy to extract features, the features reflecting the feature identification information are extracted from the time domain and the frequency domain, meanwhile, redundant features in the feature identification information are removed, different features are fused into a fusion feature according to a certain proportion, the server classifies the feature identification information by using an extreme learning machine to obtain a target feature class, the server further derives the feature identification information into training data in a preset format, the server sets the feature identification information to be forward propagation, the model type to be Mask-RCNN, the batch size to be 500, the optimal learning rate is calculated through a learning rate formula, wall column labeling information image conversion is conducted through the information, and a training image with a label is obtained.
203. Inputting the training image into a preset neural network model for model training to obtain a wall column recognition model;
specifically, a training image is input into a preset neural network model, and image feature extraction is carried out on the training image through the neural network model to obtain a training feature image; wall column recognition classification and feature point selection are carried out on the training feature images to obtain a target training result; and performing parameter tuning on the neural network model according to the target training result, and taking the neural network model after parameter tuning as a wall column recognition model.
The method comprises the steps that a server inputs a training image into a preset neural network model, extracts an effective area image and pixel point color value data of the training image and performs color block segmentation to obtain image data of the effective area image, performs multiple equal subdivision on the effective area image to obtain a subdivided area of the effective area image, performs connected domain confirmation, line segment identification and line length measurement on the subdivided area to obtain image characteristic data of the subdivided area to obtain a training characteristic image, performs statistics and combination processing on the obtained training characteristic image to obtain image characteristic points, describes image characteristics by using the image characteristic points to obtain a target training result, and finally performs parameter optimization on the neural network model according to the target training result and uses the neural network model with the optimized parameters as a wall column identification model.
204. Acquiring a target original drawing to be recognized, inputting the target original drawing into a wall column recognition model for intelligent wall column recognition, and obtaining target line segment data;
specifically, a target original drawing to be identified is searched from an assembly type collaborative design platform; inputting a target original drawing into a wall column identification model for intelligent wall column identification to obtain initial line segment data; and predicting and marking the position data of the initial line segment data to obtain target line segment data.
The method comprises the steps that a server searches a target original drawing to be identified from an assembled collaborative design platform, identifies the target original drawing, carries out data identification and analysis by combining a preset identification strategy, plans a target area as initial line segment data according to a target object in the target original drawing, carries out fusion identification on the initial line segment data according to preset position parameters, carries out position data prediction and marking on the initial line segment data, and obtains the target line segment data.
205. Parallel line selection is carried out on the target line segment data according to a preset parallel line selection rule to obtain a parallel pairing line;
specifically, the maximum thickness of the wall is generated according to the target line segment data; performing parallel line extraction on the target line segment data according to the maximum thickness of the wall body to obtain an initial parallel line; performing distance distribution operation on the initial parallel lines to obtain distance distribution data; and screening the initial parallel lines according to the distance distribution data to obtain parallel paired lines.
Specifically, the server determines the maximum thickness of the wall body through the target line segment data, counts the distance distribution of all parallel alignment lines, wherein the requirements of the parallel alignment lines are opposite in required direction, the ratio of the length to the distance is larger than four, and then selects the parallel alignment lines meeting the requirements to obtain the parallel alignment lines.
206. Wall edge and column edge classification extraction is carried out on the parallel alignment lines to obtain a wall edge array and a column edge array;
specifically, the server divides the parallel paired lines by wall and column, then processes the single outer contour, starts from the longest line in the unknown attributes, judges that the longest line is the wall if the parallel paired lines meeting the requirements exist, and continues the algorithm. If not, the longest line is the column edge, and then classification and extraction are continuously carried out along the longest line to obtain a wall edge array and a column edge array.
207. And carrying out wall column cutting on the wall side array and the column side array to generate a target wall column drawing.
Specifically, wall column distinguishing is carried out on the wall side array and the column side array to obtain a plurality of outer contours; respectively carrying out attribute analysis on each outer contour to obtain attribute data of each outer contour; and generating a target wall column drawing according to the attribute data of each outer contour.
The classified data are further subjected to corresponding wall column cutting, it is to be noted that the cutting points are concave points of the outer contour, and the front and back of the concave points are required to be a wall edge and a column edge.
