CN107103603A - A kind of Objects extraction method of inclination measurement scene - Google Patents

A kind of Objects extraction method of inclination measurement scene Download PDF

Info

Publication number
CN107103603A
CN107103603A CN201710350830.0A CN201710350830A CN107103603A CN 107103603 A CN107103603 A CN 107103603A CN 201710350830 A CN201710350830 A CN 201710350830A CN 107103603 A CN107103603 A CN 107103603A
Authority
CN
China
Prior art keywords
atural object
line
region
contour line
inclination measurement
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
Application number
CN201710350830.0A
Other languages
Chinese (zh)
Inventor
隗刚
李伟
段业兴
贺蓉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Dao Heng Time Technology Co Ltd
Original Assignee
Beijing Dao Heng Time Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Dao Heng Time Technology Co Ltd filed Critical Beijing Dao Heng Time Technology Co Ltd
Priority to CN201710350830.0A priority Critical patent/CN107103603A/en
Publication of CN107103603A publication Critical patent/CN107103603A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • G06T7/0004Industrial image inspection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/12Edge-based segmentation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20112Image segmentation details
    • G06T2207/20116Active contour; Active surface; Snakes

Abstract

The present invention discloses the Objects extraction method of inclination measurement scene, and initial dip measurement scene uses tree-like storage structured data, and whole scene is root node, is made up of multiple scenario block child nodes, this method includes:(1) setting Objects extraction region;(2) setting height filtering surface;(3) atural object triangular facet pel is extracted, is handled by the way of recombination region union operation;(4) atural object is separated, all triangular facets carried out first using contour line to the inclination measurement data triangulation network after triangulation, subdivision in contour line are atural object pel;Secondly in former scenario node, all triangular facets in current atural object profile are extracted, single atural object pel packet are stored as, and label using the ID of atural object;(5) atural object is operated, including:Obtain the addition and search operaqtion of atural object primitive data, distance calculating, ray collision detection ground object model associative operation, and terrain object attribute information.

