CN109934911A - Mobile terminal high-precision oblique photograph three-dimensional modeling method based on OpenGL - Google Patents
Mobile terminal high-precision oblique photograph three-dimensional modeling method based on OpenGL Download PDFInfo
- Publication number
- CN109934911A CN109934911A CN201910196627.1A CN201910196627A CN109934911A CN 109934911 A CN109934911 A CN 109934911A CN 201910196627 A CN201910196627 A CN 201910196627A CN 109934911 A CN109934911 A CN 109934911A
- Authority
- CN
- China
- Prior art keywords
- model
- oblique photograph
- fragment
- lod
- 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
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses a kind of mobile terminal high-precision oblique photograph three-dimensional modeling method based on OpenGL, belongs to Surveying Engineering and computer combination field；This method is first with unmanned plane oblique photograph technology, quick obtaining region to be measured has the image data of spatial positional information, then to image by carrying out Geometry rectification, oblique photograph model batch processing is carried out to image data after correction to generate, then LOD classification is carried out to the model data of generation, data compression is carried out to the model data of generation later, merges adjacent node, is finally used model based on OpenGL technology in mobile terminal dynamically load；The present invention is able to achieve mobile terminal high-volume load high-precision oblique photograph model, it can be achieved that on mobile terminal load browsing oblique photograph measurement three-dimensional digital city, and has certain social benefit and economic interests by combining oblique photograph technology with mobile technology.
Description
Technical field:
The present invention relates to mapping and digial earth Rendering field, specifically a kind of mobile terminals based on OpenGL
High-precision oblique photograph three-dimensional modeling method.
Background technique:
Java programming language possesses complete ecological chain, has safe and efficient, cross-platform operation characteristic, in mobile application field
Using extensive.The OpenGL of advantage using Java in mobile application field, the encapsulation of the java3d used in eclipse is tied up
Fixed SFML Kucheng is building three-dimensional digital city model, digital earth, the hot spot of digital city developer.With inclination
The development of photogrammetric technology, present digital city have begun similar in this true to nature and true atural object of large-scale use
The higher model of the degree of automation.Compared with the Digital Campus of traditional-handwork modeling, digital city, it is based on oblique photograph model structure
Build Digital Campus, city can save a large amount of cost of labor, oblique photograph whole process no manual intervention, effect is truer, and precision can
To reach mapping rank, data presentation mode is total factor presentation, has the characteristics that high-efficient, the period is short.Oblique photograph model is answered
Can solve the problem within the scope of industry field for mobile field, as urban planning, the detection of city side shape, major project engineering,
Emergency disaster relief, new rural village and town development, smart city etc. can play great economic results in society.But because of oblique photograph
Generation model data amount is huge, and the rendering task of model is heavier.Therefore the model of high-precision oblique photograph at this stage can not still advise greatly
Mould is loaded onto mobile device.The solution in industry is that model is carried out physical extent cutting at present, by a monolith model point
Muti-piece submodel is cut, load rendering then is carried out to the model grouping of well cutting, physical extent cutting is carried out to model and is not only broken
The integrality of bad initial data also destroys original model texture, forms numerous broken floating materials, still generates to the precision of model
It influences, causes model to match with corresponding satellite image inaccurate.And there is no first carry out base to raw video in current industry
Correction and noise-filtering in central pixel offset only carry out geometric correction in Mass production prototype software, joint is put down
Difference, this is not enough to reach mapping precision.Therefore how the oblique photograph model true to nature to high-volume, high-precision, texture carries out wash with watercolours
Dye/load strategy/method improves, and is allowed to be able to satisfy and is added by mobile device to extensive high-precision oblique photograph model
Carry, meet in mobile field to oblique photograph model browsing need become a urgent need to solve the problem.
Summary of the invention:
The purpose of the present invention is overcoming the shortcomings of above-mentioned prior art, and provide it is a kind of based on OpenGL mobile terminal high-precision incline
Tiltedly photography three-dimensional modeling method；Mobile device can not be loaded on a large scale by mainly solving existing high-precision oblique photograph model
Problem.
The technical scheme is that the mobile terminal high-precision oblique photograph three-dimensional modeling method based on OpenGL, special
Different place is, comprising the following steps:
A is based on shadow according to formula 1-8 to the 60% Duplication image data in area to be measured acquired in five camera of UAV flight
Inconocenter picture point is corrected, and carries out the filtering processing of Kalman filtering algorithm noise to image after correction, obtains base image；
Formula 1:A=mx cos t
Formula 2:B=my (k cos t-sin t)
Formula 3:D=mx sin t
Formula 4:E=- 1 my (k sin t+cos t)
The translational movement in the direction formula 5:C=x
The translational movement in the direction formula 6:F=y
Formula 7:x1=Ax+By+C
Formula 8:y1=Dx+Ey+F
Wherein, scale factor, B and the D that A is X are the negative value for rotating the scale factor that item, C and F are translation item, E is y, and mx is
Ratio variation on the direction x, my are the ratio variation on the direction y, and t is depended on yaw angle, surveyed counterclockwise using x-axis by starting point
Value, since aircraft axes are different from the definition of abovementioned mathematical plane coordinate system, need to be solved for two-dimensional surface just with calculated by coordinate
True yaw angle, t here are by the yaw angle after calculated by coordinate；K is shear factor=tan (u) along x-axis, and u is inclined
The difference at boat angle and y-axis, tilt angle at this time is measured relative to y-axis；
B uses image processing software MATLAB combination Kalman filtering algorithm, filters out influence of the noise to picture quality；Kalman
The step of filtering removal noise factor are as follows:
1) noise state equation and observational equation are first determined；
2) elements of exterior orientation of image data is imported；
3) input parameters: latitude, longitude, process-noise variance, observation noise variance, filter vector initial value, it is any when
Carve observation state variance；
4) coefficient for finding out predictive equation is realized finally by difference equation and is filtered；
Base image is imported the end PC automation modeling software Smart 3D by c, by further to the base image after correction
Geometric correction, simultaneous adjustment process flow are generated the ultra high density point cloud based on real image, and are generated with this based on true shadow
As the high-resolution outdoor scene threedimensional model of texture；
D is to high-precision oblique photograph model is made after completing image correction, by the oblique photograph model data of generation according to LOD
Rank classification, the careful degree of texture that can be used to indicate to generate oblique photograph model using LOD classification, LOD are classified model
It is divided into pyramid rank, every level-one corresponds to the accuracy class of oblique photograph model；
E carries out fragment according to rank is divided to the oblique photograph threedimensional model after classification, each fragment model corresponds to one
Root node；Possess numerous child nodes under each fragment model；Each fragment model is divided into the lesser inclination of several block amounts to take the photograph
Shadow model corresponds to a child node under each sectional pattern；To according to the good fragment number of LOD grade classification, fragment model root section
Point and corresponding latitude and longitude coordinates, sectional pattern child node and latitude and longitude coordinates save；Load grouping is carried out to the model of fragment,
Steps are as follows for grouping:
1) entire survey section model is determined first, and model grade is divided according to LOD technology；
2) the model fragment number under each grade is then determined, the fragment number under each series is determined according to LOD divided rank
It is fixed；
3) it then determines the regional scope that respectively shows of fragment pattern number under the grade, determines that each fragment model is corresponding
Heart node longitude and latitude corresponding with its；
4) child node of several pieces of submodels and corresponding longitude and latitude under fragment model are finally determined；
F carries out data compression to the high-precision oblique photograph model for completing model classification fragment, reduces the data volume of model, should
The nodal information for being interpreted as current oblique photograph model is tree, and entire hierarchy model possesses a central node, right
Each fragment model answered has root node, and several block models possess corresponding child node under each fragment model, by modifying entire mould
Type central node merges adjacent node, vacuates node, texture compression to realize the fast browsing load of mobile terminal, data compression
Method and step are as follows:
1) entire oblique photograph model center point is modified；
2) merge adjacent node；
3) node is vacuated；
4) texture compression；
G carries out texture building and rendering using high-precision oblique photograph model of the OpenGL open graphic library interface to generation, and
The original data format of oblique photograph is converted, realizes that mobile terminal browses the load of high-precision oblique photograph model, it is right
Model carries out the method and step of building rendering are as follows:
1) application programming interface provided with open graphic library, will be good based on LOD grade classification by routine interface
Oblique photograph model carries out building rendering in OpenGL shape library；
2) data conversion is carried out to the high-precision oblique photograph model of high-volume that rendering is built, by oblique photograph original data lattice
Formula (OBJ, OSGB, DAE), which is converted to, is converted to mobile device format by OpenGL shape library binding SFML, and rendering is built
Data are uploaded to server or by data configuration to engineering local, are completed by mobile device application program to index file
It reads, completes the load browsing of mobile device；
3) then to the elements of exterior orientation of UAV flight's video camera, the central point including entire model, the attitude angle pair of aircraft
Degree is answered, the attitude angle of aircraft includes pitch angle, course angle, roll angle, calculates determination within the scope of camera lens and needs to load
The oblique photograph model that is classified based on LOD of model；
4) when mobile zoom operations occur for the oblique photograph model of mobile device end, first judge that grade locating for current model is
No is maximum model grade and model loading capacity is more than 60%, is then calculated by model grade to be loaded, if on mobile terminal
What is carried out is amplification oblique photograph model manipulation, then loads new Grade Model, and dynamic deletes the model loaded before, if
What mobile terminal carried out is to reduce oblique photograph model manipulation, then does not delete the oblique photograph model loaded, and final realize is moved
State loads high-volume high-precision oblique photograph model.
Further, the step f carries out data compression and merge node tool is three-dimensional process software；And by data into
The information of row texture compression and merge node is recorded in config.scp configuration file.
Further, it is formulated in the step d by import information and is based on LOD hierarchical policy, meet following condition:
1) the hierarchical model quantity under corresponding LOD grade is no more than 15；
2) each hierarchical model is adjacent, forms complete oblique photograph model；
3) collapsible mould type quantity corresponding to the model of difference LOD grade has significant difference；
4) the minimum model fragment minimum number of LOD grade classification.
Compared with the prior art mobile terminal high-precision oblique photograph three-dimensional modeling method based on OpenGL of the invention has
There are substantive distinguishing features and marked improvement following prominent:
1, because each of region to be measured image data all has elements of exterior orientation, elements of exterior orientation and affine transform algorithm are based on
Combine, central pixel can be solved relative to practical center picture point offset, come eliminate aspect and relief because
Element influences, and improves the precision of oblique photograph model, and oblique photograph model accuracy is about equal to three times of engineering orthography resolution ratio,
Model accuracy improves 20-30% compared with tradition directly generates oblique photograph model without image correction；Formula (1-8) algorithm
The difference is that the offset (relative to route track) of calculated image center picture point come to image center picture point into
Row is corrected, and the central pixel of each image corresponds to the central node of each sectional pattern, this nodes records longitude and latitude and
Elevation information, model essence can integrally be improved by carrying out oblique photograph modelling using the data after correcting based on image center picture point
Degree；
2, because base image is there is also noise effect after central point corrects, noise will lead to image and periodical item occurs
Line, spot etc. use kalman filter method to longitude and latitude, the elements of exterior orientation of the image data combination central point after correction
Noise is removed, compared to the processing method of conventional method, is theed improvement is that in conjunction with the filter processing method of elements of exterior orientation
It to base image noise suppression and is filtered out in conjunction with the longitude and latitude of base image central point, attitude angle and flying height, aircraft
Attitude angle includes pitch angle, course angle, roll angle；Traditional remote sensing images filter out noise using high pass or low-pass filtering to realize
The correction of data, but high pass or low-pass filtering realization noise are carried out for base image data required for oblique photograph model
Filtration result is not obvious, and the space bit of filming image has been fully considered using elements of exterior orientation combination Kalman filter
It sets, the factors such as posture, noise can be reduced to greatest extent to the imaging results of base image；Relative to without noise-filtering
Model, the model texture fitting effect after filtering out noise are more preferable；
3, carry out partitioning model method relative to traditional region Slicing Model for Foreign, Slicing Model for Foreign not only breaks up data integrity and causes original
Beginning loss of data, suspended matter easy to form, model accuracy are low；This method is theed improvement is that according to primary model data format
(OBJ, OSGB, DAE) characteristic is divided oblique photograph model according to LOD model grade, and according to corresponding nodal information
Grade fragment piecemeal, saves complete primary model data information, enables model accuracy in the case where not destroying initial data structure
Reach 0.3-0.5CM, meets Surveying and Mapping Industry required precision.
Detailed description of the invention:
Fig. 1 is for image rectification method course diagram of the invention；
Fig. 2 is model scalable compression method flow chart of the invention；
Fig. 3 is in the present invention based on OpenGL dynamically load oblique photograph model flow figure；
Fig. 4 is overview flow chart of the invention.
Specific embodiment:
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, to this hair
It is bright to be further elaborated.It should be appreciated that described herein, the specific embodiments are only for explaining the present invention, is not used to
It is defined in the present invention.
Embodiment 1, referring to Fig. 1,2,3,4；Central pixel correction, filtering removal noise step, the present invention are carried out to image
In Fig. 1 is shown in image rectification method:
60% Duplication image data is obtained by region to be measured using unmanned plane first, is imaged using five camera lens of UAV flight
Machine, since image correction and image need to be carried out to the image data got by aspect, hypsography, noise effect
Transformation；Each accessed at this time image data all has center point coordinate, records unmanned plane when shooting the region
Longitude and latitude；
Using the elements of exterior orientation combination affine transformation formula of each image data, the offset of image center point is calculated, and
Correct central point is calculated according to central point offset；The precision of oblique photograph model is improved, oblique photograph model accuracy is about
Equivalent three times of engineering orthography resolution ratio；The algorithm the difference is that calculated image center picture point offset
(relative to route track) corrects image center picture point, and the central pixel of each image corresponds to each piecemeal mould
The central node of type, this nodes records longitude and latitude and elevation information, using based on image center picture point correct after data into
The production of line tilt photography model can integrally improve model accuracy；
The formula that image calculates the offset of central point is as follows:
A = mx · cos t
B = my · (k · cos t - sin t)
D = mx · sin t
E = -1 · my · (k · sin t + cos t)
The translational movement in the direction C=x
The translational movement in the direction F=y
x1 = Ax + By + C
y1 = Dx + Ey + F
Wherein, scale factor, B and the D that A is X are the negative value for rotating the scale factor that item, C and F are translation item, E is y, and mx is
Ratio variation on the direction x, my are the ratio variation on the direction y, and t is depended on yaw angle, surveyed counterclockwise using x-axis by starting point
Value.Since body coordinate system is different from the definition of abovementioned mathematical plane coordinate system, need to be solved for two-dimensional surface just with calculated by coordinate
True yaw angle, t here are by the yaw angle after calculated by coordinate；K is shear factor=tan (u) along x-axis, and u is inclined
The difference at boat angle and y-axis, tilt angle at this time is measured relative to y-axis；By the correction based on image center point, image number
According to precision significantly improve, make high-precision oblique photograph model, Surveying and Mapping Industry had an important influence；
There is also noise effect, noise will lead to image and periodic stripe, spot occurs image after central point corrects
Deng being removed to longitude and latitude, the elements of exterior orientation of the image data combination central point after correction using kalman filter method
Noise, and image processing software MATLAB combination Kalman filter algorithm is used, influence of the noise to picture quality is filtered out, is reached
To filtering out interference of the noise to picture quality；
Kalman filtering removes the step of noise factor are as follows:
1) state equation and observational equation are first determined；
2) elements of exterior orientation of image data is imported；
3) input parameters: latitude, longitude, process-noise variance, observation noise variance, filter vector initial value, it is any when
Carve observation state variance；
4) coefficient for finding out predictive equation is realized finally by difference equation and is filtered；
The High-precision image for completing Geometry rectification and noise-filtering is imported into the end PC automation modeling software, automatic Building in batches
Mould software is completed aerial triangle and is resolved by process flows such as further geometric correction, simultaneous adjustments, and generates based on true
The ultra high density point cloud of image, and the high-resolution outdoor scene threedimensional model based on real image texture is generated with this；
It is general formats such as (OBJ, OSGB, DAE) by the oblique photograph model data export format of generation.Then model is carried out
Classification processing.The careful degree of texture that can be used to indicate to generate oblique photograph model using LOD classification, LOD are classified model
It is divided into pyramid rank, every level-one corresponds to the accuracy class of oblique photograph model；
Model implements spatial scalable compression step, the present invention in realize Fig. 2 is shown in model scalable compression method:
Fragment is carried out according to rank is divided to the oblique photograph threedimensional model after classification, each fragment model corresponds to a root
Node；Possess numerous child nodes under each fragment model；Each fragment model is divided into several lesser oblique photographs of block amount
Model corresponds to a child node under each sectional pattern；To according to the good fragment number of LOD grade classification, fragment model root node
It is saved with corresponding latitude and longitude coordinates, sectional pattern child node and latitude and longitude coordinates.Load grouping is carried out to the model of fragment, point
Steps are as follows for group:
Entire survey section model is determined first, and model grade is divided according to LOD industrial grade；
Then the model fragment number under each grade is determined, the fragment number under each series is determined according to LOD divided rank；
Then it determines the regional scope that fragment pattern number is respectively shown under the grade, determines each corresponding center of fragment model
Node longitude and latitude corresponding with its；
Finally determine the child node of several pieces of submodels and corresponding longitude and latitude under fragment model；
Information, which is formulated, is based on LOD hierarchical policy, and the principle mainly followed has: 1) the hierarchical model quantity under corresponding LOD grade is not
More than 15；2) each hierarchical model is adjacent, forms complete oblique photograph model；3) model pair of difference LOD grade
Fragment model quantity is answered to have significant difference；4) the minimum model fragment minimum number of LOD grade classification；
Information preservation after model to be classified to fragment is xml document, each nodal information of the in store oblique model of this file, this text
Part is index file, for obtaining the absolute position of every block models；
The bibliographic structure of XML file are as follows:
{FileName:/Tile_+000_+000.osgb:centerX,centerY,centerZ:Radius},
Relative path of the model file relative to central node under same catalogue, only loads the osgb file of a master control；
Wherein Tile file stores oblique photograph data, X, Y, and Z corresponds to node longitude and latitude and elevation information,
Radius corresponds to range distance, and every primary information is all stored in XML index file with such information；
Three-dimensional process software is imported to the xml document for completing model classification fragment, data compression is carried out, reduces the data of model
Amount, it should be understood that the nodal information of current oblique photograph model is tree, and entire hierarchy model possesses a center
Node, corresponding each fragment model have root node, and several block models possess corresponding child node under each fragment model, pass through modification
Entire model center node merges adjacent node, vacuates node, texture compression, and data compression can be wanted according to detailed programs
It asks, nodal information is divided into quaternary tree, Octree, any.The method and step of implements spatial scalable compression are as follows:
1) entire oblique photograph model center point is modified；
2) merge adjacent node；
3) node is vacuated；
4) texture compression；
After completing data compression, the number of folders of model, which is merged, to be vacuated, and generates configuration file config.scp, and with this point
The data compression under different LOD grades is not completed, and model is divided into pyramid rank by LOD classification, and every level-one corresponds to inclination and takes the photograph
The accuracy class of shadow model；
Model dynamically load step, the present invention in the batch high-precision oblique photograph model dynamic loading method based on OpenGL see
Fig. 3:
5) texture building and rendering are carried out using high-precision oblique photograph model of the OpenGL open graphic library high-volume to generation,
And convert the original data format of oblique photograph, realize that mobile terminal browses the load of high-precision oblique photograph model,
The method and step of building rendering is carried out to model are as follows:
5.1) Java programming language, development platform eclipse, the application program provided in conjunction with OpenGL open graphic library are provided
Programming interface carries out building rendering to model；
5.2) data conversion is carried out to the high-precision oblique photograph model of high-volume that rendering is built, by oblique photograph original data
Format (OBJ, OSGB, DAE) is converted to by the format after the binding SFML rendering building of OpenGL shape library.Rendering is built
Data are uploaded to server or by data configuration to engineering local, are completed by mobile device application program to index file
It reads, completes the load browsing of mobile device；
5.3) then to the elements of exterior orientation of UAV flight's video camera, the central point including entire model, the attitude angle of aircraft
Corresponding degree；The attitude angle of aircraft includes pitch angle, course angle, roll angle, then calculates determining within the scope of camera lens need
The oblique photograph model that model to be loaded is classified based on LOD；
6) when mobile zoom operations occur for the oblique photograph model in mobile device, first judge that grade locating for current model is
No is maximum model grade and model loading capacity is more than 60%, is then calculated by model grade to be loaded, if on mobile terminal
What is carried out is amplification oblique photograph model manipulation, then loads new Grade Model, and dynamic deletes the model loaded before, if
What mobile terminal carried out is to reduce oblique photograph model manipulation, then does not delete the oblique photograph model loaded.Final realize is moved
State loads high-volume high-precision oblique photograph model.
