CN103914600B - A kind of road alignment method for evaluating similarity and device based on perspective view principle - Google Patents
A kind of road alignment method for evaluating similarity and device based on perspective view principle Download PDFInfo
- Publication number
- CN103914600B CN103914600B CN201410165776.9A CN201410165776A CN103914600B CN 103914600 B CN103914600 B CN 103914600B CN 201410165776 A CN201410165776 A CN 201410165776A CN 103914600 B CN103914600 B CN 103914600B
- Authority
- CN
- China
- Prior art keywords
- similarity
- line
- module
- road
- node
- 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.)
- Active
Links
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a kind of road alignment method for evaluating similarity and device based on perspective view principle.On the basis of road perspective graph model is set up, using two-dimentional similarity calculating method, point similarity, line similarity and comprehensive similarity are calculated respectively, with the similitude of quantitatively evaluating road alignment inducing effect.Include according to the device designed by the method for invention：Alignment elements extraction module, parameter setting module, data import modul, point similarity calculation module, line similarity calculation module, comprehensive similarity computing module, data memory module and the part of data outputting module eight.Wherein, point similarity calculation module, line similarity calculation module and comprehensive similarity computing module are core places.Based on the device invented, it is easy to implement using invented method the evaluation of different road alignment similarities, improves convenience and accuracy that linear inducing effect is calculated.
Description
Technical field
The present invention relates to a kind of road alignment method for evaluating similarity and device based on perspective view principle, belong to road friendship
Logical, technical field of data recognition.
Background technology
Correction in Road Alignment Design is to ensure that the key factor of road safety operation, different road alignment condition or varying environment
Lower same link alignment condition, the inducing action to driver's driving behavior has differences.In existing Specification,
The regulation of correlation is carried out primarily directed to the linear design objective of surface road, from objective design objective angle, for not
With the inducing effect difference caused by linear index, systematic quantitative research is not carried out.As highway layout becomes more meticulous reason
That reads gradually gos deep into, and under certain path space yardstick, which kind of linear inducing effect is even more ideal, will turn into designers and close
The focus of note.When answering the problem, it is necessary first to it is clear which kind of quantitative differences different alignment conditions have.In addition, with
Urbanization and the raising of motor vehicle level, many cities of China all occur in that Urban Traffic Jam Based, and this is existing
As, such as Beijing, Shanghai, Guangzhou ground especially prominent in megalopolis.In this context, building road tunnel turns into city
The inexorable trend of transport development.Yet with the particularity of road tunnel, the index of correlation of original surface road design is being answered
When using road tunnel design, it might not be applicable completely, and different inducing effect will be produced to driver.At present from road
The objective angle of road design objective, systematic quantitative research is there is no for this aspect.Therefore, scientific and rational quantization road
Linear difference, that is, study the similarity of road alignment, is designed for improving road minute design level and road tunnel
Security have very important significance.
The achievement in research on road alignment is mainly appeared in existing codes and standards at present, for example《Urban road work
Journey design specification》CJJ37-2012、《Urban road highway route design specification》CJJ193-2012、《Highway technical standard》JTG
B01-2003、《Specification of the highway route design》JTG D20-2006 etc., the leading indicator being related to includes：Road plane design objective
(circular curve, easement curve, maximum superelevation slope etc.), vertical alignment design index (maximum longitudinal grade, length of grade, resultant gradient, perpendicular song
Line etc.) and linear combination design principle in terms of.However, there is no quantitative research for different linear difference.On phase
Like the calculating of degree, correlative study is concentrated mainly on：The fields such as text retrieval, image retrieval, image comparison, images match, communication,
Also systematicness is not applied to highway layout field.In summary, from objective design objective angle, it there is no at present for Road
The index and method of shape similarity evaluation.
The content of the invention
It is an object of the invention to provide a kind of road alignment method for evaluating similarity and device based on perspective view principle,
With technical problems such as the objectivity of solving road alignment evaluation, systematicness, versatility and efficiencies.
The technical solution adopted for the present invention to solve the technical problems is：
A kind of road alignment method for evaluating similarity based on perspective view principle, its step is as follows：
Step one：Set up road perspective graph model；Based on road three-dimensional scene models, according to highway layout size, in phase
Under same engineer's scale and viewing angles, the perspective view for studying road is built；
Step 2：Alignment elements is extracted；Road perspective graph model based on structure, emphasis is sat to the X of road alignment, Y-axis
Mark carries out calculating extraction；Signified road alignment is the linear concept of broad sense herein, is linear, the abbreviation being made up of induction element
Induce line；Wherein, induction element can be driver including street lamp, greening installation in the range of roadmarking, path space etc.