Description

A kind of Objects extraction method of inclination measurement scene
Technical field
The invention belongs to the technical field of power circuit design, more particularly to a kind of Objects extraction of inclination measurement scene Method, it can automatically extract house tree under the premise of data precision and the atural object profile degree of accuracy is ensured from inclination measurement scene The atural objects such as wood.
Background technology
Oblique photograph dimensional Modeling Technology and application are the focuses of survey field concern in recent years, due to its data precision Outstanding advantages are widely used in each side such as smart city, wisdom water conservancy, smart travel to high, ground mulching comprehensively etc..The present invention Relate generally to its application in power circuit design aspect.
Require to obtain the scissors crossing atural objects such as house, trees, road in the range of line corridor in power circuit design to believe Breath, measures these atural objects to the distance of power line, shaft tower etc., according to measurement reasonable selection shaft tower position, shaft tower model, frame Line parameter etc. meets the check requirement of power circuit design specification with guarantee.But the initial data of inclination measurement is typically using certainly Dynamicization mode generates three-dimensional scenic achievement, due to its formation mechanism, and obtained model is a continuous TIN (irregular triangle Net) to stick on figure, i.e. atural object, landform etc. be integrally provided as one, be not divided into atural object can be with not individually operated Object, can not enter row distance calculation and check in this case.Therefore, inclination measurement scene is used in power circuit design When, it is necessary first to singulation processing is carried out to the atural object that the power circuits such as the house in scene, trees are concerned about.
The atural object singulation mode on the market in inclination measurement scene three-dimensional applications mainly has three kinds at present:One kind is on ground Supporting vector face is superimposed on thing, is monomer effect to the eye, actually simply building profile point is simply queried Vector face, rather than monomer separation truly;Two be simple triangular facet packet, and original triangular facet in scene is direct Atural object or landform are divided into, but such processing can cause atural object border that zigzag, and the atural object side of different accuracy grade is presented Boundary is inconsistent, and pole unsightly, and can not be with terrain scene when needing to replace with atural object the high-precision model of manual manufacture Realize seamless spliced;Three be that artificial setting atural object border carries out triangular facet cutting separation atural object, and such a method atural object border is compared Neat appearance, different accuracy level boundaries are consistent, and do not damage original data precision.But this method operation is comparatively laborious, both takes Manpower, is extremely difficult to atural object profile and accurately matches again.
The content of the invention
The technology of the present invention solves problem:Overcoming the deficiencies in the prior art, there is provided a kind of atural object of inclination measurement scene Extracting method, it has higher automaticity, it is only necessary to which simple operations can obtain atural object precise boundary, with house, tree The function that the atural objects such as wood are really separated with inclination measurement scene, it is ensured that atural object has the accuracy of initial data, Ke Yifang Atural object is just replaced with to the model of artificial fine modeling, edge is accurately spliced, can be in the case of retention data be high-precision Realize rapid loading display.
The present invention technical solution be:The Objects extraction method of this inclination measurement scene, initial dip measurement field Scape use tree-like storage structured data, whole scene be root node, be made up of multiple scenario block child nodes, each child node except Possess the triangular facet pel collection of itself, also comprising several higher precision grade scenario block child nodes, the inclination measurement scene Objects extraction method comprises the following steps:
(1) setting Objects extraction region, Objects extraction region is will extract the scope of atural object, and the region is taken turns by some borders Profile and some empty contour line compositions, boundary contour define the foreign range boundary line for extracting region, empty contour line definition Extract the internal range boundary line in region, for the part that need not be extracted in exclusionary zone, within border beyond cavity for ground Thing extracts region;
(2) setting height filtering surface, highly filtered face is such a curved surface, and its horizontal extent is limited by Objects extraction region System, just change rises and falls (or being constant) physical features of the elevation in the range of, and the filtering surface is higher than background terrain and is less than and to carry The minimum altitude of atural object main part is taken, atural object forms isolated island triangular facet set one by one higher than the part of filtering surface, by institute Some isolated island triangular facets take out and project to horizontal plane, and it projects the boundary line that outline is atural object;
(3) atural object triangular facet pel is extracted, all atural object triangular facets obtained according to step (2) calculate atural object boundary profile Line, is handled by the way of recombination region union operation here;
(4) atural object is separated, using the atural object contour line obtained by step (3) calculating from inclination measurement extracting data atural object, All triangular facets after triangulation, subdivision in contour line are carried out to the inclination measurement data triangulation network using contour line first For atural object pel;Secondly in former scenario node, all triangular facets in current atural object profile are extracted, are stored as individually The packet of atural object pel, and labelled using the ID of atural object;
(5) atural object is operated, and atural object operation includes:Atural object primitive data is obtained, distance is calculated, ray collision detection atural object mould Type associative operation, and terrain object attribute information addition and search operaqtion.
, only need to be to original because the present invention retains Raw scene data, it is possible to easily change or delete atural object Contextual data is cut again according to amended all relative profiles.Because the present invention saves all atural object number of contours According to when needing to change the contextual data of higher precision, only need to reading all outline datas automatically and cut again, it is not necessary to again Secondary setting atural object profile by hand.The present invention obtains atural object border using atural object contour extraction method, simple to operate, and accuracy is high, And multiple atural objects can be extracted simultaneously, also substantially increase operating efficiency while accuracy is improved.Due to present invention use The method of actual separation initial dip measurement scene pel carries out singulation, realizes singulation truly.Singulation Atural object afterwards can be individually operated to its as independent individual, such as sets the colors different from background, individually chooses, and obtains summit Data etc., meet power circuit design requirement.The present invention, will be with wheel by the way of atural object profile wire cutting triangle primitives The intersecting triangle of profile is rebuild as the combination of multiple triangles, then chooses triangle in contour line as atural object number According to not only ensure that the neat in edge of atural object profile is attractive in appearance, and conveniently replace with higher precision, more attractive artificial model. All accuracy class primitive datas of the original scene of present invention cutting, the atural object primitive data after cutting retains in pel group form In former scenario node, identified by atural object ID.Before the accuracy class and precision switchover policy and singulation of line map tuple Do not change, so as to ensure that higher atural object display precision and rendering speed.Ensure that with any zoom state simultaneously Thing border is constant.Therefore, the present invention has higher automaticity, it is only necessary to which simple operations can obtain atural object precise boundary, The function being really separated with atural objects such as house, trees with inclination measurement scene, it is ensured that atural object has the essence of initial data Exactness, can facilitate the model that atural object is replaced with to artificial fine modeling, and edge is accurately spliced, can be high-precision in retention data Rapid loading display is realized in the case of degree.
Brief description of the drawings
Fig. 1 is the main flow chart of the Objects extraction method of the inclination measurement scene according to the present invention.
Fig. 2 is the schematic diagram of the recombination region of the present invention.
Fig. 3 is the flow chart of the atural object contours extract of the present invention.
Embodiment
As shown in figure 1, the Objects extraction method of this inclination measurement scene, initial dip measurement scene uses tree-like storage Structured data, whole scene is root node, is made up of multiple scenario block child nodes, triangle of each child node except possessing itself Face pel collection, also comprising several higher precision grade scenario block child nodes, the Objects extraction method bag of the inclination measurement scene Include following steps:
(1) setting Objects extraction region, Objects extraction region is will extract the scope of atural object, and the region is taken turns by some borders Profile and some empty contour line compositions, boundary contour define the foreign range boundary line for extracting region, empty contour line definition Extract the internal range boundary line in region, for the part that need not be extracted in exclusionary zone, within border beyond cavity for ground Thing extracts region;
(2) setting height filtering surface, highly filtered face is such a curved surface, and its horizontal extent is limited by Objects extraction region System, just change rises and falls (or being constant) physical features of the elevation in the range of, and the filtering surface is higher than background terrain and is less than and to carry The minimum altitude of atural object main part is taken, atural object forms isolated island triangular facet set one by one higher than the part of filtering surface, by institute Some isolated island triangular facets take out and project to horizontal plane;
(3) atural object triangular facet pel is extracted, all atural object triangular facets obtained according to step (2) calculate atural object boundary profile Line, is handled by the way of recombination region union operation here;
(4) atural object is separated, using the atural object contour line obtained by step (3) calculating from inclination measurement extracting data atural object, All triangular facets after triangulation, subdivision in contour line are carried out to the inclination measurement data triangulation network using contour line first As atural object pel;Secondly in former scenario node, all triangular facets in current atural object profile is extracted, list is stored as Only atural object pel packet, and labelled using the ID of atural object;
(5) atural object is operated, and atural object operation includes:Atural object primitive data is obtained, distance is calculated, ray collision detection atural object mould Type associative operation, and terrain object attribute information addition and search operaqtion.
, only need to be to original because the present invention retains Raw scene data, it is possible to easily change or delete atural object Contextual data is cut again according to amended all relative profiles.Because the present invention saves all atural object number of contours According to when needing to change the contextual data of higher precision, only need to reading all outline datas automatically and cut again, it is not necessary to again Secondary setting atural object profile by hand.The present invention obtains atural object border using atural object contour extraction method, simple to operate, and accuracy is high, And multiple atural objects can be extracted simultaneously, also substantially increase operating efficiency while accuracy is improved.Due to present invention use The method of actual separation initial dip measurement scene pel carries out singulation, realizes singulation truly.Singulation Atural object afterwards can be individually operated to its as independent individual, such as sets the colors different from background, individually chooses, and obtains summit Data etc., meet power circuit design requirement.The present invention, will be with wheel by the way of atural object profile wire cutting triangle primitives The intersecting triangle of profile is rebuild as the combination of multiple triangles, then chooses triangle in contour line as atural object number According to not only ensure that the neat in edge of atural object profile is attractive in appearance, and conveniently replace with higher precision, more attractive artificial model. All accuracy class primitive datas of the original scene of present invention cutting, the atural object primitive data after cutting retains in pel group form In former scenario node, identified by atural object ID.Before the accuracy class and precision switchover policy and singulation of line map tuple Do not change, so as to ensure that higher atural object display precision and rendering speed.Ensure that with any zoom state simultaneously Thing border is constant.Therefore, the present invention has higher automaticity, it is only necessary to which simple operations can obtain atural object precise boundary, The function being really separated with atural objects such as house, trees with inclination measurement scene, it is ensured that atural object has the essence of initial data Exactness, can facilitate the model that atural object is replaced with to artificial fine modeling, and edge is accurately spliced, can be high-precision in retention data Rapid loading display is realized in the case of degree.
In addition, in the step (1), setting Objects extraction region includes:Boundary contour and empty profile are chosen by hand Line, by choosing intersecting point one by one in three-dimensional scenic during first extraction atural object, defines some closing multi-line sections, closes multi-line section For contour line, each contour line represents a border or cavity in region, and a border and its internal some cavities define one Continuum, all contour lines are simple polygon, and phase friendship is not present between same polygonal any two lines section Condition, any two contour line is non-intersect;Two be that the area data prestored is loaded from file, and the extraction area data of manual definition is protected It is stored in file, it is necessary to be loaded when reusing from file and directly act on inclination measurement scene.
In addition, in the step (2), highly filtered face burns sth to the ground two kinds of the highly filtered face in highly filtered face and mountain region; If landform residing for atural object be level land or atural object main part far above the microrelief on ground and ground, using horizontal plane conduct Filtering surface, constructs all summits on the triangulation network, and the triangulation network according to zone boundary contour line and empty contour line and is all set to same One height, the plane is to extract the highly filtered face in level land in region;If landform is not smooth enough residing for atural object, or atural object main body Highly relatively low, it is not easy to mutually to be distinguished with earth bulge, using the highly filtered face in the mountain region being defined by the user, the filtering surface is with physical features Fluctuations, first define previous step the boundary contour and empty contour line that extract region by certain horizontal interval progress Interpolation processing, then the elevation of each point of inclination measurement data acquisition is inquired about, added by operating personnel by way of mouse is clicked Plus some ground points, ground point is evenly distributed in extraction region planted agent and covers all ground as far as possible, for wanting what is filtered out Microspike, also gathers its summit, and it is irregular that above-mentioned all contour line interpolation points and discrete ground point are merged together into construction The triangulation network, this triangulation network is the highly filtered face in mountain region.
In addition, for the highly filtered face in level land, directly comparing summit elevation and horizontal plane elevation;It is highly filtered for mountain region Face, height is inquired about using the apex horizontal coordinate of inclination measurement data on this filtering surface, if apex height is more than filtering surface height Angle value, then this point is atural object summit, otherwise it is assumed that being ground point, is ignored;If all summits of a triangular facet are all higher than Filtering surface, then be labeled as atural object triangular facet by this triangular facet.
In addition, in the step (3), use the mode of recombination region union operation for:
(a) recombination region A and B are made into standardization processing respectively, boundary contour is processed as summit by arranged counterclockwise Closing multi-line section, empty contour line is processed as closing multi-line section of the summit by arranged clockwise;
(b) intersection point of each contour line in the A of zoning and each contour line in the B of region, and each handing over Point is separately recorded in two profile cable architectures, and each intersection point is identified using globally unique intersection point number;
(c) every contour line is split as some multi-line sections according to intersection point, every multi-thread segment record its start-stop intersection point number, institute There is multi-line section to be included into two multi-line section groups of A, B respectively by its original affiliated area;
(d) multi-line section is filtered, A group multi-line sections are filtered using region B, the multi-line section being in the B of region is rejected, reservation is in Multi-line section outside B;B group multi-line sections are similarly filtered using region A;
(e) alternately take out multi-line section from two groups of AB according to start-stop intersection point number to be spliced, the new contour line of formation;
(f) union recombination region C is constructed, the contour line that step (e) is obtained adds this region, the profile of arranged counterclockwise Wire tag is border, and the contour line of arranged clockwise is labeled as cavity;
(g) all outermost layer boundary contours in recombination region C are obtained, atural object boundary line is used as.
In addition, in the step (4), for a triangular facet to be split, triangle and current outline line are calculated first Intersection point, this triangle is ignored if without intersection point;Contour line is split as by some multi-line sections according to intersection point, retains triangle scope Interior multi-line section, deletes other multi-line sections;The summit for some multi-line sections that 3 summits of former triangle and step (3) are obtained Reconfigure the triangulation network, it is ensured that each multi-line section is appeared in as the side of triangle in the triangulation network of neotectonics, these broken lines Some sections can be divided into, but there can not be missing;After the completion of triangle subdivision, former triangle is deleted from the scene triangulation network Remove, instead all new triangles of triangulation generation.
In addition, in the step (4), each triangle in being grouped for scene pel calculates triangle level first Central point, the plane coordinates (XY) on Atria summit is separately summed again divided by 3, triangle center point coordinates is obtained;Secondly Judge triangle center point whether in atural object profile;Atural object pel is finally separating, if current triangle is in outline polygon It is interior, then removed during it is grouped from the landform of world subdivision, and be added in current atural object packet, if not in outline polygon It is interior, then ignore.