When the condition is met, one edge is parallel and reverse to one edge of the cutting point, and the other edge is collinear and homodromous with the other edge of the cutting point. And a cutting line is formed by connecting two points to cut the outer contour, the two cut sub-outer contours return to a link for distinguishing the wall side and the column side again, the second cutting condition is a cutting point with an acute or right-angled included angle, the column side is an extension side, if the column side is a front line of a concave point, the column side is extended in the forward direction, otherwise, the column side is extended in the reverse direction, the extension is stopped when other line segments are met, the shortest line which is extended is used as the cutting line to cut the outer contour, the two cut sub-outer contours return to the link for distinguishing the wall side and the column side again, the third cutting condition is a cutting point with an obtuse included angle, the vertical line of the wall side of the cutting point is used as an extension line, the rest processing is the same as the second cutting condition, and the condition for finishing the cutting does not exist in the three conditions.
Further, wall columns are distinguished according to the edge attributes of the outer contour, if one column edge exists, the wall column is regarded as a column, otherwise, the wall column is regarded as a wall column to generate the whole outer contour as an outer contour line wall body of the column, one-to-one paired wall edges are sequentially combed according to the parallel pairing number of the wall edges, the wall edges are uniquely paired with one another, and one side of the wall edges is connected with one another and has a wall end; and connecting the other ends of the two walls to generate a complete wall, continuously processing the rest parts, and finally generating a target wall stud drawing by the server according to the attribute data of each outer contour, wherein the diagram 7 is a schematic diagram of the target wall stud drawing.
In the embodiment of the invention, a sample drawing is obtained based on an assembled collaborative design platform, and line extraction is carried out on the sample drawing to obtain sample line data; carrying out wall column marking information identification on the sample line data to obtain a training image with a label; inputting the training image into a preset neural network model for model training to obtain a wall column recognition model; acquiring a target original drawing to be recognized, inputting the target original drawing into a wall column recognition model for intelligent wall column recognition, and obtaining target line segment data; and carrying out wall column drawing conversion on the target line segment data to obtain a target wall column drawing. According to the method, when the target original drawing is recognized, the wall and the column can be very quickly and accurately distinguished in the wall column recognition model, and the drawing outline is completely restored.
With reference to fig. 3, the wall stud recognition method based on the assembly type building AI design in the embodiment of the present invention is described above, and a wall stud recognition apparatus based on the assembly type building AI design in the embodiment of the present invention is described below, and an embodiment of the wall stud recognition apparatus based on the assembly type building AI design in the embodiment of the present invention includes:
the acquisition module 301 is configured to acquire a sample drawing based on a preset assembly type collaborative design platform, and perform line extraction on the sample drawing to obtain sample line data;
the identification module 302 is configured to perform wall column labeling information identification on the sample line data to obtain a training image with a label;
the training module 303 is configured to input the training image into a preset neural network model for model training to obtain a wall column recognition model;
the processing module 304 is configured to obtain a target original drawing to be recognized, input the target original drawing into the wall stud recognition model, and perform intelligent wall stud recognition to obtain target line segment data;
and the output module 305 is configured to perform wall column drawing conversion on the target line segment data to obtain a target wall column drawing.
In the embodiment of the invention, a sample drawing is obtained based on an assembled collaborative design platform, and line extraction is carried out on the sample drawing to obtain sample line data; carrying out wall column marking information identification on the sample line data to obtain a training image with a label; inputting the training image into a preset neural network model for model training to obtain a wall column recognition model; acquiring a target original drawing to be recognized, and inputting the target original drawing into the wall column recognition model for intelligent wall column recognition to obtain target line segment data; and converting the wall column drawing of the target line segment data to obtain a target wall column drawing. According to the method, when the target original drawing is recognized, the wall and the column can be very quickly and accurately distinguished in the wall column recognition model, and the drawing outline is completely restored.