The element of inducing action is provided；
Step 3：Parameter setting；Set parameter includes the unit distance values of road alignment, chooses some road alignment
The start node of contrast, the weight of X-axis similarity and Y-axis similarity are taken as basic induction line, for basic induction line selection
Weight；Wherein, the selection principle of basic induction line is：Choose the maximum conduct of Y value and induce line substantially；
Step 4：Similarity Measure；
(1) Similarity Measure is put；Using start node as the first of similarity comparison node, according to the unit spacing of setting
Value, with reference to the road alignment X, Y-axis coordinate initially obtained, is calculated to obtain inducing substantially on line and owned respectively based on linear interpolation method
Contrast X, the Y-axis coordinate of node；For a certain node on basic induction line, it will be to the air line distance for comparing induction line
The node and the minimum value for being compared all nodal distances on induction line；If the node apart from minimum value is comparing the linear of induction line
In the range of, then calculate to obtain and compare the X of the node, Y-axis coordinate on induction line, and obtained a little according to two-dimentional calculating formula of similarity
Similarity；Otherwise, the node is not present, it is impossible to obtain X, Y coordinate, then the point similarity of the node is 0；
(2) line Similarity Measure；After all point Similarity values are obtained, line is obtained using the calculating of arithmetic mean method similar
Angle value；
(3) comprehensive similarity is calculated；Obtaining after wired Similarity value, calculated using arithmetic mean method and obtain comprehensive phase
Like angle value, it can be used for the assessment of the linear similarity in two road or a plurality of road alignment similarity.
A kind of road alignment similarity evaluation device based on perspective view principle, including：Alignment elements extraction module, parameter
Setup module, data import modul, point similarity calculation module, line similarity calculation module, comprehensive similarity computing module, number
According to memory module and data outputting module；
Alignment elements extraction module, the road perspective view based on structure, Road is extracted under same ratio, same view angle
The X of shape, Y-axis coordinate；
Parameter setting module, sets the unit distance values of road alignment according to user's request, chooses some road alignment work
The start node of contrast is taken for basic induction line, for basic induction line selection, the weight and Y-axis similarity of X-axis similarity are set
Weight；
Data import modul, import the X of road alignment, Y-axis coordinate, the unit distance values of road alignment, substantially induction line,
The weight of the basic start node for inducing line, the weight of X-axis similarity and Y-axis similarity；
Point similarity calculation module, on the basis of inducing the start node of line substantially, X, Y-axis with reference to linear elements recognition
Coordinate, according to the unit distance values of the road alignment of setting, calculates and obtains institute X, Y in need for contrasting node on basic induction line
Axial coordinate；Based on this, for the individual node of basic induction line, searching is compared most short with the presence or absence of air line distance on induction line
Node, and the X of node, Y-axis coordinate on induction line are compared in acquisition, and a point similarity is calculated according to two-dimentional similarity calculating method；
Line similarity calculation module, two induction lines of result calculating or a plurality of induction line based on all node similarities
Similarity；
Comprehensive similarity computing module, result calculating two road or a plurality of road based on all induction line similarities
Comprehensive similarity；
Data memory module, realizes that each data processing stage data are preserved；
Data outputting module, according to user's request output node X, Y-axis coordinate, the unit distance values of road alignment, point phase
Like degree calculated value, line similarity value calculation or comprehensive similarity calculated value；
Above-mentioned module is mutually related according to the processing direction of evaluating data stream；Wherein, alignment elements extraction module is
Carry out the first module evaluated, it provides data basis for next stage Similarity Measure, and it has flowed directly into data importing
Module；In parameter setting module, the parameter according to necessary to user's request sets Similarity Measure, it has flowed directly into data
Import modul；In data import modul, the data for Similarity Measure are imported, its data flow point similarity
Computing module；In a similarity calculation module, based on two-dimentional similarity calculating method, the Similarity value of each node is obtained,
And data storage device is stored in, it is to carry out the basis that line Similarity Measure and comprehensive similarity are calculated next stage, and directly flow
To line similarity calculation module；In online similarity calculation module, based on the similarity value calculation of each node, line phase is integrated into
Like angle value, and data storage device is stored in, and has flowed directly into comprehensive similarity computing module；In comprehensive similarity computing module
In, comprehensive similarity value is obtained, preservation is realized respectively via data memory module and data outputting module and exports.