By the present invention in that being combined with the elements of exterior orientation of aircraft with affine transformation projection algorithm, imago in accurate calculating
The offset of point realizes the geometric correction to image, filters out noise band and spot caused by image using Kalman filter
Point influences；Finally by classification dynamically load strategy the high-precision oblique photograph threedimensional model of high-volume is led in Surveying Engineering
Mock-up form shows with high precision in domain, and in mobile field in the form of APP fast browsing, on the basis of the present invention can be real
The customization and exploitation of existing geography in formation software, realize the cross-platform displaying of Intelligent campus, digital city.
It should be understood that can be improved or converted according to appeal explanation for those skilled in the art,
And all these modifications and variations all should belong to scope of the appended claims of the present invention.
Claims (3)
1. the mobile terminal high-precision oblique photograph three-dimensional modeling method based on OpenGL, which comprises the following steps:
A is based on shadow according to formula 1-8 to the 60% Duplication image data in area to be measured acquired in five camera of UAV flight
Inconocenter picture point is corrected, and carries out the filtering processing of Kalman filtering algorithm noise to image after correction, obtains base image；
Formula 1:A=mx cos t
Formula 2:B=my (k cos t-sin t)
Formula 3:D=mx sin t
Formula 4:E=- 1 my (k sin t+cos t)
The translational movement in the direction formula 5:C=x
The translational movement in the direction formula 6:F=y
Formula 7:x1=Ax+By+C
Formula 8:y1=Dx+Ey+F
Wherein, scale factor, B and the D that A is X are the negative value for rotating the scale factor that item, C and F are translation item, E is y, and mx is
Ratio variation on the direction x, my are the ratio variation on the direction y, and t is depended on yaw angle, surveyed counterclockwise using x-axis by starting point
Value, since aircraft axes are different from the definition of abovementioned mathematical plane coordinate system, need to be solved for two-dimensional surface just with calculated by coordinate
True yaw angle, t here are by the yaw angle after calculated by coordinate；K is shear factor=tan (u) along x-axis, and u is inclined
The difference at boat angle and y-axis, tilt angle at this time is measured relative to y-axis；
B uses image processing software MATLAB combination Kalman filtering algorithm, filters out influence of the noise to picture quality；Kalman
The step of filtering removal noise factor are as follows:
1) noise state equation and observational equation are first determined；
2) elements of exterior orientation of image data is imported；
3) input parameters: latitude, longitude, process-noise variance, observation noise variance, filter vector initial value, it is any when
Carve observation state variance；
4) coefficient for finding out predictive equation is realized finally by difference equation and is filtered；
Base image is imported the end PC automation modeling software Smart 3D by c, by further to the base image after correction
Geometric correction, simultaneous adjustment process flow are generated the ultra high density point cloud based on real image, and are generated with this based on true shadow
As the high-resolution outdoor scene threedimensional model of texture；
D is to high-precision oblique photograph model is made after completing image correction, by the oblique photograph model data of generation according to LOD
Rank classification, the careful degree of texture that can be used to indicate to generate oblique photograph model using LOD classification, LOD are classified model
It is divided into pyramid rank, every level-one corresponds to the accuracy class of oblique photograph model；
E carries out fragment according to rank is divided to the oblique photograph threedimensional model after classification, each fragment model corresponds to one
Root node；Possess numerous child nodes under each fragment model；Each fragment model is divided into the lesser inclination of several block amounts to take the photograph
Shadow model corresponds to a child node under each sectional pattern；To according to the good fragment number of LOD grade classification, fragment model root section
Point and corresponding latitude and longitude coordinates, sectional pattern child node and latitude and longitude coordinates save；Load grouping is carried out to the model of fragment,
Steps are as follows for grouping:
1) entire survey section model is determined first, and model grade is divided according to LOD technology；
2) the model fragment number under each grade is then determined, the fragment number under each series is determined according to LOD divided rank
It is fixed；