Referring to fig. 4, another embodiment of the wall stud recognition apparatus designed based on the prefabricated building AI according to the embodiment of the present invention includes:
the acquisition module 301 is configured to acquire a sample drawing based on a preset assembly type collaborative design platform, and perform line extraction on the sample drawing to obtain sample line data;
the identification module 302 is configured to perform wall column labeling information identification on the sample line data to obtain a training image with a label;
the training module 303 is configured to input the training image into a preset neural network model for model training to obtain a wall stud recognition model;
the processing module 304 is configured to obtain a target original drawing to be recognized, input the target original drawing into the wall stud recognition model, and perform intelligent wall stud recognition to obtain target line segment data;
and the output module 305 is configured to perform wall column drawing conversion on the target line segment data to obtain a target wall column drawing.
Optionally, the identification module 302 is specifically configured to: calling a preset depth analysis detection model to perform line feature recognition on the sample line data to obtain feature identification information; carrying out feature classification on the feature identification information to obtain a target feature class; and performing wall column labeling information image conversion on the sample line data according to the target feature class and the feature identification information to obtain a training image with a label.
Optionally, the training module 303 is specifically configured to: inputting the training image into a preset neural network model, and performing image feature extraction on the training image through the neural network model to obtain a training feature image; wall column recognition classification and feature point selection are carried out on the training feature images to obtain a target training result; and performing parameter tuning on the neural network model according to the target training result, and taking the neural network model after parameter tuning as the wall column recognition model.
Optionally, the processing module 304 is specifically configured to: searching a target original drawing to be identified from the assembly type collaborative design platform; inputting the target original drawing into the wall column identification model for intelligent wall column identification to obtain initial line segment data; and predicting and marking the position data of the initial line segment data to obtain target line segment data.
Optionally, the output module 305 further includes:
the selecting unit 3051 is configured to perform parallel line selection on the target line segment data according to a preset parallel line selection rule to obtain a parallel paired line;
the extracting unit 3052 is configured to perform wall-edge and column-edge classification extraction on the parallel alignment line to obtain a wall-edge array and a column-edge array;
and the cutting unit 3053 is used for cutting the wall column of the wall edge array and the column edge array to generate a target wall column drawing.
Optionally, the selecting unit 3051 is specifically configured to: generating the maximum thickness of the wall according to the target line segment data; performing parallel line extraction on the target line segment data according to the maximum wall thickness to obtain initial parallel lines; performing distance distribution operation on the initial parallel lines to obtain distance distribution data; and screening the initial parallel lines according to the distance distribution data to obtain parallel paired lines.
Optionally, the cutting unit 3053 is specifically configured to: wall column distinguishing is carried out on the wall side array and the column side array, and a plurality of outer profiles are obtained; respectively carrying out attribute analysis on each outer contour to obtain attribute data of each outer contour; and generating a target wall column drawing according to the attribute data of each outer contour.
In the embodiment of the invention, a sample drawing is obtained based on an assembled collaborative design platform, and line extraction is carried out on the sample drawing to obtain sample line data; carrying out wall column marking information identification on the sample line data to obtain a training image with a label; inputting the training image into a preset neural network model for model training to obtain a wall column recognition model; acquiring a target original drawing to be recognized, and inputting the target original drawing into the wall column recognition model for intelligent wall column recognition to obtain target line segment data; and converting the wall column drawing of the target line segment data to obtain a target wall column drawing. According to the method, when the target original drawing is recognized, the wall and the column can be very quickly and accurately distinguished in the wall column recognition model, and the drawing outline is completely restored.
Fig. 3 and 4 describe the wall post recognition device based on the assembly type building AI design in the embodiment of the present invention in detail from the perspective of the modular functional entity, and the wall post recognition device based on the assembly type building AI design in the embodiment of the present invention is described in detail from the perspective of the hardware processing.
Fig. 5 is a schematic structural diagram of a wall stud recognition device designed based on an assembly building AI according to an embodiment of the present invention, where the wall stud recognition device 500 designed based on an assembly building AI may generate relatively large differences due to different configurations or performances, and may include one or more processors (CPUs) 510 (e.g., one or more processors) and a memory 520, and one or more storage media 530 (e.g., one or more mass storage devices) storing an application program 533 or data 532. Memory 520 and storage media 530 may be, among other things, transient or persistent storage. The program stored in the storage medium 530 may include one or more modules (not shown), each of which may include a series of instruction operations in the wall stud recognition device 500 designed based on the prefabricated building AI. Still further, the processor 510 may be configured to communicate with the storage medium 530, and execute a series of instruction operations in the storage medium 530 on the wall stud recognition device 500 designed based on the prefabricated building AI.