Advantages of the present invention is as follows：
Due to still lacking index and method for road alignment similarity evaluation at present, in order to contrast two roads exactly
The difference of route shape or a plurality of road alignment, provides quantization contrasting foundation for Correction in Road Alignment Design from now on, has invented institute of the present invention
A kind of the road alignment method for evaluating similarity and device based on perspective view principle stated.The user of this method and device is according to need
The linear inducing effect difference of same link under the inducing effect difference or varying environment that contrast different road alignments is sought, is based on
Perspective view extracts linear X, Y-axis coordinate, and incorporating parametric is set, and point similarity, line similarity and comprehensive similarity are calculated respectively,
It is used as the numerical indication for quantifying linear difference.The beneficial effects of the invention are as follows from application oriented angle, index of similarity
With easy quantization, the easily characteristic such as contrast, computational efficiency is improved, the convenience of appraisal is enhanced；Similarity Measure is covered
Point, line and comprehensive three aspects, systematicness are strong；In addition, the highly versatile of the present invention, it is adaptable to different road alignment phases
Like the evaluation of identical linear similarity under degree or varying environment.
Brief description of the drawings
Fig. 1 is the structure drawing of device of the present invention.
Fig. 2 is the device flow chart of the present invention.
Fig. 3-a are the schematic perspective views of the present invention.
Fig. 3-b are the induction elemental maps of the present invention.
Embodiment
Embodiment is given below to illustrate the embodiment of invention：
A kind of road alignment method for evaluating similarity based on perspective view principle of the present invention, is specifically included：Set up
Reuter's view model；Extract X, the Y-axis coordinate of road alignment；Arrange parameter, specify linear unit distance values, substantially induction line,
The weight of the basic start node for inducing line, the weight of X-axis similarity and Y-axis similarity；Propose two-dimentional Similarity Measure side
Method, calculates point similarity；Based on this, line similarity and comprehensive similarity are calculated, two road or many are evaluated from the angle of quantization
The linear similitude of bar road, is easy to designer to understand difference of the different road alignments in terms of inducing effect, available for from now on
The minute design of surface road and selection for the linear index of road tunnel provide reference.The invention mainly comprises following step
Suddenly：
1st, road perspective graph model is set up.
Road perspective view includes three kinds of perspective views：Perspective effect figure, schematic perspective view and perspective elemental map.Utilize conventional three
Virtual emulation modeling software is tieed up, such as UC-WinRoad, Creator, road three-dimensional scene models are set up, according to highway layout
Size, under identical engineer's scale and viewing angles, the interception image from road three-dimensional scenic forms perspective effect figure；Profit
Pattern vector processing is carried out to perspective effect figure with Adobe illustrator, schematic perspective view is formed；Enter for schematic perspective view
The stripping of row non-induced element, emphasis retains road alignment key element and obtains induction elemental map.Based on inducing elemental map, carry out
The quantitative analysis and differentiation of similarity.Herein signified induction element include roadmarking, it is the street lamp in the range of path space, green
The element of inducing action can be provided for driver by changing facility etc..
2nd, road alignment X, the extraction of Y-axis coordinate.
Road perspective graph model based on structure, emphasis carries out calculating extraction to the X of road alignment, Y-axis coordinate.Based on figure
Induction elemental map after shape vector quantization, obtains X, the Y-axis coordinate of road alignment.Signified road alignment is that broad sense is linear herein
Concept, is linear, the abbreviation induction line being made up of induction element.
3rd, parameter setting.
The unit distance values of set parameter including road alignment, choose some road alignment as basic induction line,
The weight of the start node of contrast, the weight of X-axis similarity and Y-axis similarity is taken for basic induction line selection.Wherein, lure substantially
The selection principle of wire is：Choose the maximum conduct of Y value and induce line substantially.
Five parameters are mainly set：
1) the unit distance values Δ d of road alignment：The numerical value is related to engineer's scale size, and numerical value is smaller, pair of acquisition
It is more than node.
2) line is induced substantially：Contrast two road or a plurality of road and induce the Y-axis coordinate of line in road the same side, selection is lured
Wire Y value maximum is basic induction line, and remaining is to compare induction line.Assuming that a plurality of road for participating in contrast is R (a), R
(b) ..., R (n), their corresponding certain side induction lines are R (a) respectivelyi、R(b)i、……、R(n)i, their linear length
Degree is d (a) respectivelyi、d(b)i、……、d(n)i.For the ease of the explanation of computational methods, it is assumed that induction line R (a)iTo lure substantially
Wire, R (b)iTo compare induction line.
