CN110096949A - Floor plan intelligent identification Method, device, equipment and computer readable storage medium - Google Patents
Floor plan intelligent identification Method, device, equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN110096949A CN110096949A CN201910205773.6A CN201910205773A CN110096949A CN 110096949 A CN110096949 A CN 110096949A CN 201910205773 A CN201910205773 A CN 201910205773A CN 110096949 A CN110096949 A CN 110096949A
- Authority
- CN
- China
- Prior art keywords
- floor plan
- wall
- dimentional
- model
- preset
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000012549 training Methods 0.000 claims description 37
- 238000013527 convolutional neural network Methods 0.000 claims description 11
- 230000006870 function Effects 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 4
- 238000000638 solvent extraction Methods 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 abstract description 3
- 230000000694 effects Effects 0.000 description 8
- 230000011218 segmentation Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000010606 normalization Methods 0.000 description 4
- 238000005034 decoration Methods 0.000 description 2
- 230000005284 excitation Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 239000010903 husk Substances 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 230000003014 reinforcing effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
- G06V30/42—Document-oriented image-based pattern recognition based on the type of document
- G06V30/422—Technical drawings; Geographical maps
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Structural Engineering (AREA)
- Civil Engineering (AREA)
- Architecture (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Graphics (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to field of artificial intelligence, disclose a kind of floor plan intelligent identification Method, comprising: obtain two-dimentional floor plan to be identified;The two-dimentional floor plan is identified based on preset wall identification model, exports the wall information in the two-dimentional floor plan;The corresponding three-dimensional house type illustraton of model of the two-dimentional floor plan is generated based on the wall information, and exports the three-dimensional house type illustraton of model.The invention also discloses a kind of floor plan intelligent identification device, equipment and computer readable storage mediums.The present invention improves the accuracy rate that wall identifies in floor plan.
Description
Technical field
The present invention relates to field of artificial intelligence more particularly to a kind of floor plan intelligent identification Method, device, equipment and
Computer readable storage medium.
Background technique
Change with the fast development and modern of China Real Estate economy to house decoration demand, the finishing of house
It is more and more personalized, but these decoration styles often with actual house type and mismatch.If first plane floor plan converted
For 3D floor plan, finishing effect is checked in 3D floor plan, then practical finishing effect can just be made to reach best.Also have at present
The floor plan of plane is much converted to the patent and method of 3D floor plan, but identifies the actual effect of wall in these patents
Barely satisfactory, therefore, how preferably to identify that wall is difficult point urgently to be resolved.
Summary of the invention
The main purpose of the present invention is to provide a kind of floor plan intelligent identification Method, device, equipment and computer-readable
Storage medium, it is intended to solve the not high technical problem of wall recognition accuracy in the prior art in floor plan.
To achieve the above object, the present invention provides a kind of floor plan intelligent identification Method, the floor plan intelligent recognition side
Method includes:
Obtain two-dimentional floor plan to be identified;
The two-dimentional floor plan is identified based on preset wall identification model, is exported in the two-dimentional floor plan
Wall information;
The corresponding three-dimensional house type illustraton of model of the two-dimentional floor plan is generated based on the wall information, and exports the three-dimensional
House type illustraton of model.
Optionally, described the step of obtaining two-dimentional floor plan to be identified, includes:
When receiving the wall identification request based on two-dimentional floor plan, obtains the wall identification and request corresponding two dimension
Floor plan.
Optionally, described that the two-dimentional floor plan is identified based on preset wall identification model, output described two
Before the step of tieing up the wall information in floor plan, further includes:
Two-dimentional floor plan sample to be trained is obtained, and mould is identified to preset wall based on the two-dimentional floor plan sample
Type carries out off-line training;
The preset wall identification model after saving off-line training.
Optionally, described to obtain two-dimentional floor plan sample to be trained, and based on the two-dimentional floor plan sample to default
Wall identification model carry out off-line training the step of include:
Two-dimentional floor plan sample to be trained is obtained, and the two-dimentional floor plan sample is pre-processed, wherein is described
Two-dimentional floor plan sample to be trained has wall markup information;
The pretreated two-dimentional floor plan sample with wall markup information is input to preset convolutional Neural
In network, so that the preset convolutional neural networks export the characteristic pattern of the two-dimentional floor plan sample;
Case-based Reasoning partitioning algorithm detects the wall target in the characteristic pattern, when determining that current detection reaches receipts
When holding back condition, confirm that the preset wall identification model off-line training is completed.
Optionally, the condition of convergence include frequency of training, the training time, training loss function.
Optionally, described that the two-dimentional floor plan is identified based on preset wall identification model, output described two
The step of tieing up the wall information in floor plan generates the corresponding three-dimensional of the two dimension floor plan based on the wall information with described
House type illustraton of model, and between the step of exporting the three-dimensional house type illustraton of model, further includes:
Preset floor plan component recognition model is obtained, and based on the floor plan component recognition model to the two-dimentional family
Component in type figure is identified, determines the component position information in the two-dimentional floor plan;
The wall information is modified based on the component position information.