Wall stud identification apparatus 500 based on the fabricated building AI design may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input-output interfaces 560, and/or one or more operating systems 531, such as Windows Server, mac OS X, unix, linux, freeBSD, and the like. It will be understood by those skilled in the art that the structure of the wall stud recognition device based on the assembly building AI design shown in fig. 5 does not constitute a limitation of the wall stud recognition device based on the assembly building AI design, and may include more or fewer components than those shown, or some components in combination, or a different arrangement of components.
The invention also provides a wall stud recognition device based on the assembly type building AI design, which comprises a memory and a processor, wherein the memory stores computer readable instructions, and when the computer readable instructions are executed by the processor, the processor executes the steps of the wall stud recognition method based on the assembly type building AI design in the above embodiments.
The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium, and which may also be a volatile computer-readable storage medium, having stored therein instructions, which, when run on a computer, cause the computer to perform the steps of the wall stud recognition method based on an assembly building AI design.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A wall column identification method based on an assembly type building AI design is characterized by comprising the following steps:
acquiring a sample drawing based on a preset assembly type collaborative design platform, and carrying out line extraction on the sample drawing to obtain sample line data;
carrying out wall column marking information identification on the sample line data to obtain a training image with a label;
inputting the training image into a preset neural network model for model training to obtain a wall column recognition model;
acquiring a target original drawing to be recognized, and inputting the target original drawing into the wall column recognition model for intelligent wall column recognition to obtain target line segment data;
and converting the wall column drawing of the target line segment data to obtain a target wall column drawing.
2. The wall post identification method based on assembly type building AI design according to claim 1, wherein the identifying the sample line data with wall post labeling information to obtain a labeled training image comprises:
calling a preset depth analysis detection model to perform line feature recognition on the sample line data to obtain feature identification information;
carrying out feature classification on the feature identification information to obtain a target feature class;
and performing wall column labeling information image conversion on the sample line data according to the target feature class and the feature identification information to obtain a training image with a label.
3. The wall post recognition method based on the assembly type building AI design as claimed in claim 1, wherein the inputting the training image into a preset neural network model for model training to obtain the wall post recognition model comprises:
inputting the training image into a preset neural network model, and performing image feature extraction on the training image through the neural network model to obtain a training feature image;
carrying out wall column recognition classification and feature point selection on the training feature images to obtain a target training result;
and performing parameter tuning on the neural network model according to the target training result, and taking the neural network model after parameter tuning as the wall column recognition model.
4. The wall stud identification method based on the assembly type building AI design according to claim 1, wherein the obtaining of the target original drawing to be identified and the inputting of the target original drawing into the wall stud identification model for intelligent wall stud identification to obtain target line segment data comprises:
searching a target original drawing to be identified from the assembly type collaborative design platform;
inputting the target original drawing into the wall column identification model for intelligent wall column identification to obtain initial line segment data;
and predicting and marking the position data of the initial line segment data to obtain target line segment data.
5. The wall stud identification method based on the assembly type building AI design of claim 1, wherein the wall stud drawing conversion of the target line segment data to obtain a target wall stud drawing comprises:
parallel line selection is carried out on the target line segment data according to a preset parallel line selection rule to obtain a parallel pairing line;
carrying out wall edge and column edge classification extraction on the parallel alignment line to obtain a wall edge array and a column edge array;
and carrying out wall column cutting on the wall side array and the column side array to generate a target wall column drawing.
6. The wall column identification method based on the assembly type building AI design of claim 5, wherein the parallel line selection of the target line segment data according to the preset parallel line selection rule to obtain the parallel paired lines comprises:
generating the maximum thickness of the wall according to the target line segment data;
performing parallel line extraction on the target line segment data according to the maximum wall thickness to obtain initial parallel lines;
performing distance distribution operation on the initial parallel lines to obtain distance distribution data;
and screening the initial parallel lines according to the distance distribution data to obtain parallel paired lines.