In addition, in the step (5), the primitive information of atural object, atural object primitive information is obtained from three-dimensional scenic to be needed to refer to Determine the accuracy class of atural object ID and model, inquired about step by step since root node, the node until reaching given level, if node Primitive data is not loaded, then is loaded from corresponding data file, and the pel group specified is obtained according to ID, is packaged as the lattice needed Formula, returns to root node step by step.
In addition, in the step (5), ray collision detection is:First by the mouse click location in 3-D graphic window The ray in three dimensions is converted to, then ray is touched with the atural object packet in each visible elements in inclination measurement scene Detection is hit, if collision distance is less than current minimum collision distance, the atural object belonging to the triangular facet of current point and hit is recorded ID, has handled the inclination measurement atural object that all visible elements are just hit.
In addition, in the step (5), distance is calculated as:The distance between the electrical bodies such as power line and atural object are calculated, is used The full accuracy pel of atural object is calculated, and for each atural object, obtains its full accuracy grade triangle network data, these are calculated successively The distance of triangular facet and electric wire simultaneously takes minimum value.
The present invention is described in detail with reference to embodiment.
Initial dip measurement scene uses tree-like storage structured data.Whole scene is root node, by multiple scenario blocks Node is constituted.Each triangular facet pel collection of the child node except possessing itself, also comprising several higher precision grade scenario blocks Child node.So successively subdivision is gone down, until highest ranking child node.Scenario node accuracy class is higher, and its details is abundanter, And areal extent is smaller.It is according to the decision of the distance of node to video camera (representing the observer in three-dimensional scenic) during render scenes Render present node pel collection and still render child node.
The inclination measurement scene house trees extracting method that the present invention is provided, the inclination for above tree-like storage structure is surveyed Measure contextual data and carry out atural object cutting lock out operation, referring to Fig. 1, comprise the following steps:
1) setting Objects extraction region
Objects extraction region defines the scope of atural object to be extracted, and the region is by some boundary contours and some empty profiles Line is constituted.Boundary contour defines the foreign range boundary line for extracting region, and empty contour line defines the internal range for extracting region Boundary line, for need not be extracted in exclusionary zone part (exclusion be not required to part to be processed can improve subsequent treatment speed Degree), i.e., within border beyond cavity for Objects extraction region.Can be using two ways setting Objects extraction region, one is hand Work chooses boundary contour and empty contour line, and two be that the area data prestored is loaded from file.
(1) manual definition Objects extraction region
Some closing multi-line sections can be defined by choosing intersecting point one by one in three-dimensional scenic during first extraction atural object, that is, taken turns Profile.Each contour line represents a border or cavity in region, and a border and its internal some cavities define one continuously Region.All contour lines should be simple polygon, i.e., phase friendship is not present between same polygonal any two lines section Condition.Any two contour line should not intersect.Each contour line answers adeditive attribute, is border or cavity to distinguish.
Generally, the extraction region of definition should surround the institute of atural object floor projection a little and area is as far as possible small, to protect Card obtains the complete and correct atural object in border, but in order to shorten the artificial treatment time, can also define relatively rough extraction area Domain, and fine processing is carried out by the follow-up link that automatically processes.In addition, in power circuit design for reduce atural object and electric wire it Between distance amount of calculation, improve check speed, can also only choose the nearer part of distance line in atural object, this is to meet electricity Line of force road design actual conditions.
(2) using existing Objects extraction area data
The extraction area data of manual definition can be saved in file, it is necessary to be loaded and direct from file when reusing Act on inclination measurement scene.It is particularly suitable for use in original contextual data being replaced by the situation of higher precision contextual data, saves Again the troublesome operation in manual definition Objects extraction region one by one.Area data in file saves atural object ID, during establishment Between etc. information, identical order is carried out when should be created with last time when choosing atural object again.
2) setting height filtering surface
Highly filtered face is such a curved surface, and its horizontal extent is limited by Objects extraction region, and elevation is in the range of Physical features height change rises and falls (or being constant).(including ground and attachment are on the ground higher than background terrain for the overanxious face Short vegetation, vehicle etc.), and less than to extract the minimum altitude of atural object main part, then atural object is higher than the part of filtering surface " isolated island " one by one is formed, all isolated island triangular facets are taken out and horizontal plane is projected to, it projects outline as The boundary line of thing.The atural objects such as the trees that such a algorithm caves inward for the building with vertical wall face and bottom are accurate Degree is particularly high.
To take into account efficiency and the degree of accuracy of Objects extraction, highly filtered face is burnt sth to the ground and two kinds of mountain region.If atural object institute Locate landform be level land or atural object main part far above the microrelief (vegetation, vehicle etc.) on ground and ground, water can be used Plane is used as filtering surface.The benefit of horizontal plane is that the summit for differentiating triangular facet is in filtering surface top or lower section and its simple fast Speed, as long as the height value of triangle vertex of surface is directly compared with overanxious face height value as size.Specific practice is according to upper All summits on zone boundary contour line and empty contour line the construction triangulation network, and the triangulation network that one step is defined all are set to same Individual height, the i.e. plane are to extract the horizontal plane in region.
If landform is not smooth enough residing for atural object, such as hillside river levee, or atural object body height are relatively low, it is not easy to ground Projection is mutually distinguished, then should be using the highly filtered face in mountain region being defined by the user, and the filtering surface is with physical features fluctuations.Such a method Speed is slower, but the degree of accuracy is high, can save adjustment correction time of the later stage to wrong data, generally efficiency is higher.Specifically Way is that the boundary contour and empty contour line that previous step is defined into extraction region first are inserted by certain horizontal interval Value processing, then the elevation of each point of inclination measurement data acquisition is inquired about, this step is automatically performed, and is intervened without operating personnel; Some ground points are added by way of mouse is clicked by operating personnel, ground point is evenly distributed in extraction region planted agent and to the greatest extent may be used All ground can be covered, for wanting the microspike filtered out, its summit should be also gathered.By above-mentioned all contour line interpolation points and Discrete ground point is merged together construction TIN, and this triangulation network is the highly filtered face in mountain region.
After filtering surface construction complete, operating personnel's manual setting this filtering surface height, until meeting following require:(1) institute The atural object main part to be extracted is located on filtering surface;(2) ground and earth bulge are all located under filtering surface.If can not reach To this requirement, then the part that region is allowed to avoid filtering out is extracted in modification, or addition ground point reconfigures filtering surface, directly Untill meeting and requiring.
For horizontal filtering face, directly compare summit elevation and horizontal plane elevation;For mountain region filtering surface, surveyed using tilting The apex horizontal coordinate of amount data inquires about height on this filtering surface.If apex height is more than filtering surface height value, this point is Atural object summit, otherwise it is assumed that being ground point, is ignored.If all summits of a triangular facet are all higher than filtering surface, by this Triangular facet is labeled as atural object triangular facet.
3) atural object border is calculated
All atural object triangular facets obtained according to previous step calculate atural object boundary contour automatically.Here recombination region is used The mode of union operation is handled.Recombination region refers to a kind of such region (see Fig. 2):
(1) recombination region is made up of at least one independent subregion;
(2) independent subregion refers to the region surrounded by a continuous outermost layer boundary contour;
(3) outermost layer border refers to the boundary contour not surrounded by other any contour lines;
(4) each boundary contour (being not limited to outermost layer boundary contour) can surround some empty contour lines, empty Hole contour line necessarily be in boundary contour;