Optionally, described that the corresponding three-dimensional house type illustraton of model of the two-dimentional floor plan is generated based on the wall information, and
The step of exporting the three-dimensional house type illustraton of model include:
Based on preset three-dimensional wall the build tool, the two-dimentional floor plan corresponding three is generated according to the wall information
House type illustraton of model is tieed up, and exports the three-dimensional house type illustraton of model.
In addition, to achieve the above object, the present invention also provides a kind of floor plan intelligent identification device, the floor plan intelligence
Identification device includes:
Floor plan obtains module, for obtaining two-dimentional floor plan to be identified;
Wall identification module is exported for being identified based on preset wall identification model to the two-dimentional floor plan
Wall information in the two dimension floor plan;
Model construction module, for generating the corresponding three-dimensional house type model of the two-dimentional floor plan based on the wall information
Figure, and export the three-dimensional house type illustraton of model.
In addition, to achieve the above object, the present invention also provides a kind of floor plan intelligent identification equipment, the floor plan intelligence
Identification equipment includes: memory, processor and is stored in the floor plan that can be run on the memory and on the processor
Intelligent recognition program, the floor plan intelligent recognition program realize floor plan intelligence as described above when being executed by the processor
The step of recognition methods.
In addition, to achieve the above object, it is described computer-readable the present invention also provides a kind of computer readable storage medium
It is stored with floor plan intelligent recognition program on storage medium, is realized such as when the floor plan intelligent recognition program is executed by processor
Above the step of floor plan intelligent identification Method.
The invention belongs to field of artificial intelligence, propose a kind of floor plan intelligent identification Method, are obtained first wait know
Other two dimension floor plan, and the two dimension floor plan is identified based on preset wall identification model, so as to preset wall
Identification model exports the wall information in the two dimension floor plan, finally generates the corresponding three-dimensional of two-dimentional floor plan according to wall information
House type illustraton of model, and export three-dimensional house type illustraton of model.Floor plan intelligent identification Method proposed by the present invention, passes through preset wall
Identification model carries out wall identification, and identification wall method is novel, recognition speed is fast and accuracy rate is high, more facilitates three-dimensional house type
The building of figure.
Detailed description of the invention
Fig. 1 is the hardware structural diagram of floor plan intelligent identification equipment involved in the embodiment of the present invention;
Fig. 2 is the flow diagram of floor plan intelligent identification Method first embodiment of the present invention;
Fig. 3 is the flow diagram of floor plan intelligent identification Method second embodiment of the present invention;
Fig. 4 is the functional block diagram of floor plan intelligent identification device first embodiment of the present invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
The primary solutions of the embodiment of the present invention are: obtaining two-dimentional floor plan to be identified;Known based on preset wall
Other model identifies the two-dimentional floor plan, exports the wall information in the two-dimentional floor plan;Believed based on the wall
Breath generates the corresponding three-dimensional house type illustraton of model of the two-dimentional floor plan, and exports the three-dimensional house type illustraton of model.Through the invention
The technical solution of embodiment solves the not high technical problem of the wall recognition accuracy of floor plan in the prior art.
As shown in FIG. 1, FIG. 1 is the hardware configurations of the floor plan intelligent identification equipment involved in the embodiment of the present invention to show
It is intended to.
The present embodiments relate to floor plan intelligent identification Method be mainly used in floor plan intelligent identification equipment, the family
Type figure intelligent identification equipment can be the equipment that PC, portable computer, mobile terminal etc. have display and processing function.
As shown in Figure 1, the floor plan intelligent identification equipment may include: processor 1001, such as CPU, communication bus
1002, user interface 1003, network interface 1004, memory 1005.Wherein, communication bus 1002 for realizing these components it
Between connection communication.User interface 1003 may include display screen (Display), input unit such as keyboard (Keyboard),
Optional user interface 1003 can also include standard wireline interface and wireless interface.Network interface 1004 optionally may include
Standard wireline interface and wireless interface (such as WI-FI interface).Memory 1005 can be high speed RAM memory, be also possible to steady
Fixed memory (non-volatile memory), such as magnetic disk storage.Memory 1005 optionally can also be independently of
The storage device of aforementioned processor 1001.
Optionally, floor plan intelligent identification equipment can also include camera, RF (Radio Frequency, radio frequency) electricity
Road, sensor, voicefrequency circuit, Wi-Fi module etc..Wherein, sensor such as optical sensor, motion sensor and other biographies
Sensor.Certainly, floor plan intelligent identification equipment can also configure gyroscope, barometer, hygrometer, thermometer, infrared sensor
Etc. other sensors, details are not described herein.
It will be understood by those skilled in the art that floor plan intelligent identification equipment structure shown in Fig. 1 is not constituted to family
The restriction of type figure intelligent identification equipment may include perhaps combining certain components or not than illustrating more or fewer components
Same component layout.