7. The wall stud recognition method based on assembly type building AI design according to claim 5, wherein the wall stud cutting of the wall edge array and the stud edge array to generate the target wall stud drawing comprises:
carrying out wall column distinguishing on the wall side array and the column side array to obtain a plurality of outer contours;
respectively carrying out attribute analysis on each outer contour to obtain attribute data of each outer contour;
and generating a target wall column drawing according to the attribute data of each outer contour.
8. A wall post recognition device based on assembly type structure AI design, its characterized in that, wall post recognition device based on assembly type structure AI design includes:
the acquisition module is used for acquiring a sample drawing based on a preset assembly type collaborative design platform and carrying out line extraction on the sample drawing to obtain sample line data;
the identification module is used for carrying out wall column marking information identification on the sample line data to obtain a training image with a label;
the training module is used for inputting the training image into a preset neural network model for model training to obtain a wall column recognition model;
the processing module is used for acquiring a target original drawing to be recognized, inputting the target original drawing into the wall column recognition model for intelligent wall column recognition, and obtaining target line segment data;
and the output module is used for carrying out wall column drawing conversion on the target line segment data to obtain a target wall column drawing.
9. A wall post recognition device based on assembly type structure AI design, its characterized in that, wall post recognition device based on assembly type structure AI design includes: a memory and at least one processor, the memory having instructions stored therein;
the at least one processor invokes the instructions in the memory to cause the wall stud recognition apparatus based on an assembled building AI design to perform the wall stud recognition method based on an assembled building AI design according to any one of claims 1-7.
10. A computer-readable storage medium having instructions stored thereon, wherein the instructions, when executed by a processor, implement the wall stud identification method based on an AI design of a fabricated building according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310056470.9A CN115797962B (en) | 2023-01-13 | 2023-01-13 | Wall column identification method and device based on assembly type building AI design |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310056470.9A CN115797962B (en) | 2023-01-13 | 2023-01-13 | Wall column identification method and device based on assembly type building AI design |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115797962A true CN115797962A (en) | 2023-03-14 |
| CN115797962B CN115797962B (en) | 2023-05-02 |
Family
ID=85429684
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310056470.9A Active CN115797962B (en) | 2023-01-13 | 2023-01-13 | Wall column identification method and device based on assembly type building AI design |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115797962B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116311336A (en) * | 2023-03-20 | 2023-06-23 | 华联世纪工程咨询股份有限公司 | Automatic beam information identification and acquisition method |
| CN116680778A (en) * | 2023-04-27 | 2023-09-01 | 清华大学 | Building structure arrangement generation method and system and construction method of generation model |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205198A (en) * | 2014-06-30 | 2015-12-30 | 上海神机软件有限公司 | Full-automatic intelligent integrated formwork arranging system and method for construction engineering fabricated formworks |
| CN108763813A (en) * | 2018-06-08 | 2018-11-06 | 杭州群核信息技术有限公司 | It is a kind of to identify the method and apparatus for copying wall in figure based on deep learning |
| CN108804815A (en) * | 2018-06-08 | 2018-11-13 | 杭州群核信息技术有限公司 | A kind of method and apparatus assisting in identifying wall in CAD based on deep learning |
| CN111832437A (en) * | 2020-06-24 | 2020-10-27 | 万翼科技有限公司 | Building drawing identification method, electronic equipment and related product |
| CN114139701A (en) * | 2021-11-15 | 2022-03-04 | 土巴兔集团股份有限公司 | Neural network model training method for boundary line extraction and related equipment |
| CN114417486A (en) * | 2022-03-14 | 2022-04-29 | 深圳市斯维尔科技股份有限公司 | Building design drawing importing method and device, electronic equipment and storage medium |
| CN114444168A (en) * | 2021-12-27 | 2022-05-06 | 万翼科技有限公司 | Method and device for identifying wall column in building drawing, electronic equipment and storage medium |