(5) some boundary contours can be surrounded in each empty contour line;
(6) cannot be directly nested between boundary contour;
(7) cannot be directly nested between empty contour line;
(8) all contour lines should be simple polygon, and all contour lines must not intersect.
Two recombination region (setting respectively A and B) union operation rules are following (referring to Fig. 3):
(1) recombination region A and B are made into standardization processing respectively, boundary contour is processed as summit by arranged counterclockwise Closing multi-line section, empty contour line is processed as closing multi-line section of the summit by arranged clockwise;
(2) intersection point of each contour line in the A of zoning and each contour line in the B of region, and each handing over Point is separately recorded in two profile cable architectures.Each intersection point is identified using globally unique intersection point number, additionally needs point coordinates, Serial number in current outline line intersection point etc.;
(3) every contour line is split as some multi-line sections according to intersection point, every multi-thread segment record its start-stop intersection point number.Institute There is multi-line section to be included into two multi-line section groups of A, B respectively by its original affiliated area;
(4) multi-line section is filtered.A group multi-line sections are filtered using region B, rejected at the multi-line section inside the B of region, reservation Multi-line section outside B;B group multi-line sections are similarly filtered using region A;
(5) splice multi-line section, form new contour line.Specific practice is that a multi-line section is taken out from A group multi-line sections a1, searched in the B of region with multi-line section a1Period is terminated as the multi-line section b of starting period1, spliced in current multi-line section Afterwards, and terminating point period is changed for b1Terminating point period, form new multi-line section a1b1;Searched again from the A of region with a1b1 Terminate multi-line section a of the period for starting period2And splice in thereafter, form multi-line section a1b1a2.Such cycle alternation is carried out, directly Multi-line section, the contour line after as one merging are closed to forming one.Every used multi-line section will be deleted from former group Remove.Repeat aforesaid operations, all multi-line sections until having handled;
(6) union recombination region C is constructed, the contour line that step (5) is obtained adds this region, the profile of arranged counterclockwise Wire tag is border, and the contour line of arranged clockwise is labeled as cavity;
(7) all outermost layer boundary contours in recombination region C, as atural object boundary line are obtained.
Specifically in this patent, an empty recombination region A is first initialized, then each triangular facet is regard as recombination region Domain B and A carries out union operation, obtains union refion C.Region C is done as new region A with next triangular facet region B again Union operation, so circulation, all triangular facets until having handled.Extract all ragged edges in final operation result region C Boundary's contour line, as atural object envelope.
4) Objects extraction
Using step 3) calculate obtained by atural object contour line from inclination measurement extracting data atural object.Contour line is used first All triangular facets carried out to the inclination measurement data triangulation network after triangulation, subdivision in contour line are atural object pel.
(1) triangulation
For a triangular facet to be split, the intersection point of triangle and current outline line is calculated first, is neglected if without intersection point Omit this triangle.Contour line is split as by some multi-line sections according to intersection point.Retain the multi-line section in the range of triangle, delete other Multi-line section.
The summit for some multi-line sections that 3 summits of former triangle and previous step are obtained reconfigures the triangulation network.Construction The triangulation network can use popular De Luonai Triangulation Methods (Delaunay triangulation).It should be noted that Ensure in the triangulation network that each multi-line section appears in neotectonics as the side of triangle, if these broken lines can be divided into Dry section, but can not have missing.
After the completion of triangle subdivision, former triangle is deleted from the scene triangulation network, instead the institute of triangulation generation There is new triangle.
The situation of one triangle of processing be these are only, it is necessary to handle other all triangles of current scene piecemeal successively Shape.It is expanded to other world subdivisions of current accuracy grade, the inclination measurement contextual data of other accuracy classes successively again.
Can be non-intersect as quick rejecting means using profile encirclement frame and world subdivision encirclement frame during subdivision, without every Individual triangle is all calculated in detail.
(2) atural object is separated
In former scenario node, all triangular facets in current atural object profile are extracted, single atural object is stored as Pel is grouped, and is labelled using the ID of atural object.The packet of this pel is pel packet of the atural object in current scene node.
Each triangle in being grouped for scene pel, performs following operation:
Triangle horizontal center point is calculated first.By the plane coordinates (XY) on Atria summit be separately summed again divided by 3, you can obtain triangle center point coordinates.
Secondly judge triangle center point whether in atural object profile.Rudimentary algorithm is:One origin of construction is triangle Central point, and direction is the ray of (1,0).Calculate ray situation about intersecting with atural object outline polygon, whenever ray with it is polygon When a line of shape is intersecting, with regard to carrying out conversion or opposite conversion once from inside to outside.Record is tracked through polygon Several parities, illustrates that triangle center point is located in polygon through odd-times, and even-times explanation is located at outside polygon.
It is finally separating atural object pel.If current triangle is in outline polygon, by it from the landform of world subdivision Remove, and be added in current atural object packet in packet.If not in outline polygon, ignored.
The every other triangle of current scene piecemeal is handled successively.Pel quilt after the completion of processing in original scenario node It is divided into a landform packet and multiple atural objects packet (being belonging respectively to multiple atural objects).Atural object pel packet atural object ID marks, and Pel extraction can be carried out by atural object ID or the operation such as color is set.
Aforesaid operations are expanded to other scenario nodes of present level, to ensure atural object integrality;It is expanded to other again Level data, so as to ensure the precision uniformity of atural object and terrain data.
5) atural object is operated
Atural object operation is mainly calculated including obtaining atural object primitive data, distance, ray collision detection ground object model is related grasps Make, and terrain object attribute information addition and search operaqtion.
1. primitive information is obtained
The primitive information of atural object is needed to use in some cases, and ground object model is such as exported as to the file of specified format for other Program use.Atural object primitive information is obtained from three-dimensional scenic to be needed to specify the accuracy class of atural object ID and model.From Root node starts to inquire about step by step, the node until reaching given level.If node primitive data is not loaded, from corresponding data File is loaded.The pel group specified is obtained according to ID, the form needed is packaged as, root node is returned to step by step.
2. ray collision detection
It is frequently necessary to choose atural object with mouse in three-dimensional circuits design to perform other operations.First by 3-D graphic window In mouse click location be converted to ray in three dimensions, then by each visible elements in ray and inclination measurement scene Atural object packet carry out collision detection, if collision distance is less than current minimum collision distance, record current point and hit Atural object ID belonging to triangular facet, the inclination measurement atural object that having handled all visible elements can be hit (may also be hit Any object in landform or no point).
3. distance is calculated
The main purpose that atural object is imported in power circuit design is to calculate the distance between the electrical bodies such as power line and atural object, And make comparisons to differentiate whether current design reasonable with the safe distance in design specification.It is high due to electrically checking precision prescribed, Therefore it should be calculated using the full accuracy pel of atural object, i.e., for each atural object, obtain its full accuracy grade triangle netting index According to calculating the distance of these triangular facets and electric wire successively and take minimum value.
4. adeditive attribute information
An attribute data structures are created for each atural object, are identified using atural object ID.Terrain object attribute can be divided into shared category Property and the class of privately owned attribute two, have attribute be applied to each atural object, comprising atural object classification, atural object title, atural object color, whether Display etc.;Privately owned attribute is applied to a certain or several atural objects, such as height of house, the owner, trees height.Terrain object attribute can Merge with atural object profile and be saved in data file, so as to reading next time.
It is described above, be only presently preferred embodiments of the present invention, any formal limitation not made to the present invention, it is every according to Any simple modification, equivalent variations and the modification made according to the technical spirit of the present invention to above example, still belong to the present invention The protection domain of technical scheme.