As shown in Figure 1, as may include that operating system, network are logical in a kind of memory 1005 of computer storage medium
Believe module, Subscriber Interface Module SIM and floor plan intelligent recognition program.In Fig. 1, after network interface 1004 is mainly used for connection
Platform server carries out data communication with background server;User interface 1003 is mainly used for connecting client (user terminal), with visitor
Family end carries out data communication;And processor 1001, memory 1005 can be set in floor plan intelligent identification device, the family
Type figure intelligent identification device calls the floor plan intelligent recognition program stored in memory 1005 by processor 1001, and executes
Floor plan intelligent identification Method provided in an embodiment of the present invention.
Scheme provided in this embodiment obtains two-dimentional floor plan to be identified first, and identifies mould based on preset wall
Type identifies the two dimension floor plan, so that preset wall identification model exports the wall information in the two dimension floor plan,
The corresponding three-dimensional house type illustraton of model of two-dimentional floor plan is finally generated according to wall information, and exports three-dimensional house type illustraton of model.This hair
The floor plan intelligent identification Method of bright proposition carries out wall identification by preset wall identification model, and identification wall method is new
Grain husk, recognition speed is fast and accuracy rate is high, more facilitates the building of three-dimensional floor plan.
Based on above-mentioned hardware configuration, floor plan intelligent identification Method embodiment of the present invention is proposed.
Referring to Fig. 2, Fig. 2 is the flow diagram of floor plan intelligent identification Method first embodiment of the present invention, in the implementation
In example, which comprises
Step S10 obtains two-dimentional floor plan to be identified;
In order to solve the problems, such as that wall discrimination is low in existing two-dimentional floor plan, the invention proposes a kind of floor plan intelligence
Recognition methods carries out wall online recognition to two-dimentional floor plan to be identified by preset wall identification model, determines wherein
Wall information so that user is passed through three-dimensional house type illustraton of model more preferable to be based on the wall information architecture three-dimensional house type illustraton of model
Ground understands house type construction.
Firstly, when user or staff need to carry out the wall identification of floor plan, it can be in wall identification terminal
It is operated, uploads the two-dimentional floor plan that need to carry out wall identification, and trigger the identification instruction of floor plan wall, wherein the wall
Identification terminal can be personal computer, mobile phone, tablet computer etc..When wall identification terminal is received based on two-dimentional floor plan
When wall identification request, the corresponding two-dimentional floor plan of wall identification request can be obtained, and put it into preset wall and know
In other model, so that the preset wall identification model identifies the wall in two-dimentional floor plan one by one.
Step S20 identifies the two-dimentional floor plan based on preset wall identification model, exports the two-dimentional family
Wall information in type figure;
In the present embodiment, wall identification is carried out using two-dimentional floor plan of the preset wall identification model to input, with
Exporting wall information therein, it is to be understood that the preset wall identification model is to obtain in advance by off-line training,
Specifically, before the step S20, further includes:
Step a obtains two-dimentional floor plan sample to be trained, and based on the two-dimentional floor plan sample to preset wall
Identification model carries out off-line training;
By obtaining two-dimentional floor plan sample largely to be trained, preset wall identification model is instructed offline
Practice, in the present embodiment, being somebody's turn to do two-dimentional floor plan sample to be trained is to have the two-dimensional surface floor plan sample marked on a small quantity,
In mark can be wall, decorative components, house type size and the housing area etc. in floor plan are labeled.Specifically,
The process for carrying out off-line training is as follows:
Firstly, being pre-processed to the two-dimentional floor plan sample being input in preset wall identification model, including picture
Unification, the pixel value normalization (such as linear function normalization, logarithmic function normalization, arc cotangent function normalization) of size
It is to carry out feature reinforcement to two-dimentional floor plan sample Deng, pretreated purpose.
Further, example segmentation carried out to pretreated two-dimentional floor plan sample, example segmentation be target identification with
One combination of semantic segmentation, wherein target identification is to detect there is what target in piece image, while using calibration frame
(a generally rectangle frame) comes out the location position of target;And semantic segmentation is to delimit each pixel in an image
Classify to fixed classification, that is, to each pixel.Example segmentation involved in this embodiment, can be and pass through
MASK R-CNN network implementations, specific implementation following steps:
Pretreated two-dimentional floor plan sample is input in preset convolutional neural networks, in this implementation by the first step
In example, preset convolutional neural networks include 13 convolutional layers, 13 excitation layers and 4 pond layers, for the convolution of convolutional layer
Core kernel is 3*3, and Filling power 1, the effect of Filling power is to change convolutional layer not and output and input matrix size;Pond
The convolution kernel for changing layer is 2*2, stride 2*2.By carrying out the operation such as convolution, excitation, pond to two-dimentional floor plan sample, can obtain
To a feature vector, this feature vector is the vector information for characterizing the corresponding characteristic pattern of two dimension floor plan sample.