| CN114936395A (en) * | 2022-04-27 | 2022-08-23 | 土巴兔集团股份有限公司 | Household type graph recognition method and device, computer equipment and storage medium |
-
2023
- 2023-01-13 CN CN202310056470.9A patent/CN115797962B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205198A (en) * | 2014-06-30 | 2015-12-30 | 上海神机软件有限公司 | Full-automatic intelligent integrated formwork arranging system and method for construction engineering fabricated formworks |
| CN108763813A (en) * | 2018-06-08 | 2018-11-06 | 杭州群核信息技术有限公司 | It is a kind of to identify the method and apparatus for copying wall in figure based on deep learning |
| CN108804815A (en) * | 2018-06-08 | 2018-11-13 | 杭州群核信息技术有限公司 | A kind of method and apparatus assisting in identifying wall in CAD based on deep learning |
| CN111832437A (en) * | 2020-06-24 | 2020-10-27 | 万翼科技有限公司 | Building drawing identification method, electronic equipment and related product |
| CN114139701A (en) * | 2021-11-15 | 2022-03-04 | 土巴兔集团股份有限公司 | Neural network model training method for boundary line extraction and related equipment |
| CN114444168A (en) * | 2021-12-27 | 2022-05-06 | 万翼科技有限公司 | Method and device for identifying wall column in building drawing, electronic equipment and storage medium |
| CN114417486A (en) * | 2022-03-14 | 2022-04-29 | 深圳市斯维尔科技股份有限公司 | Building design drawing importing method and device, electronic equipment and storage medium |
| CN114936395A (en) * | 2022-04-27 | 2022-08-23 | 土巴兔集团股份有限公司 | Household type graph recognition method and device, computer equipment and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116311336A (en) * | 2023-03-20 | 2023-06-23 | 华联世纪工程咨询股份有限公司 | Automatic beam information identification and acquisition method |
| CN116680778A (en) * | 2023-04-27 | 2023-09-01 | 清华大学 | Building structure arrangement generation method and system and construction method of generation model |
| CN116680778B (en) * | 2023-04-27 | 2024-03-12 | 清华大学 | Building structure arrangement generation method and system and construction method of generation model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115797962B (en) | 2023-05-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110738207B (en) | Character detection method for fusing character area edge information in character image | |
| US20210342586A1 (en) | Systems and methods for extracting and vectorizing features of satellite imagery | |
| CN115797962A (en) | Wall column identification method and device based on assembly type building AI design | |
| Shafait et al. | Performance evaluation and benchmarking of six-page segmentation algorithms | |
| CN110838105B (en) | Business process model image recognition and reconstruction method | |
| Natali et al. | Graph-based representations of point clouds | |
| CN114926699B (en) | Indoor three-dimensional point cloud semantic classification method, device, medium and terminal | |
| CN103106265A (en) | Method and system of classifying similar images | |
| JP7173309B2 (en) | LEARNING METHOD, LEARNING PROGRAM AND LEARNING APPARATUS | |
| CN113239818B (en) | Table cross-modal information extraction method based on segmentation and graph convolution neural network | |
| CN115586749B (en) | Workpiece machining track control method based on machine vision and related device | |
| Russakovsky et al. | A steiner tree approach to efficient object detection | |
| Moitra et al. | Cluster-based data reduction for persistent homology | |
| CN114387608A (en) | Table structure identification method combining convolution and graph neural network | |
| CN107533760A (en) | A kind of image partition method and device | |
| CN116945258A (en) | Die cutting machine control system and method thereof | |
| Peng et al. | Hers superpixels: Deep affinity learning for hierarchical entropy rate segmentation | |
| Chiang et al. | Voxel-based shape decomposition for feature-preserving 3D thumbnail creation | |
| Ramel et al. | Automatic reading of handwritten chemical formulas from a structural representation of the image | |
| CN114387600A (en) | Text feature recognition method and device, computer equipment and storage medium | |
| Robert Raj et al. | Offline tamil handwritten character recognition using statistical based quad tree | |
| Wu | Research on embroidery image recognition based on deep learning | |
| Jayarathna et al. | A junction based segmentation algorithm for offline handwritten connected character segmentation | |
| Ito et al. | Interactive region segmentation for manga | |
| MAHURKAR et al. | Novel Outline Tracing Techniques for Leaf Species Identification from Shrouded Leaves |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |