CN108592789A - A kind of steel construction factory pre-assembly method based on BIM and machine vision technique - Google Patents
A kind of steel construction factory pre-assembly method based on BIM and machine vision technique Download PDFInfo
- Publication number
- CN108592789A CN108592789A CN201810272596.9A CN201810272596A CN108592789A CN 108592789 A CN108592789 A CN 108592789A CN 201810272596 A CN201810272596 A CN 201810272596A CN 108592789 A CN108592789 A CN 108592789A
- Authority
- CN
- China
- Prior art keywords
- bim
- camera
- point
- coordinate
- machine vision
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
Abstract
The invention belongs to steel structure technology fields, more particularly to a kind of steel construction factory pre-assembly method based on BIM and machine vision technique, installation horizontal photographing camera C1 and vertical photographing camera C2 forms outdoor scene and takes pictures submodule on processing factory fixed position or gate-type purlin vehicle；Factory's BIM models are created, the model space are divided by three dimensions small cubes one by one according to camera pixel product, its clear centre of form coordinate values of each cube；The calculating of real space coordinate and virtual BIM models spatial coordinates is carried out by transition matrix, i.e., is that shoot on location photograph image target point assigns predefined actual spatial coordinates by BIM model scenes；According to spatial position optimization algorithm Fitting Analysis, single member dimensional measurement and digitlization preassembling are realized.The present invention comprehensively utilizes Machine Vision Detection, BIM information models, computer intelligence optimization algorithm, increases substantially factory's steel member size detection and digitizes the precision and efficiency of preassembling.
Description
Technical field
The invention belongs to steel structure technology fields, and in particular to a kind of steel construction work based on BIM and machine vision technique
Factory's pre-assembly method.
Background technology
In existing, polymorphic structure is more and more, and the accurate measurement of scantling is increasingly difficult to, it is necessary to the quality of stiffener
Control and take preassembling just to can guarantee field erected smooth realization.
Vision-based detection is exactly to replace human eye with machine to measure and judge.Mechanical vision inspection technology is built upon calculating
An emerging detection technique on the basis of machine theories of vision has fast precision height, speed, non-contact, flexible, automation and intelligence
Many advantages, such as energyization is horizontal high, the requirement for well adapting to and meeting modern manufacturing industry to detection technique.
BIM is the abbreviation of Building Information Modeling, i.e. Building Information Model, is that will build knot
The geological information and non-geometry information of structure merge in three-dimension visualization digital model, for building Life cycle application.
By information knots such as the architecture information position dimensions for including of class eye recognition ability and BIM possessed by computer vision
It closes, the positioning of virtual reality space coordinate points may be implemented.It is regarded based on BIM and machine in consideration of it, present applicant has proposed one kind
The steel construction factory number pre-assembly method of feel technology, provide implementation for steel structure member size detection and digitlization preassembling
Path and integrated system solution.
Invention content
The invention discloses a kind of steel construction factory pre-assembly method based on BIM and machine vision technique, by computer
The information such as the architecture information position dimension that class eye recognition ability and BIM possessed by vision include combine, and realize virtual reality
The accurate positionin of 3 d space coordinate point carries out steel member size detection and digitlization preassembling, passes through whole system and method
Improve the precision and efficiency of steel construction quality control.
In order to achieve the above objectives, specific technical solution of the present invention is as follows：
The first step：Two high-precision cameras are disposed on factory structure fixed position or gate-type purlin vehicle, wherein one is taken pictures for level
Camera C1, another is vertical photographing camera C2, and two cameras composition outdoor scenes are taken pictures submodule.Outdoor scene is taken pictures horizontal in submodule
Photographing camera C1 and the visual angles vertical photographing camera C2 are mutually perpendicular to, and processing factory fixed position are may be mounted at, such as roof system steel truss
On brace steel post；It can also be mounted on the vehicle of gate-type purlin, may be implemented to be taken pictures and surveyed in assembly line any position in this way
Amount operation；
Second step：Factory's BIM models are created, the horizontal photographing camera C1 of accurate calibration and vertical phase of taking pictures are needed in BIM models
The position of machine C2 forms virtual camera visual angle；
Third walks：The meter of real space coordinate and virtual BIM models spatial coordinates is carried out by camera positioning and demarcating submodule
It calculates, as target point assigns predefined actual spatial coordinates.It is reference with BIM model local Coordinate Systems, is shone in outdoor scene camera
Coordinate of ground point p in piece（X, y, z）Transition matrix is created between actual coordinate P (X, Y, Z)：P (X, Y, Z)=△ × p（X, y,
z）, wherein △ includes camera parameter matrix, distortion matrix, and spatial parameter matrix can read outdoor scene camera by matrix conversion
The corresponding true three-dimension coordinate of any point in photo；
4th step：Steel member to be measured is placed on moulding bed and is taken pictures by camera C1, C2；
5th step：Structure is extracted by photo captured by two high-precision cameras in the conversion of component coordinate and data processing submodule
The true three-dimension coordinate value of part measurement point determines measurement point by photo captured by horizontal photographing camera C1（X、Y）Coordinate passes through
Vertical photographing camera C2 determines measurement point（X、Z）Coordinate obtains the 3 d space coordinate value for being measured point according to common coordinate value X
P (X, Y, Z)；
6th step：The feature point for calibration in single member BIM models, by corresponding spy in the characteristic point taken pictures and BIM models
After sign point fitting, the variance analysis of the arbitrary measurement point of single member can be carried out；
7th step：In digitizing preassembling submodule, the practical measurement control point data of single member is fitted to successively pre-splicing
It fills on the corresponding control point of unit B IM models, is digitized preassembling separate-blas estimation；
8th step：Single member testing result, digitlization preassembling result are direct by system in display and mobile terminal device
Examining report is generated, factory is instructed to carry out the modification for not conforming to lattice piece.
The present invention is applied in steel member size detection and digitlization preassembling has notable benefit, specific as follows:
1, detection efficiency is high：It can be obtained the 3 d space coordinate of detected component arbitrary point upon taking a picture, change traditional whole station
Instrument is got ready or gradually the handheld three-dimensional laser scanning of prevalence carries out the low problem of full component Scanning Detction efficiency in recent years, substantially
Shorten component detection and preassembling duration.
2, accuracy of detection is high：The precision of measurement point is related to camera pixel, gradually with industrial equipment field camera precision
It improves, accuracy of detection can further increase, and accuracy of detection is can be controlled within 1mm.
3, equipment cost is low：This system vision collecting module capital equipment is two sets of high-precision cameras, relative to hand-held
The equipment such as 3 D laser scanning, this system software and hardware is cheap, and lower cost is conducive to the popularization of the technology of the present invention system.
4, intelligent measurement：Detection means can be fitted according to space coordinate automatic optimal, by the algorithm of artificial intelligence, be improved
Detection efficiency and precision.
5, simple and safe operation is constructed convenient for assembly line.It such as secures the equipment on the vehicle of gate-type purlin, purlin vehicle moves in workshop
To any position, it can carry out component and take pictures and size detection, be highly suitable for processing factory's assembly line construction.
Description of the drawings
Fig. 1 is the example schematic of the present invention；
Fig. 2 is the system flow chart of the present invention；
Fig. 3 is the detection operational flowchart of the present invention.
Specific implementation mode
The present invention is further described with case study on implementation below in conjunction with the accompanying drawings.
As shown in Figure 1, a kind of steel construction factory pre-assembly method based on BIM and machine vision technique, matches in processing factory
Set corresponding vision collecting module：Two high-precision cameras are disposed on factory structure fixed position or gate-type purlin vehicle, wherein one
For horizontal photographing camera C1, another is vertical photographing camera C2, and two cameras composition outdoor scenes are taken pictures submodule.
As shown in Fig. 2, a kind of steel construction factory pre-assembly method based on BIM and machine vision technique, it includes as follows
Step：
Step 1：Debugging two cameras composition outdoor scenes are taken pictures submodule；
Step 2：Factory's BIM models are created, the horizontal photographing camera C1 of accurate calibration and vertical photographing camera are needed in BIM models
The position of C2 forms virtual camera visual angle；
Step 3：The meter of real space coordinate and virtual BIM models spatial coordinates is carried out by camera positioning and demarcating submodule
It calculates, as target point assigns predefined actual spatial coordinates；
Step 4：Steel member to be measured is placed on moulding bed and is taken pictures by camera C1, C2；
Step 5：Structure is extracted by photo captured by two high-precision cameras in the conversion of component coordinate and data processing submodule
The true three-dimension coordinate value of part measurement point；
Step 6：The feature point for calibration in single member BIM models, by corresponding feature in the characteristic point taken pictures and BIM models
After point fitting, the variance analysis of the arbitrary measurement point of single member can be carried out；
Step 7：In digitizing preassembling submodule, the practical measurement control point data of single member is fitted to preassembling successively
On the corresponding control point of unit B IM models, it is digitized preassembling separate-blas estimation；
Step 8：Single member testing result, digitlization preassembling result are direct by system in display and mobile terminal device
Generate examining report.
To sum up, a kind of steel construction factory number pre-assembly method based on BIM and machine vision technique of the invention and it is
System relies on shoot on location photograph image and target point space-location method in BIM model scenes, increases substantially factory's steel member
Size detection and digitlization preassembling precision and efficiency, while have many advantages, such as it is easy to operate, it is of low cost, be very suitable for
Factory lines construction operation environment is transformed traditional architecture industry by industrialization, intelligentized means, there is preferable popularization to answer
With foreground and value.
The above case study on implementation is only used to illustrate the technical scheme of the present invention and unrestricted, and those skilled in the art is to this hair
Other modification or replacements that bright technical solution is done, the core concept without departing from technical solution of the present invention and technology model
It encloses, should all cover in scope of the presently claimed invention.
Claims (6)
1. a kind of steel construction factory pre-assembly method based on BIM and machine vision technique, which is characterized in that detailed process includes
Following steps:
Step 1：Two high-precision cameras are disposed on factory structure fixed position or gate-type purlin vehicle, wherein one is taken pictures phase for level
Machine C1, another is vertical photographing camera C2, and two cameras composition outdoor scenes are taken pictures submodule；
Step 2：Factory's BIM models are created, the horizontal photographing camera C1 of accurate calibration and vertical photographing camera are needed in BIM models
The position of C2 forms virtual camera visual angle；
Step 3：The meter of real space coordinate and virtual BIM models spatial coordinates is carried out by camera positioning and demarcating submodule
It calculates, as target point assigns predefined actual spatial coordinates；
Step 4：Steel member to be measured is placed on moulding bed and is taken pictures by camera C1, C2；
Step 5：Structure is extracted by photo captured by two high-precision cameras in the conversion of component coordinate and data processing submodule
The true three-dimension coordinate value of part measurement point；
Step 6：The feature point for calibration in single member BIM models, by corresponding feature in the characteristic point taken pictures and BIM models
After point fitting, the variance analysis of the arbitrary measurement point of single member is carried out；
Step 7：In digitizing preassembling submodule, the practical measurement control point data of single member is fitted to preassembling successively
On the corresponding control point of unit B IM models, it is digitized preassembling separate-blas estimation；
Step 8：Single member testing result, digitlization preassembling result are direct by system in display and mobile terminal device
Generate examining report.
2. a kind of steel construction factory pre-assembly method based on BIM and machine vision technique according to claim 1, special
Sign is：Take pictures horizontal photographing camera C1 and the visual angles vertical photographing camera C2 in submodule of outdoor scene is mutually perpendicular to, and is mounted on processing
Factory fixed position：On roof system steel truss and brace steel post；On the vehicle of gate-type purlin.
3. a kind of steel construction factory pre-assembly method based on BIM and machine vision technique according to claim 1, special
Sign is：In factory's BIM models, the entire BIM model spaces are divided by " camera C1 pixels × camera C2 according to actual camera pixel
A three dimensions small cubes of pixel ", its clear centre of form coordinate values of each cube, i.e. actual coordinate P (X, Y, Z).
4. a kind of steel construction factory pre-assembly method based on BIM and machine vision technique according to claim 3, special
Sign is：Coordinate of ground point p in outdoor scene camera photos（X, y, z）Transition matrix is created between actual coordinate P (X, Y, Z)：P
(X, Y, Z)=△ × p（X, y, z）, wherein △ includes camera parameter matrix, distortion matrix, and spatial parameter matrix is turned by matrix
The corresponding true three-dimension coordinate of any point in outdoor scene camera photos can be read by changing.
5. a kind of steel construction factory pre-assembly method based on BIM and machine vision technique according to claim 1, special
Sign is：Measurement point is determined by photo captured by horizontal photographing camera C1 in the conversion of component coordinate and data processing submodule
（X, Y）Coordinate determines measurement point by vertical photographing camera C2（X, Z）Coordinate obtains being measured point according to common coordinate value X
3 d space coordinate value P (X, Y, Z).
6. a kind of steel construction factory pre-assembly method based on BIM and machine vision technique according to claim 1, special
Sign is：It needs to demarcate three or more features respectively in BIM models and shooting photo in single member size detection submodule
Then characteristic point is fitted by point according to spatial position optimization algorithm, the deviation point of arbitrary measurement point can be carried out after the completion of fitting
Analysis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810272596.9A CN108592789A (en) | 2018-03-29 | 2018-03-29 | A kind of steel construction factory pre-assembly method based on BIM and machine vision technique |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810272596.9A CN108592789A (en) | 2018-03-29 | 2018-03-29 | A kind of steel construction factory pre-assembly method based on BIM and machine vision technique |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108592789A true CN108592789A (en) | 2018-09-28 |
Family
ID=63623856
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810272596.9A Pending CN108592789A (en) | 2018-03-29 | 2018-03-29 | A kind of steel construction factory pre-assembly method based on BIM and machine vision technique |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108592789A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111238389A (en) * | 2020-01-20 | 2020-06-05 | 上海海事大学 | Box body detection system and method based on binocular vision |
| CN111472486A (en) * | 2020-04-23 | 2020-07-31 | 浙江精工钢结构集团有限公司 | Steel construction and curtain intelligent monitoring maintenance system based on BIM |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1878318A (en) * | 2006-05-17 | 2006-12-13 | 浙江大学 | Three-dimensional small-sized scene rebuilding method based on dual-camera and its device |
| CN101032830A (en) * | 2006-03-08 | 2007-09-12 | 上海市基础工程公司 | Technique of producing steel-concrete connecting beam |
| CN101261738A (en) * | 2008-03-28 | 2008-09-10 | 北京航空航天大学 | A camera marking method based on double 1-dimension drone |
| CN101413348A (en) * | 2008-11-28 | 2009-04-22 | 中冶京唐建设有限公司 | Steel structure three-dimensional scanning observe and control method |
| CN101590725A (en) * | 2008-05-30 | 2009-12-02 | 索尼株式会社 | Printing equipment, printing process and program |
| CN101900875A (en) * | 2010-06-04 | 2010-12-01 | 南开大学 | High-magnification three-dimensional imaging microscope based on double-light source off-axis illumination and imaging method |
| CN104729534A (en) * | 2015-03-12 | 2015-06-24 | 北京空间飞行器总体设计部 | Monocular visual error measurement system for cooperative target and error limit quantification method |
| WO2015153393A1 (en) * | 2014-04-02 | 2015-10-08 | Faro Technologies, Inc. | Registering of a scene disintegrating into clusters with visualized clusters |