Second step carries out the judgement and classification of target area using RPN (region candidate network).Specifically, to this feature
The ROI (region of interest, area-of-interest) of each of figure point setting predetermined number, therefore can get multiple
ROI.Further, two-value classification is carried out to it using RPN network and BB is returned, a part of invalid ROI is filtered out, is had
The ROI region of effect, the region in the present embodiment where described target area, that is, wall.
Third step does semantic segmentation to effective ROI region, specifically, using ROI Align to the area of each ROI
Characteristic of field figure carry out pixel correction, each ROI is predicted according to the provincial characteristics figure of each ROI, obtain its classification of each ROI with
And the boundary of each ROI, it is pre- by preset FCN (Fully Convolutional Network, simply comb full convolutional network)
Each pixel generic in each ROI is surveyed, wall position is determined according to class prediction result.
Based on above procedure, preset wall identification model is constantly instructed offline by two-dimentional floor plan sample
Practice, during off-line training, when detecting that current training has reached the condition of convergence, that is, can determine that present offline training is completed,
Specifically, the condition of convergence can include but is not limited to: frequency of training, the training time, training loss function, can be according to practical feelings
Condition is configured.
Step b, the preset wall identification model after saving off-line training.
After the completion of the off-line training for determining preset wall identification model, the wall after saving the off-line training identifies mould
Type, to be used in the online wall identification of two-dimentional floor plan.
Specifically, the step of two-dimentional floor plan being identified based on the wall identification model after off-line training, and it is above-mentioned
The step of during off-line training, is identical, first by pre-processing to two-dimentional floor plan, carries out feature to two-dimentional floor plan
Reinforce;Secondly, pretreated two-dimentional floor plan is input in preset convolutional neural networks, the extraction of characteristic pattern is carried out;
Further, wall and its region in characteristic pattern are determined using RPN;Finally, to the area comprising wall identified
Domain carries out semantic segmentation, final to determine wall position, i.e. output wall position information.
In the present embodiment, in order to improve the discrimination of wall, except through the MASK in preset wall identification model
Except R-CNN identifies wall, can also by multistage refinement network RefineNet to wall recognition result do into
The amendment of one step.Specifically, the wall feature in two-dimentional floor plan is extracted by residual error convolutional network first, then passes through MPR (multichannel
Diameter RefineNet) use high-resolution feature to improve the wall feature that said extracted goes out, i.e., the different resolution that will be extracted
Characteristic pattern merged, a fused characteristic pattern is obtained, finally by being connected to max layers of soft and bilinearity difference
Wall recognition result is exported, based on the wall recognition result of RefineNet output, to the wall of preset wall identification model output
Body information is modified, and obtains the high wall recognition result of final accuracy rate.
Step S30 generates the corresponding three-dimensional house type illustraton of model of the two-dimentional floor plan based on the wall information, and exports
The three-dimensional house type illustraton of model.
It is after going out the wall information in two-dimentional floor plan by preset wall identification model online recognition, its is corresponding
Wall information is converted to coordinate json file, to carry out three-dimensional wall building based on coordinate json file, and exports building
Three-dimensional house type illustraton of model afterwards makes user be best understood from house type construction by three-dimensional house type illustraton of model.
Specifically, two-dimentional floor plan is identified based on preset wall identification model, exports the location information of wall,
Then according to preset wall height and the wall position, three-dimensional house type illustraton of model is constructed, i.e., is erect in wall position default
To generate the three-dimensional house type illustraton of model of wall, above-mentioned wall height can be based on the wall of wall height by those skilled in the art
Actual conditions are configured.After three-dimensional house type illustraton of model constructs successfully, i.e., the exportable three-dimensional house type illustraton of model is looked into convenient for user
See 3D effect floor plan.
In the present embodiment, two-dimentional floor plan to be identified is obtained first, and based on preset wall identification model to this
Two-dimentional floor plan is identified, so that preset wall identification model exports the wall information in the two dimension floor plan, last root
The corresponding three-dimensional house type illustraton of model of two-dimentional floor plan is generated according to wall information, and exports three-dimensional house type illustraton of model.The present invention proposes
Floor plan intelligent identification Method, pass through preset wall identification model and carry out wall identification, identification wall method is novel, identification
Speed is fast and accuracy rate is high, more facilitates the building of three-dimensional floor plan.
Further, referring to Fig. 3, based on the above embodiment, propose that floor plan intelligent identification Method second of the present invention is implemented
Example, in the present embodiment, between the step S20 and the step S30 further include:
Step S40 obtains preset floor plan component recognition model, and based on the floor plan component recognition model to institute
The component stated in two-dimentional floor plan is identified, determines the component position information in the two-dimentional floor plan;
It in the present embodiment, can also be by identifying two-dimentional family in order to further improve the accuracy rate of wall identification
Decorative components information in type figure judges information by accident to wipe the wall generated during wall Direct Recognition.
After getting two-dimentional floor plan to be identified, it is input in preset floor plan component recognition model, with
Just the floor plan component recognition model identifies the module information in two-dimentional floor plan.Specifically, preset floor plan component is known
The process that other model identifies module information is similar to the process of above-mentioned carry out wall identification, firstly, to two-dimentional floor plan
It is pre-processed, is reinforced to carry out the feature of two-dimentional floor plan;Further, by preset convolutional neural networks to pre- place
Two-dimentional floor plan after reason carries out feature extraction, obtains corresponding characteristic pattern, and sets to each of characteristic pattern point pre-
If the ROI of number, to obtain multiple a ROI, then multiple ROI is screened, filter out a part of ROI;Finally, to effective
ROI region does semantic segmentation, identifies decorative components in two-dimentional floor plan, including furniture, cooker, electric appliance etc..
It is understood that the floor plan component recognition model of above-mentioned carry out component recognition, be also possible in advance by from
What line training obtained, the process for carrying out off-line training to floor plan component recognition model is carried out with to preset wall identification model
The process of off-line training is similar, and details are not described herein, and by off-line training, the available higher model of recognition accuracy is mentioned
The high subsequent recognition speed and accuracy for carrying out wall or component online recognition.
Step S50 is modified the wall information based on the component position information.
Further, by the module information in the two-dimentional floor plan that identifies, preset wall identification model is identified
Wall information out is modified, for example, similarly knowing at the wall position that the preset wall identification model of discovery identifies
Chu there are not decorative components, such as furniture, cooker, electric appliance, show that the regional location can not have wall, then to preset wall
The wall information that body identification model identifies is wiped, and the accuracy of wall identification is improved.
In the present embodiment, when the wall gone out in two-dimentional floor plan by preset wall identification model online recognition is believed
Breath, and after being modified based on the module information that floor plan component recognition model identifies to it, accurate wall information is obtained,
And the building of three-dimensional wall is carried out based on preset three-dimensional wall the build tool, to obtain three-dimensional house type illustraton of model.Meanwhile may be used also
3D visualization is carried out with components such as furniture, cooker, electric appliances in the two-dimentional floor plan that will identify that, so that three-dimensional house type illustraton of model
In reality it is more three-dimensional, true.
Specifically, in the three-dimensional house type illustraton of model comprising three-dimensional wall constructed in the first embodiment, in addition identifying
Module information, and 3D show on the basis of, in addition the Changes in weather such as rain, snow, frost and early, middle and late light variation, realize
The real reinforcing effect of three-dimensional house type illustraton of model can also check decorative effect in 3D floor plan, then can just make reality
Decorative effect reaches best.The specific implementation process is as follows:
Corresponding wall information and module information are converted into coordinate json file, and utilize three-dimensional wall the build tool
Unity carries out the building of three-dimensional house type illustraton of model, includes wall and decorative components in the three-dimensional house type illustraton of model of output, as bed,
Sofa etc. can also correspond to the respective text information of mark, dimension information;Meanwhile it can be with for the three-dimensional house type illustraton of model
In such a way that screen slides, the model in rotation and scaling scene can also pass through simulation in the three-dimensional house type illustraton of model
It the special efficacys such as rains, snow, hazing, can also simulate morning, in, the light variation caused by the time changes such as evening in illustraton of model,
Keep user experience more rich and varied.
In the present embodiment, by preset floor plan component recognition model, to the group in two-dimentional floor plan to be identified
Part is identified, to determine the component position information in two-dimentional floor plan, and based on the component position information to the wall identified
Body position information is wiped, is corrected, and the accuracy rate of wall identification is improved;At the same time it can also pass through wall information architecture
Three-dimensional house type illustraton of model in, to identifying that the component of location information also carries out 3D displaying, and increased by addition weather, light etc.
Strong reality special efficacy, keeps three-dimensional house type illustraton of model more three-dimensional, true, improves the experience of user.
In addition, the embodiment of the present invention also provides a kind of floor plan intelligent identification device.
It is the functional block diagram of floor plan intelligent identification device first embodiment of the present invention referring to Fig. 4, Fig. 4.
In the present embodiment, the floor plan intelligent identification device includes:
Floor plan obtains module 10, for obtaining two-dimentional floor plan to be identified;
Wall identification module 20, it is defeated for being identified based on preset wall identification model to the two-dimentional floor plan
Wall information in the two-dimentional floor plan out;
Model construction module 30, for generating the corresponding three-dimensional house type mould of the two-dimentional floor plan based on the wall information
Type figure, and export the three-dimensional house type illustraton of model.
Further, the floor plan obtains module 10 and specifically includes:
Floor plan acquiring unit, for obtaining the wall when receiving the wall identification request based on two-dimentional floor plan
The corresponding two-dimentional floor plan of body identification request.
Further, the floor plan intelligent identification device further include:
Off-line training unit, for obtaining two-dimentional floor plan sample to be trained, and based on the two-dimentional floor plan sample
Off-line training is carried out to preset wall identification model;
Model storage unit, for saving the preset wall identification model after off-line training.
Further, the off-line training unit specifically includes:
Sample preprocessing subelement, for obtaining two-dimentional floor plan sample to be trained, and to the two-dimentional house type pattern
This is pre-processed, wherein the two-dimentional floor plan sample to be trained has wall markup information;
Characteristic pattern determines subelement, for by the pretreated two-dimentional floor plan sample with wall markup information
It is input in preset convolutional neural networks, so that the preset convolutional neural networks export the two-dimentional floor plan sample
Characteristic pattern;
Wall target detection subelement examines the wall target in the characteristic pattern for Case-based Reasoning partitioning algorithm
It surveys, when determining that current detection reaches the condition of convergence, confirms that the preset wall identification model off-line training is completed.
Further, the floor plan intelligent identification device further include:
Component recognition unit is known for obtaining preset floor plan component recognition model, and based on the floor plan component
Other model identifies the component in the two-dimentional floor plan, determines the component position information in the two-dimentional floor plan;
Wall Information revision unit, for being modified based on the component position information to the wall information.
Further, the model construction module 30 specifically includes:
Model construction unit, for being based on preset three-dimensional wall the build tool, according to wall information generation
The corresponding three-dimensional house type illustraton of model of two-dimentional floor plan, and export the three-dimensional house type illustraton of model.
Wherein, in above-mentioned floor plan intelligent identification device in modules and above-mentioned floor plan intelligent identification Method embodiment
Each step is corresponding, and function and realization process no longer repeat one by one here.
In addition, the embodiment of the present invention also proposes a kind of computer readable storage medium, the computer readable storage medium
On be stored with floor plan intelligent recognition program, realized when the floor plan intelligent recognition program is executed by processor as described above
The step of floor plan intelligent identification Method.
Wherein, floor plan intelligent recognition program, which is performed realized method, can refer to floor plan intelligent recognition of the present invention
Each embodiment of method, details are not described herein again.
Scheme provided in this embodiment obtains two-dimentional floor plan to be identified first, and identifies mould based on preset wall
Type identifies the two dimension floor plan, so that preset wall identification model exports the wall information in the two dimension floor plan,
The corresponding three-dimensional house type illustraton of model of two-dimentional floor plan is finally generated according to wall information, and exports three-dimensional house type illustraton of model.This hair
The floor plan intelligent identification Method of bright proposition carries out wall identification by preset wall identification model, and identification wall method is new
Grain husk, recognition speed is fast and accuracy rate is high, more facilitates the building of three-dimensional floor plan.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or the system that include a series of elements not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or system institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, method of element, article or system.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior art
The part contributed out can be embodied in the form of software products, which is stored in one as described above
In storage medium (such as ROM/RAM, magnetic disk, CD), including some instructions are used so that terminal device (it can be mobile phone,
Computer, server, air conditioner or network equipment etc.) execute method described in each embodiment of the present invention.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of floor plan intelligent identification Method, which is characterized in that the floor plan intelligent identification Method the following steps are included:
Obtain two-dimentional floor plan to be identified;
The two-dimentional floor plan is identified based on preset wall identification model, exports the wall in the two-dimentional floor plan
Information;
The corresponding three-dimensional house type illustraton of model of the two-dimentional floor plan is generated based on the wall information, and exports the three-dimensional house type
Illustraton of model.
2. floor plan intelligent identification Method as described in claim 1, which is characterized in that described to obtain two-dimentional house type to be identified
The step of figure includes:
When receiving the wall identification request based on two-dimentional floor plan, the corresponding two-dimentional house type of the wall identification request is obtained
Figure.
3. floor plan intelligent identification Method as described in claim 1, which is characterized in that described to identify mould based on preset wall
Before the step of type identifies the two-dimentional floor plan, exports the wall information in the two-dimentional floor plan, further includes:
Obtain two-dimentional floor plan sample to be trained, and based on the two-dimentional floor plan sample to preset wall identification model into
Row off-line training;
The preset wall identification model after saving off-line training.
4. floor plan intelligent identification Method as claimed in claim 3, which is characterized in that described to obtain two-dimentional house type to be trained
Pattern sheet, and the step of carrying out off-line training to preset wall identification model based on the two-dimentional floor plan sample includes:
Two-dimentional floor plan sample to be trained is obtained, and the two-dimentional floor plan sample is pre-processed, wherein is described wait instruct
Experienced two-dimentional floor plan sample has wall markup information;
The pretreated two-dimentional floor plan sample with wall markup information is input to preset convolutional neural networks
In, so that the preset convolutional neural networks export the characteristic pattern of the two-dimentional floor plan sample;
Case-based Reasoning partitioning algorithm detects the wall target in the characteristic pattern, when determining that current detection reaches convergence item
When part, confirm that the preset wall identification model off-line training is completed.