Claims (10)

1. a kind of Objects extraction method of inclination measurement scene, it is characterised in that:Initial dip measurement scene uses tree-like storage Structured data, whole scene is root node, is made up of multiple scenario block child nodes, triangle of each child node except possessing itself Face pel collection, also comprising several higher precision grade scenario block child nodes, the Objects extraction method bag of the inclination measurement scene Include following steps:
(1) setting Objects extraction region, Objects extraction region is will extract the scope of atural object, and the region is by some boundary contours Constituted with some empty contour lines, boundary contour defines the foreign range boundary line for extracting region, empty contour line definition is extracted The internal range boundary line in region, for the part that need not be extracted in exclusionary zone, being carried for atural object beyond cavity within border Take region;
(2) setting height filtering surface, highly filtered face is such a curved surface, and its horizontal extent is limited by Objects extraction region, Physical features height change of the elevation in the range of rises and falls, and the filtering surface, which is higher than background terrain and is less than, will extract atural object main part Minimum altitude, atural object forms isolated island triangular facet set one by one higher than the part of filtering surface, and all isolated island triangular facets are taken Go out and project to horizontal plane, it projects the boundary line that outline is atural object;
(3) atural object triangular facet pel is extracted, all atural object triangular facets obtained according to step (2) calculate atural object boundary contour, Here handled by the way of recombination region union operation;
(4) separate atural object, using step (3) calculate obtained by atural object contour line from inclination measurement extracting data atural object, first It is ground that all triangular facets after triangulation, subdivision in contour line are carried out to the inclination measurement data triangulation network using contour line Thing pel;Secondly in former scenario node, all triangular facets in current atural object profile is extracted, are stored as individually Thing pel is grouped, and is labelled using the ID of atural object;
(5) atural object is operated, and atural object operation includes:Atural object primitive data is obtained, distance is calculated, ray collision detection ground object model phase Close operation, and terrain object attribute information addition and search operaqtion.
2. the Objects extraction method of inclination measurement scene according to claim 1, it is characterised in that:In the step (1), Setting Objects extraction region includes:Boundary contour and empty contour line are chosen by hand, by three-dimensional during first extraction atural object Intersecting point is chosen in scene one by one, some closing multi-line sections are defined, closing multi-line section is contour line, and each contour line represents region A border or cavity, a border and its internal some cavities define a continuums, and all contour lines are simple Intersecting situation is not present between polygon, same polygonal any two lines section, any two contour line is non-intersect;Two are Load the area data that prestores from file, the extraction area data of manual definition be saved in file, it is necessary to when reusing from File loads and directly acts on inclination measurement scene.
3. the Objects extraction method of inclination measurement scene according to claim 2, it is characterised in that:In the step (2), Highly filtered face burns sth to the ground two kinds of the highly filtered face in highly filtered face and mountain region;If landform residing for atural object is level land or ground owner Body portion is far above the microrelief on ground and ground, using horizontal plane as filtering surface, according to zone boundary contour line and All summits on empty contour line the construction triangulation network, and the triangulation network are all set to same height, and the plane is extraction region The highly filtered face in level land;If landform is not smooth enough residing for atural object, or atural object body height is relatively low, it is not easy to earth bulge phase Distinguish, using the not contour filtering surface being defined by the user, previous step is defined extract first by the filtering surface with physical features fluctuations The boundary contour in region and empty contour line carry out interpolation processing by certain horizontal interval, then inquire about inclination measurement data and obtain The elevation for taking each to put, some ground points are added by operating personnel by way of mouse is clicked, and ground point is extracting region Planted agent is evenly distributed and covers all ground as far as possible, for wanting the microspike filtered out, also gathers its summit, by above-mentioned institute There are contour line interpolation point and discrete ground point to be merged together construction TIN, this triangulation network is that mountain region is highly filtered Face.
4. the Objects extraction method of inclination measurement scene according to claim 3, it is characterised in that:Spent for level land height Filtering surface, directly compares summit elevation and horizontal plane elevation;For the highly filtered face in mountain region, using the summit water of inclination measurement data Flat seat, which is marked on this filtering surface, inquires about height, if apex height is more than filtering surface height value, and this point is atural object summit, is otherwise recognized To be ground point, ignored;If all summits of a triangular facet are all higher than filtering surface, this triangular facet is labeled as atural object Triangular facet.
5. the Objects extraction method of inclination measurement scene according to claim 4, it is characterised in that:In the step (3), Use the mode of recombination region union operation for:
(a) recombination region A and B are made into standardization processing respectively, boundary contour is processed as envelope of the summit by arranged counterclockwise Multi-line section is closed, empty contour line is processed as closing multi-line section of the summit by arranged clockwise;
(b) intersection point of each contour line in the A of zoning and each contour line in the B of region, and each intersection point is divided Do not record in two profile cable architectures, each intersection point is identified using globally unique intersection point number;
(c) every contour line is split as some multi-line sections according to intersection point, every multi-thread segment record its start-stop intersection point number, Suo Youduo Line segment is included into two multi-line section groups of A, B respectively by its original affiliated area;
(d) multi-line section is filtered, A group multi-line sections are filtered using region B, the multi-line section inside the B of region is rejected, retains and is in B Outside multi-line section;B group multi-line sections are similarly filtered using region A;
(e) alternately take out multi-line section from two groups of AB according to start-stop intersection point number to be spliced, the new contour line of formation;
(f) union recombination region C is constructed, the contour line that step (e) is obtained adds this region, the contour line mark of arranged counterclockwise Border is designated as, the contour line of arranged clockwise is labeled as cavity;
(g) all outermost layer boundary contours in recombination region C are obtained, atural object boundary line is used as.
6. the Objects extraction method of inclination measurement scene according to claim 5, it is characterised in that:In the step (4), For a triangular facet to be split, the intersection point of triangle and current outline line is calculated first, this triangle is ignored if without intersection point Shape;Contour line is split as by some multi-line sections according to intersection point, retains the multi-line section in the range of triangle, deletes other multi-line sections; The summit for some multi-line sections that 3 summits of former triangle and step (3) are obtained reconfigures the triangulation network, it is ensured that each is multi-thread The side of Duan Zuowei triangles is appeared in the triangulation network of neotectonics, and these broken lines can be divided into some sections, but can not have Missing;After the completion of triangle subdivision, former triangle is deleted from the scene triangulation network, instead triangulation generation is all new Triangle.
7. the Objects extraction method of inclination measurement scene according to claim 6, it is characterised in that:In the step (4), Each triangle in being grouped for scene pel, calculates triangle horizontal center point first, by the flat of Atria summit Areal coordinate (XY) is separately summed again divided by 3, obtains triangle center point coordinates;Secondly judge triangle center point whether in atural object wheel In wide;Atural object pel is finally separating, if current triangle is in outline polygon, it is grouped from the landform of world subdivision It is middle to remove, and be added in current atural object packet, if not in outline polygon, ignored.
8. the Objects extraction method of inclination measurement scene according to claim 7, it is characterised in that:In the step (5), The primitive information of atural object, atural object primitive information is obtained from three-dimensional scenic to be needed to specify the accuracy class of atural object ID and model, Inquired about step by step since root node, the node until reaching given level, if node primitive data is not loaded, from corresponding number Loaded according to file, the pel group specified is obtained according to ID, be packaged as the form needed, root node is returned to step by step.
9. the Objects extraction method of inclination measurement scene according to claim 7, it is characterised in that:In the step (5), Ray collision detection is:Mouse click location in 3-D graphic window is converted to the ray in three dimensions first, then will Atural object packet in ray and inclination measurement scene in each visible elements carries out collision detection, if collision distance is less than currently most Small collision distance, then record the atural object ID belonging to the triangular facet of current point and hit, having handled all visible elements must To the inclination measurement atural object of hit.
10. the Objects extraction method of inclination measurement scene according to claim 7, it is characterised in that:The step (5) In, distance is calculated as:The distance between the electrical bodies such as power line and atural object are calculated, is calculated using the full accuracy pel of atural object, For each atural object, its full accuracy grade triangle network data is obtained, the distance of these triangular facets and electric wire is calculated successively and is taken Minimum value.
CN201710350830.0A 2017-05-18 2017-05-18 A kind of Objects extraction method of inclination measurement scene Pending CN107103603A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710350830.0A CN107103603A (en) 2017-05-18 2017-05-18 A kind of Objects extraction method of inclination measurement scene