| CN106354950A (en) * | 2016-08-31 | 2017-01-25 | 华中科技大学 | Steel structure pre-splicing method and system |
-
2018
- 2018-03-29 CN CN201810272596.9A patent/CN108592789A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101032830A (en) * | 2006-03-08 | 2007-09-12 | 上海市基础工程公司 | Technique of producing steel-concrete connecting beam |
| CN1878318A (en) * | 2006-05-17 | 2006-12-13 | 浙江大学 | Three-dimensional small-sized scene rebuilding method based on dual-camera and its device |
| CN101261738A (en) * | 2008-03-28 | 2008-09-10 | 北京航空航天大学 | A camera marking method based on double 1-dimension drone |
| CN101590725A (en) * | 2008-05-30 | 2009-12-02 | 索尼株式会社 | Printing equipment, printing process and program |
| CN101413348A (en) * | 2008-11-28 | 2009-04-22 | 中冶京唐建设有限公司 | Steel structure three-dimensional scanning observe and control method |
| CN101900875A (en) * | 2010-06-04 | 2010-12-01 | 南开大学 | High-magnification three-dimensional imaging microscope based on double-light source off-axis illumination and imaging method |
| WO2015153393A1 (en) * | 2014-04-02 | 2015-10-08 | Faro Technologies, Inc. | Registering of a scene disintegrating into clusters with visualized clusters |
| CN104729534A (en) * | 2015-03-12 | 2015-06-24 | 北京空间飞行器总体设计部 | Monocular visual error measurement system for cooperative target and error limit quantification method |
| CN106354950A (en) * | 2016-08-31 | 2017-01-25 | 华中科技大学 | Steel structure pre-splicing method and system |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111238389A (en) * | 2020-01-20 | 2020-06-05 | 上海海事大学 | Box body detection system and method based on binocular vision |
| CN111472486A (en) * | 2020-04-23 | 2020-07-31 | 浙江精工钢结构集团有限公司 | Steel construction and curtain intelligent monitoring maintenance system based on BIM |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105758426B (en) | The combined calibrating method of the multisensor of mobile robot | |
| CN110146099B (en) | Synchronous positioning and map construction method based on deep learning | |
| CN104330074B (en) | Intelligent surveying and mapping platform and realizing method thereof | |
| US6917702B2 (en) | Calibration of multiple cameras for a turntable-based 3D scanner | |
| CN112894832A (en) | Three-dimensional modeling method, three-dimensional modeling device, electronic equipment and storage medium | |
| CN107255476A (en) | A kind of indoor orientation method and device based on inertial data and visual signature | |
| Li | Camera calibration of a head-eye system for active vision | |
| CN106971408B (en) | A kind of camera marking method based on space-time conversion thought | |
| CN110580723A (en) | method for carrying out accurate positioning by utilizing deep learning and computer vision | |
| CN110400315A (en) | A kind of defect inspection method, apparatus and system | |
| JP2013257349A (en) | Camera calibration result verification device and method | |
| CN108227929A (en) | Augmented reality setting-out system and implementation method based on BIM technology | |
| CN108958256A (en) | A kind of vision navigation method of mobile robot based on SSD object detection model | |
| CN105023294A (en) | Fixed point movement augmented reality method combining sensors and Unity3D | |
| CN109903382A (en) | The fusion method and device of point cloud data | |
| CN110044266B (en) | Photogrammetry system based on speckle projection | |
| Mi et al. | A vision-based displacement measurement system for foundation pit | |
| CN115187612A (en) | Plane area measuring method, device and system based on machine vision | |
| CN114792345A (en) | Calibration method based on monocular structured light system | |
| CN108592789A (en) | A kind of steel construction factory pre-assembly method based on BIM and machine vision technique | |
| CN114663473A (en) | Personnel target positioning and tracking method and system based on multi-view information fusion | |
| CN114266823A (en) | Monocular SLAM method combining SuperPoint network characteristic extraction | |
| CN116228870B (en) | Mapping method and system based on two-dimensional code SLAM precision control | |
| CN112348941A (en) | Real-time fusion method and device based on point cloud and image data | |
| CN107274449B (en) | Space positioning system and method for object by optical photo |
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 |