3) it then determines the regional scope that respectively shows of fragment pattern number under the grade, determines that each fragment model is corresponding
Heart node longitude and latitude corresponding with its；
4) child node of several pieces of submodels and corresponding longitude and latitude under fragment model are finally determined；
F carries out data compression to the high-precision oblique photograph model for completing model classification fragment, reduces the data volume of model, should
The nodal information for being interpreted as current oblique photograph model is tree, and entire hierarchy model possesses a central node, right
Each fragment model answered has root node, and several block models possess corresponding child node under each fragment model, by modifying entire mould
Type central node merges adjacent node, vacuates node, texture compression to realize the fast browsing load of mobile terminal, data compression
Method and step are as follows:
1) entire oblique photograph model center point is modified；
2) merge adjacent node；
3) node is vacuated；
4) texture compression；
G carries out texture building and rendering using high-precision oblique photograph model of the OpenGL open graphic library interface to generation, and
The original data format of oblique photograph is converted, realizes that mobile terminal browses the load of high-precision oblique photograph model, it is right
Model carries out the method and step of building rendering are as follows:
1) application programming interface provided with open graphic library, will be good based on LOD grade classification by routine interface
Oblique photograph model carries out building rendering in OpenGL shape library；
2) data conversion is carried out to the high-precision oblique photograph model of high-volume that rendering is built, by oblique photograph original data lattice
Formula (OBJ, OSGB, DAE), which is converted to, is converted to mobile device format by OpenGL shape library binding SFML, and rendering is built
Data are uploaded to server or by data configuration to engineering local, are completed by mobile device application program to index file
It reads, completes the load browsing of mobile device；
3) then to the elements of exterior orientation of UAV flight's video camera, the central point including entire model, the attitude angle pair of aircraft
Degree is answered, the attitude angle of aircraft includes pitch angle, course angle, roll angle, calculates determination within the scope of camera lens and needs to load
The oblique photograph model that is classified based on LOD of model；
4) when mobile zoom operations occur for the oblique photograph model of mobile device end, first judge that grade locating for current model is
No is maximum model grade and model loading capacity is more than 60%, is then calculated by model grade to be loaded, if on mobile terminal
What is carried out is amplification oblique photograph model manipulation, then loads new Grade Model, and dynamic deletes the model loaded before, if
What mobile terminal carried out is to reduce oblique photograph model manipulation, then does not delete the oblique photograph model loaded, and final realize is moved
State loads high-volume high-precision oblique photograph model.
2. the mobile terminal high-precision oblique photograph three-dimensional modeling method according to claims 1 based on OpenGL, special
Sign is that the step f carries out data compression and merge node tool is three-dimensional process software；And data are subjected to texture compression
It is recorded in config.scp configuration file with the information of merge node.
3. the mobile terminal high-precision oblique photograph three-dimensional modeling method according to claim 1 based on OpenGL, special
Sign is, is formulated in the step d by import information and is based on LOD hierarchical policy, meets following condition:
1) the hierarchical model quantity under corresponding LOD grade is no more than 15；
2) each hierarchical model is adjacent, forms complete oblique photograph model；
3) collapsible mould type quantity corresponding to the model of difference LOD grade has significant difference；
4) the minimum model fragment minimum number of LOD grade classification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910196627.1A CN109934911B (en) | 2019-03-15 | 2019-03-15 | OpenGL-based three-dimensional modeling method for high-precision oblique photography of mobile terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910196627.1A CN109934911B (en) | 2019-03-15 | 2019-03-15 | OpenGL-based three-dimensional modeling method for high-precision oblique photography of mobile terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109934911A true CN109934911A (en) | 2019-06-25 |
| CN109934911B CN109934911B (en) | 2022-12-13 |
Family
ID=66987188