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710350830.0A CN107103603A (en) 2017-05-18 2017-05-18 A kind of Objects extraction method of inclination measurement scene

Publications (1)

Publication Number Publication Date
CN107103603A true CN107103603A (en) 2017-08-29

Family

ID=59668957

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710350830.0A Pending CN107103603A (en) 2017-05-18 2017-05-18 A kind of Objects extraction method of inclination measurement scene

Country Status (1)

Country Link
CN (1) CN107103603A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108038900A (en) * 2017-12-06 2018-05-15 浙江科澜信息技术有限公司 Oblique photograph model monomerization approach, system and computer-readable recording medium
CN109064482A (en) * 2018-08-16 2018-12-21 武大吉奥信息技术有限公司 A kind of method and device obtaining outline of house automatically in three-dimensional tilt photographed data scene
CN109492606A (en) * 2018-11-26 2019-03-19 上海同繁勘测工程科技有限公司 Multispectral vector picture capturing method and system, three dimensional monolithic method and system
CN110648401A (en) * 2019-08-30 2020-01-03 贝壳技术有限公司 Method and device for unitizing oblique photography model, electronic equipment and storage medium
CN112330791A (en) * 2020-09-18 2021-02-05 广州中望龙腾软件股份有限公司 Three-dimensional model creation method, intelligent terminal and storage device
CN113593020A (en) * 2021-08-10 2021-11-02 四川易方智慧科技有限公司 Large-scale three-dimensional city scene generation method based on ArcGIS