3) start node：For basic induction line R (a)i, choose certain point and be used as start node P (a)i1For similarity
Contrast.
4) the weight ρ of X-axis similarityX, general recommendations value 0.5.
5) the weight ρ of Y-axis similarityY, general recommendations value 0.5.
ρXYρY=1
4th, Similarity Measure is put
Point similarity is calculated using two-dimentional similarity calculating method.
1) using start node as the first of similarity comparison node P (a)i1, according to the unit distance values Δ d of setting, knot
The road alignment X initially obtained, Y-axis coordinate are closed, calculates obtain inducing the X of all contrast nodes, Y-axis on line to sit substantially respectively
Mark.Assuming that induction line R (a) substantiallyiIt is upper contrast node number be：Then own on basic induction line
The coordinate for contrasting node is { X (a)i1, Y (a)i1, { X (a)i2, Y (a)i2... ..., { X (a)iW, Y (a)iW}。
2) with i-th basic induction line R (a)iOn w-th of node P (a)iw{X(a)iw, Y (a)iwIt is research object, it is false
If node P (a)iwLine R (b) is induced to comparingiAir line distance be DZ (a-b)iwIf this is apart from corresponding node coordinate P (b)ih
{X(b)ih, Y (b)ih}∈{X(b)i, Y (b)i, then obtain w on i-th bar of basic induction line using two-dimentional similarity calculating method
The point similarity of individual node：
SPiw=ρX×SP(X)iw+ρY×SP(Y)iw
In formula, SPiwFor the point similarity of w-th of node on i-th basic induction line；SP(X)iwFor i-th basic induction
Point similarity of w-th of node in X-axis on line；SP(Y)iwIt is similar in the point of Y-axis for w-th of node on i-th basic induction line
Degree；ρXFor the weight of basic induction line X-axis similarity；ρYFor the weight of basic induction line Y-axis similarity.
In formula, X is the length of induction elemental map horizontal boundary；X(a)iwFor w on i-th bar of basic induction line of road R (a)
The X-axis coordinate of individual node；X(b)ihCompare the X-axis coordinate of h-th of node on induction line for road R (b) i-th.
In formula, Y is the length of induction elemental map longitudinal boundary；Y(a)iwFor w on i-th bar of basic induction line of road R (a)
The Y-axis coordinate of individual node；Y(b)ihCompare the Y-axis coordinate of h-th of node on induction line for road R (b) i-th.
If comparing induction line R (b)iUpper all nodes induce line R (a) to basiciUpper P (a)iwDistance be all higher than straight line away from
From for DZ (a-b)iw, then the node coordinate P (b) with straight line beeline is illustratedih{X(b)ih, Y (b)ihIt is not belonging to { X (b)i,
Y(b)i, then on i-th of basic induction line w-th of node point similarity：
SPiw=0
It follows that to induce line R (a) substantiallyiUpper W node is reference point, compares induction line R with the nodal distance
(b)iAir line distance be the scale of measurement, then obtain compare induction line R (b)iUpper corresponding node coordinate { X (b)i1, Y (b)i1,
{X(b)i2, Y (b)i2... ..., { X (b)iH, Y (b)iH, and there is H≤W.Calculate obtained point similarity SPiw∈[0,1]。
5th, line Similarity Measure
The similarity of induction line is the integrated of multiple similarities, and computing formula is as follows：
In formula, SLiFor the similarity of i-th induction line；W is the number of contrast node.
6th, comprehensive similarity is calculated
Comprehensive similarity is the integrated of a plurality of line similarity, and computing formula is as follows：
In formula, S is comprehensive similarity；I is the bar number of induction line.
A kind of road alignment similarity evaluation device based on perspective view principle is devised according to the method for the present invention, is wrapped
Include：Alignment elements extraction module, parameter setting module, data import modul, point similarity calculation module, line Similarity Measure mould
Block, comprehensive similarity computing module, data memory module and the part of data outputting module eight are constituted.This eight parts are mutually interconnected
System, is combined together, and use in order according to the data flow in evaluation.Wherein, point similarity calculation module, line similarity
Computing module and comprehensive similarity computing module are the cores of the present apparatus.