5. floor plan intelligent identification Method as claimed in claim 4, which is characterized in that the condition of convergence includes training time
Number, the training time, training loss function.
6. floor plan intelligent identification Method as described in claim 1, which is characterized in that described to identify mould based on preset wall
The step of type identifies the two-dimentional floor plan, exports the wall information in the two-dimentional floor plan is with described based on described
Wall information generates the corresponding three-dimensional house type illustraton of model of the two-dimentional floor plan, and the step of exporting the three-dimensional house type illustraton of model
Between, further includes:
Preset floor plan component recognition model is obtained, and based on the floor plan component recognition model to the two-dimentional floor plan
In component identified, determine the component position information in the two-dimentional floor plan;
The wall information is modified based on the component position information.
7. floor plan intelligent identification Method as claimed in claim 6, which is characterized in that described to be generated based on the wall information
The corresponding three-dimensional house type illustraton of model of the two dimension floor plan, and the step of exporting the three-dimensional house type illustraton of model includes:
Based on preset three-dimensional wall the build tool, the corresponding three-dimensional family of the two-dimentional floor plan is generated according to the wall information
Type illustraton of model, and export the three-dimensional house type illustraton of model.
8. a kind of floor plan intelligent identification device, which is characterized in that the floor plan intelligent identification device includes:
Floor plan obtains module, for obtaining two-dimentional floor plan to be identified;
Wall identification module, for being identified based on preset wall identification model to the two-dimentional floor plan, described in output
Wall information in two-dimentional floor plan;
Model construction module, for generating the corresponding three-dimensional house type illustraton of model of the two-dimentional floor plan based on the wall information,
And export the three-dimensional house type illustraton of model.
9. a kind of floor plan intelligent identification equipment, which is characterized in that the floor plan intelligent identification equipment includes: memory, place
Reason device and the floor plan intelligent recognition program that is stored on the memory and can run on the processor, the floor plan
The floor plan intelligent recognition as described in any one of claims 1 to 7 is realized when intelligent recognition program is executed by the processor
The step of method.
10. a kind of computer readable storage medium, which is characterized in that be stored with floor plan on the computer readable storage medium
Intelligent recognition program is realized when the floor plan intelligent recognition program is executed by processor such as any one of claims 1 to 7 institute
The step of floor plan intelligent identification Method stated.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910205773.6A CN110096949A (en) | 2019-03-16 | 2019-03-16 | Floor plan intelligent identification Method, device, equipment and computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910205773.6A CN110096949A (en) | 2019-03-16 | 2019-03-16 | Floor plan intelligent identification Method, device, equipment and computer readable storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110096949A true CN110096949A (en) | 2019-08-06 |
Family
ID=67443425
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910205773.6A Pending CN110096949A (en) | 2019-03-16 | 2019-03-16 | Floor plan intelligent identification Method, device, equipment and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110096949A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110634100A (en) * | 2019-08-07 | 2019-12-31 | 贝壳技术有限公司 | Household type graph generation method and device, electronic equipment and storage medium |
CN110781826A (en) * | 2019-10-25 | 2020-02-11 | 珠海格力电器股份有限公司 | Three-dimensional stereogram construction method and system, storage medium and processor |
CN110826627A (en) * | 2019-11-06 | 2020-02-21 | 广东三维家信息科技有限公司 | Image similarity measuring method and device and electronic equipment |
CN111310755A (en) * | 2020-01-13 | 2020-06-19 | 江苏艾佳家居用品有限公司 | Household type graph identification method and device based on target detection and computer equipment |
CN111612890A (en) * | 2020-05-21 | 2020-09-01 | 广东三维家信息科技有限公司 | Method and device for automatically generating three-dimensional model from two-dimensional house type diagram and electronic equipment |
CN113591929A (en) * | 2021-07-05 | 2021-11-02 | 华南师范大学 | Family pattern recognition method combining direction sensing kernel cluster |
CN114973297A (en) * | 2022-06-17 | 2022-08-30 | 广州市圆方计算机软件工程有限公司 | Wall area identification method, system, equipment and medium for planar house type graph |
CN116137066A (en) * | 2023-03-17 | 2023-05-19 | 东易日盛家居装饰集团股份有限公司 | Machine learning-based intelligent identification method and device for house type map functional area |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103971098A (en) * | 2014-05-19 | 2014-08-06 | 北京明兰网络科技有限公司 | Method for recognizing wall in house type image and method for automatically correcting length ratio of house type image |
CN104821011A (en) * | 2015-05-20 | 2015-08-05 | 郭小虎 | Method of generating 3D house type model by 2D house type model based on camera shooting |