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101976452A (en) * 2010-08-23 2011-02-16 武汉大学 Integrated filtering method of airborne laser scanning spot clouds based on contour line cluster analysis
US20130094768A1 (en) * 2011-10-14 2013-04-18 Cywee Group Limited Face-Tracking Method with High Accuracy
CN104050474A (en) * 2014-06-10 2014-09-17 上海海洋大学 Method for automatically extracting island shoreline based on LiDAR data
CN104200212A (en) * 2014-06-25 2014-12-10 西安煤航信息产业有限公司 Building outer boundary line extraction method based on onboard LiDAR (Light Detection and Ranging) data
CN104463857A (en) * 2014-11-27 2015-03-25 南宁市界围工程咨询有限公司 Extracting method for structure data in geographic data of oblique aviation photography
CN104484668A (en) * 2015-01-19 2015-04-01 武汉大学 Unmanned aerial vehicle multi-overlapped-remote-sensing-image method for extracting building contour line
CN106251331A (en) * 2016-07-15 2016-12-21 北京道亨时代科技有限公司 The extracting method of atural object in a kind of inclination measurement scene

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101976452A (en) * 2010-08-23 2011-02-16 武汉大学 Integrated filtering method of airborne laser scanning spot clouds based on contour line cluster analysis
US20130094768A1 (en) * 2011-10-14 2013-04-18 Cywee Group Limited Face-Tracking Method with High Accuracy
CN104050474A (en) * 2014-06-10 2014-09-17 上海海洋大学 Method for automatically extracting island shoreline based on LiDAR data
CN104200212A (en) * 2014-06-25 2014-12-10 西安煤航信息产业有限公司 Building outer boundary line extraction method based on onboard LiDAR (Light Detection and Ranging) data
CN104463857A (en) * 2014-11-27 2015-03-25 南宁市界围工程咨询有限公司 Extracting method for structure data in geographic data of oblique aviation photography
CN104484668A (en) * 2015-01-19 2015-04-01 武汉大学 Unmanned aerial vehicle multi-overlapped-remote-sensing-image method for extracting building contour line
CN106251331A (en) * 2016-07-15 2016-12-21 北京道亨时代科技有限公司 The extracting method of atural object in a kind of inclination measurement scene

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
李乐林: ""基于等高线族的机载LiDAR数据建筑物三维模型重建方法研究"", 《中国博士学位论文全文数据库基础科学辑》 *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108038900A (en) * 2017-12-06 2018-05-15 浙江科澜信息技术有限公司 Oblique photograph model monomerization approach, system and computer-readable recording medium
CN109064482A (en) * 2018-08-16 2018-12-21 武大吉奥信息技术有限公司 A kind of method and device obtaining outline of house automatically in three-dimensional tilt photographed data scene
CN109064482B (en) * 2018-08-16 2021-10-22 武大吉奥信息技术有限公司 Method and device for automatically acquiring house outline in three-dimensional oblique photography scene
CN109492606A (en) * 2018-11-26 2019-03-19 上海同繁勘测工程科技有限公司 Multispectral vector picture capturing method and system, three dimensional monolithic method and system
CN110648401A (en) * 2019-08-30 2020-01-03 贝壳技术有限公司 Method and device for unitizing oblique photography model, electronic equipment and storage medium
CN110648401B (en) * 2019-08-30 2023-10-20 贝壳技术有限公司 Oblique photography model singulation method, oblique photography model singulation device, electronic equipment and storage medium
CN112330791A (en) * 2020-09-18 2021-02-05 广州中望龙腾软件股份有限公司 Three-dimensional model creation method, intelligent terminal and storage device
CN113593020A (en) * 2021-08-10 2021-11-02 四川易方智慧科技有限公司 Large-scale three-dimensional city scene generation method based on ArcGIS
CN113593020B (en) * 2021-08-10 2023-05-23 四川易方智慧科技有限公司 ArcGIS-based large-scale three-dimensional urban scene generation method

Similar Documents

Publication Publication Date Title
CN107103603A (en) A kind of Objects extraction method of inclination measurement scene
CN105513127B (en) Shaft regularization three-dimensional modeling method and system based on density peaks cluster
CN106815847B (en) Trees dividing method and single tree extracting method based on laser radar point cloud
CN109242862B (en) Real-time digital surface model generation method
CN106251331A (en) The extracting method of atural object in a kind of inclination measurement scene
CN113034689B (en) Laser point cloud-based terrain three-dimensional model, terrain map construction method and system, and storage medium
CN108734728A (en) A kind of extraterrestrial target three-dimensional reconstruction method based on high-resolution sequence image
CN104463164B (en) It is a kind of based on umbrella frame method and crown height than trees canopy structure information extracting method
CN106780524A (en) A kind of three-dimensional point cloud road boundary extraction method
CN107025685A (en) Airborne building summit cloud modeling method under topology ambiguity
CN107146280A (en) A kind of point cloud building method for reconstructing based on cutting
CN106570468A (en) Method for reconstructing LiDAR original point cloud building contour
CN103871102B (en) A kind of road three-dimensional fine modeling method based on elevational point and road profile face
CN106340061B (en) A kind of mountain area point cloud filtering method
CN103942838A (en) Point cloud data based single tree three-dimensional modeling and morphological parameter extracting method
CN105354883A (en) 3ds Max fast and precise three-dimensional modeling method and system based on point cloud
CN109118588B (en) Automatic color LOD model generation method based on block decomposition
CN115205690B (en) Method and device for extracting street tree in monomer mode based on MLS point cloud data
CN108198194B (en) Oblique photography single object extraction method based on lateral multi-plane cutting
CN104751479A (en) Building extraction method and device based on TIN data
CN110032771A (en) A kind of accurate method of cutting out of open coal mine DEM for taking local detail feature into account
CN111754618A (en) Object-oriented live-action three-dimensional model multilevel interpretation method and system
CN112241661A (en) Urban ground feature fine classification method combining airborne LiDAR point cloud data and aerial image
CN109146990A (en) A kind of calculation method for building profile
CN109872390A (en) A kind of method for organizing of magnanimity threedimensional model

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20170829