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910196627.1A Active CN109934911B (en) | 2019-03-15 | 2019-03-15 | OpenGL-based three-dimensional modeling method for high-precision oblique photography of mobile terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109934911B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110634184A (en) * | 2019-09-11 | 2019-12-31 | 西安恒歌数码科技有限责任公司 | Loading method of mass oblique photography data |
| CN110796152A (en) * | 2020-01-06 | 2020-02-14 | 杭州鲁尔物联科技有限公司 | Group building earthquake damage extraction method and system based on oblique photography |
| CN111325848A (en) * | 2020-03-02 | 2020-06-23 | 中国电建集团昆明勘测设计研究院有限公司 | Automatic batch deletion method for suspended matters of OSGB inclined model |
| CN111583404A (en) * | 2020-05-08 | 2020-08-25 | 广西壮族自治区自然资源遥感院 | Massive oblique photography three-dimensional model data scheduling method and system |
| CN112365598A (en) * | 2020-10-29 | 2021-02-12 | 深圳大学 | Method, device and terminal for converting oblique photography data into three-dimensional data |
| CN113868349A (en) * | 2021-08-31 | 2021-12-31 | 广东省测绘工程有限公司 | Rural house-ground integrated investigation and library building method based on hybrid network |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002074323A (en) * | 2000-09-01 | 2002-03-15 | Kokusai Kogyo Co Ltd | Method and system for generating three-dimensional urban area space model |
| JPWO2008152740A1 (en) * | 2007-06-13 | 2010-08-26 | 株式会社情報科学テクノシステム | Digital aerial 3D measurement system |
| WO2012132237A1 (en) * | 2011-03-31 | 2012-10-04 | パナソニック株式会社 | Image drawing device for drawing stereoscopic image, image drawing method, and image drawing program |
| CN103617650A (en) * | 2013-11-29 | 2014-03-05 | 哈尔滨工程大学 | Displaying method for complex three-dimensional terrain |
| CN104766360A (en) * | 2015-03-09 | 2015-07-08 | 腾讯科技（深圳）有限公司 | Rendering method and device of three-dimensional space data |
| US20150371431A1 (en) * | 2013-01-29 | 2015-12-24 | Andrew Robert Korb | Methods for analyzing and compressing multiple images |
| CN106327573A (en) * | 2016-08-25 | 2017-01-11 | 成都慧途科技有限公司 | Real scene three-dimensional modeling method for urban building |
| CN106373199A (en) * | 2016-08-31 | 2017-02-01 | 中测新图(北京)遥感技术有限责任公司 | Rapid oblique photography building model extraction method |
| CN107590854A (en) * | 2017-09-15 | 2018-01-16 | 福建四创软件有限公司 | Reservoir region three-dimensional live methods of exhibiting based on WEBGIS |
| CN107862744A (en) * | 2017-09-28 | 2018-03-30 | 深圳万图科技有限公司 | Aviation image three-dimensional modeling method and Related product |
| CN108090956A (en) * | 2017-11-10 | 2018-05-29 | 中国科学院空间应用工程与技术中心 | A kind of geometric correction method and system |
| CN108171787A (en) * | 2017-12-18 | 2018-06-15 | 桂林电子科技大学 | A kind of three-dimensional rebuilding method based on the detection of ORB features |
| CN108665536A (en) * | 2018-05-14 | 2018-10-16 | 广州市城市规划勘测设计研究院 | Three-dimensional and live-action data method for visualizing, device and computer readable storage medium |
| CN109186551A (en) * | 2018-08-08 | 2019-01-11 | 广州市城市规划勘测设计研究院 | Oblique photograph measures building feature point extracting method, device and storage medium |
| CN109238239A (en) * | 2018-09-12 | 2019-01-18 | 成都坤舆空间科技有限公司 | Digital measurement three-dimensional modeling method based on aeroplane photography |
-
2019
- 2019-03-15 CN CN201910196627.1A patent/CN109934911B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002074323A (en) * | 2000-09-01 | 2002-03-15 | Kokusai Kogyo Co Ltd | Method and system for generating three-dimensional urban area space model |
| JPWO2008152740A1 (en) * | 2007-06-13 | 2010-08-26 | 株式会社情報科学テクノシステム | Digital aerial 3D measurement system |
| WO2012132237A1 (en) * | 2011-03-31 | 2012-10-04 | パナソニック株式会社 | Image drawing device for drawing stereoscopic image, image drawing method, and image drawing program |
| US20150371431A1 (en) * | 2013-01-29 | 2015-12-24 | Andrew Robert Korb | Methods for analyzing and compressing multiple images |
| CN103617650A (en) * | 2013-11-29 | 2014-03-05 | 哈尔滨工程大学 | Displaying method for complex three-dimensional terrain |
| CN104766360A (en) * | 2015-03-09 | 2015-07-08 | 腾讯科技（深圳）有限公司 | Rendering method and device of three-dimensional space data |
| CN106327573A (en) * | 2016-08-25 | 2017-01-11 | 成都慧途科技有限公司 | Real scene three-dimensional modeling method for urban building |
| CN106373199A (en) * | 2016-08-31 | 2017-02-01 | 中测新图(北京)遥感技术有限责任公司 | Rapid oblique photography building model extraction method |
| CN107590854A (en) * | 2017-09-15 | 2018-01-16 | 福建四创软件有限公司 | Reservoir region three-dimensional live methods of exhibiting based on WEBGIS |
| CN107862744A (en) * | 2017-09-28 | 2018-03-30 | 深圳万图科技有限公司 | Aviation image three-dimensional modeling method and Related product |
| CN108090956A (en) * | 2017-11-10 | 2018-05-29 | 中国科学院空间应用工程与技术中心 | A kind of geometric correction method and system |
| CN108171787A (en) * | 2017-12-18 | 2018-06-15 | 桂林电子科技大学 | A kind of three-dimensional rebuilding method based on the detection of ORB features |