CN107798725A (en) * | 2017-09-04 | 2018-03-13 | 华南理工大学 | The identification of two-dimentional house types and three-dimensional rendering method based on Android |
CN108717726A (en) * | 2018-05-11 | 2018-10-30 | 北京家印互动科技有限公司 | Three-dimensional house type model generating method and device |
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 |
-
2019
- 2019-03-16 CN CN201910205773.6A patent/CN110096949A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103971098A (en) * | 2014-05-19 | 2014-08-06 | 北京明兰网络科技有限公司 | Method for recognizing wall in house type image and method for automatically correcting length ratio of house type image |
CN104821011A (en) * | 2015-05-20 | 2015-08-05 | 郭小虎 | Method of generating 3D house type model by 2D house type model based on camera shooting |
CN107798725A (en) * | 2017-09-04 | 2018-03-13 | 华南理工大学 | The identification of two-dimentional house types and three-dimensional rendering method based on Android |
CN108717726A (en) * | 2018-05-11 | 2018-10-30 | 北京家印互动科技有限公司 | Three-dimensional house type model generating method and device |
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 |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110634100A (en) * | 2019-08-07 | 2019-12-31 | 贝壳技术有限公司 | Household type graph generation method and device, electronic equipment and storage medium |
CN110781826A (en) * | 2019-10-25 | 2020-02-11 | 珠海格力电器股份有限公司 | Three-dimensional stereogram construction method and system, storage medium and processor |
CN110826627A (en) * | 2019-11-06 | 2020-02-21 | 广东三维家信息科技有限公司 | Image similarity measuring method and device and electronic equipment |
CN111310755A (en) * | 2020-01-13 | 2020-06-19 | 江苏艾佳家居用品有限公司 | Household type graph identification method and device based on target detection and computer equipment |
CN111310755B (en) * | 2020-01-13 | 2022-07-08 | 江苏艾佳家居用品有限公司 | Household type graph identification method and device based on target detection and computer equipment |
CN111612890A (en) * | 2020-05-21 | 2020-09-01 | 广东三维家信息科技有限公司 | Method and device for automatically generating three-dimensional model from two-dimensional house type diagram and electronic equipment |
CN111612890B (en) * | 2020-05-21 | 2023-08-22 | 广东三维家信息科技有限公司 | Method and device for automatically generating three-dimensional model by two-dimensional house type graph and electronic equipment |
CN113591929A (en) * | 2021-07-05 | 2021-11-02 | 华南师范大学 | Family pattern recognition method combining direction sensing kernel cluster |
CN113591929B (en) * | 2021-07-05 | 2023-07-25 | 华南师范大学 | House pattern recognition method combining direction perception kernel cluster |
CN114973297A (en) * | 2022-06-17 | 2022-08-30 | 广州市圆方计算机软件工程有限公司 | Wall area identification method, system, equipment and medium for planar house type graph |
CN116137066A (en) * | 2023-03-17 | 2023-05-19 | 东易日盛家居装饰集团股份有限公司 | Machine learning-based intelligent identification method and device for house type map functional area |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110096949A (en) | Floor plan intelligent identification Method, device, equipment and computer readable storage medium | |
CN109636905B (en) | Environment semantic mapping method based on deep convolutional neural network | |
US11798132B2 (en) | Image inpainting method and apparatus, computer device, and storage medium | |
US10679046B1 (en) | Machine learning systems and methods of estimating body shape from images | |
EP3644277A1 (en) | Image processing system, image processing method, and program | |
CN108734120A (en) | Mark method, apparatus, equipment and the computer readable storage medium of image | |
CN104978762A (en) | Three-dimensional clothing model generating method and system | |
CN113496507A (en) | Human body three-dimensional model reconstruction method | |
CN106875203A (en) | A kind of method and device of the style information for determining commodity picture | |
CN108763813A (en) | It is a kind of to identify the method and apparatus for copying wall in figure based on deep learning | |
CN109800653A (en) | A kind of characteristics of human body's parameter extracting method and system based on image analysis | |
CN107506792B (en) | Semi-supervised salient object detection method | |
CN113139543A (en) | Training method of target object detection model, target object detection method and device | |
CN109920018A (en) | Black-and-white photograph color recovery method, device and storage medium neural network based | |
CN114332977A (en) | Key point detection method and device, electronic equipment and storage medium | |
CN114723888A (en) | Three-dimensional hair model generation method, device, equipment, storage medium and product | |
CN116229066A (en) | Portrait segmentation model training method and related device | |
CN114820292A (en) | Image synthesis method, device, equipment and storage medium | |
CN110059721A (en) | Floor plan area recognizing method, device, equipment and computer readable storage medium | |
CN111079807A (en) | Ground object classification method and device | |
CN113658195A (en) | Image segmentation method and device and electronic equipment | |
CN112750155A (en) | Panoramic depth estimation method based on convolutional neural network | |
CN117252988A (en) | Image data processing method, device and computer readable storage medium | |
CN117173595A (en) | Unmanned aerial vehicle aerial image target detection method based on improved YOLOv7 | |
CN111199248A (en) | Clothing attribute detection method based on deep learning target detection algorithm |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190806 |