| CN108665536A (en) * | 2018-05-14 | 2018-10-16 | 广州市城市规划勘测设计研究院 | Three-dimensional and live-action data method for visualizing, device and computer readable storage medium |
| CN109186551A (en) * | 2018-08-08 | 2019-01-11 | 广州市城市规划勘测设计研究院 | Oblique photograph measures building feature point extracting method, device and storage medium |
| CN109238239A (en) * | 2018-09-12 | 2019-01-18 | 成都坤舆空间科技有限公司 | Digital measurement three-dimensional modeling method based on aeroplane photography |
Non-Patent Citations (2)
| Title |
|---|
| 杨争艳: "倾斜摄影测量三维重建中纹理映射的研究", 《中国优秀硕士学位论文全文数据库基础科学辑》 * |
| 谭仁春 等: "无人机倾斜摄影的城市三维建模方法优化", 《测绘通报》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110634184A (en) * | 2019-09-11 | 2019-12-31 | 西安恒歌数码科技有限责任公司 | Loading method of mass oblique photography data |
| CN110634184B (en) * | 2019-09-11 | 2023-01-17 | 西安恒歌数码科技有限责任公司 | Loading method of mass oblique photography data |
| CN110796152A (en) * | 2020-01-06 | 2020-02-14 | 杭州鲁尔物联科技有限公司 | Group building earthquake damage extraction method and system based on oblique photography |
| CN111325848A (en) * | 2020-03-02 | 2020-06-23 | 中国电建集团昆明勘测设计研究院有限公司 | Automatic batch deletion method for suspended matters of OSGB inclined model |
| CN111325848B (en) * | 2020-03-02 | 2022-04-15 | 中国电建集团昆明勘测设计研究院有限公司 | Automatic batch deletion method for suspended matters of OSGB inclined model |
| CN111583404A (en) * | 2020-05-08 | 2020-08-25 | 广西壮族自治区自然资源遥感院 | Massive oblique photography three-dimensional model data scheduling method and system |
| CN111583404B (en) * | 2020-05-08 | 2021-08-31 | 广西壮族自治区自然资源遥感院 | Massive oblique photography three-dimensional model data scheduling method and system |
| CN112365598A (en) * | 2020-10-29 | 2021-02-12 | 深圳大学 | Method, device and terminal for converting oblique photography data into three-dimensional data |
| CN112365598B (en) * | 2020-10-29 | 2022-09-20 | 深圳大学 | Method, device and terminal for converting oblique photography data into three-dimensional data |
| CN113868349A (en) * | 2021-08-31 | 2021-12-31 | 广东省测绘工程有限公司 | Rural house-ground integrated investigation and library building method based on hybrid network |
| CN113868349B (en) * | 2021-08-31 | 2023-01-13 | 广东省测绘工程有限公司 | Rural house-ground integrated investigation and library building method based on hybrid network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109934911B (en) | 2022-12-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109934911A (en) | Mobile terminal high-precision oblique photograph three-dimensional modeling method based on OpenGL | |
| WO2020192354A1 (en) | Blended urban design scene simulation method and system | |
| CN111629193B (en) | Live-action three-dimensional reconstruction method and system | |
| CN108919944B (en) | Virtual roaming method for realizing data lossless interaction at display terminal based on digital city model | |
| Stoter et al. | 3D GIS, where are we standing | |
| CN109242862B (en) | Real-time digital surface model generation method | |
| Pepe et al. | Techniques, tools, platforms and algorithms in close range photogrammetry in building 3D model and 2D representation of objects and complex architectures | |
| CN113112603B (en) | Method and device for optimizing three-dimensional model | |
| CN112113544B (en) | Remote sensing data processing method and system based on unmanned aerial vehicle image | |
| CN112150629A (en) | Vision-based coal inventory system and method | |
| CN110298136A (en) | Application based on BIM technology scene method of construction and system and in garden landscape digital modeling | |
| CN117315146B (en) | Reconstruction method and storage method of three-dimensional model based on trans-scale multi-source data | |
| CN115393240A (en) | GIS and BIM fusion method and device, electronic equipment and readable storage medium | |
| Wang et al. | Fast 3D reconstruction method based on UAV photography | |
| CN112200906B (en) | Entity extraction method and system for inclined three-dimensional model | |
| Kumar et al. | A CityGML extension for handling very large TINs | |
| CN114119892B (en) | Three-dimensional digital road network construction method based on BIM and GIS technologies | |
| CN110634184A (en) | Loading method of mass oblique photography data | |
| KR102587445B1 (en) | 3d mapping method with time series information using drone | |
| Pacina et al. | Accuracy of digital surface models derived from archival aerial photographs. Case study for the Czech Republic. | |
| CN114387488A (en) | Road extraction system and method based on Potree point cloud image fusion | |
| Cera et al. | Fast survey procedures in urban scenarios: some tests with 360 cameras | |
| Yang et al. | Study on 3D printing based on UAV oblique photogrammetry | |
| Gruber et al. | Urban data management—A modern approach | |
| Harshit et al. | Advancements in open-source photogrammetry with a point cloud standpoint |
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 |