It is a kind of road alignment similarity evaluation based on perspective view principle designed according to the inventive method shown in Fig. 1
The structure chart of device, by alignment elements extraction module, parameter setting module, data import modul, point similarity calculation module, line
Similarity calculation module, comprehensive similarity computing module, data memory module and the part of data outputting module eight are constituted.
The function of alignment elements extraction module is the road perspective view based on structure, is carried under same ratio, same view angle
Take X, the Y-axis coordinate of road alignment.
The function of parameter setting module is to set the unit distance values of road alignment according to user's request, choose some road
It is linear to take the start node of contrast as basic induction line, for basic induction line selection, the weight and Y-axis of X-axis similarity are set
The weight of similarity.
The function of data import modul is introduced into the X of road alignment, Y-axis coordinate, the unit distance values of road alignment, basic
Induce the weight of line, the start node of basic induction line, the weight of X-axis similarity and Y-axis similarity.
The function of point similarity calculation module is on the basis of inducing the start node of line substantially, with reference to linear elements recognition
X, Y-axis coordinate, according to the unit distance values of the road alignment of setting, calculating obtains induce in need contrast of institute on line to save substantially
The X of point, Y-axis coordinate.Based on this, for the individual node of basic induction line, searching compare on induction line with the presence or absence of straight line away from
From most short node, and the X of node, Y-axis coordinate on induction line are compared in acquisition, and a point phase is calculated according to two-dimentional similarity calculating method
Like degree.
The function of line similarity calculation module is that the result based on all node similarities calculates two induction lines or a plurality of
Induce the similarity of line.
The function of comprehensive similarity computing module is that the result based on all induction line similarities calculates two road or many
The comprehensive similarity of bar road.
The function of data memory module is to realize that each data processing stage data are preserved.
The function of data outputting module be according to user's request output node X, Y-axis coordinate, road alignment unit spacing
Value, point similarity value calculation, line similarity value calculation or comprehensive similarity calculated value.
This eight part is connected each other, is combined together and is used in order according to evaluation flow direction.Its running is as follows：When true
After the evaluation object for determining road alignment similarity, road perspective graph model is set up, road is obtained by alignment elements extraction module
X, Y-axis coordinate；Alignment elements extraction module Connecting quantity setup module and data import modul, in parameter setting module,
According to evaluation object, the relevant parameter of Similarity Measure is set, data import modul is connected；Data import modul tie point is similar
Computing module is spent, obtained point Similarity value is calculated according to two-dimentional similarity calculating method, and be stored in data memory module；Point phase
Like degree computing module connecting line similarity calculation module, calculated using arithmetic mean method and obtain line Similarity value, and be stored in data
Memory module；Line similarity calculation module connects comprehensive similarity computing module, is calculated using arithmetic mean method and obtains comprehensive phase
Like angle value；Comprehensive similarity module connects data memory module and data outputting module, and the comprehensive similarity value of acquisition is via number
The function of preserving and export is realized according to memory module and data outputting module.
It is the detail flowchart of apparatus of the present invention shown in Fig. 2.User is setting up road perspective view according to evaluation demand first
On the basis of model, by alignment elements extraction module, X, the Y-axis coordinate of road alignment are obtained, and sets up " initial X, Y-axis seat
Table indexOri " is marked, subsequently into parameter setting module, five parameters are inputted：Linear unit distance values, substantially induction line, rise
The weight of beginning node, the weight of X-axis similarity and Y-axis similarity；Alignment elements extraction module and parameter setting module all enter
Data import modul, initial X, Y-axis coordinates table indexOri and five parameters are imported to a similarity calculation module；Point is similar
Degree computing module is directed to the data that data import modul is imported, and recalculates X, the Y for obtaining and node being contrasted on basic induction line
Axial coordinate, and set up " it is basic to induce X, the Y-axis coordinates table indexOriBasCon " that node is contrasted on line, calculate basic induction line
Upper contrast nodal distance compares the air line distance of induction line, and sets up that " air line distance table linearDis ", on this basis, is searched
The comparison induction X of line corresponding node corresponding to rope air line distance, Y-axis coordinate, judge the X of corresponding node, Y-axis coordinate whether
In the linear range for comparing induction line；" compare the X of corresponding node, Y-axis coordinates table on induction line if so, calculating and setting up
InderOriCon ", is calculated using two-dimentional similarity calculating method and obtains a Similarity value；Otherwise, point similarity is 0；In this base
On plinth, set up " point Similarity value tables of data resultPoint ", and result is stored in data storage device and issued the user with a little
Similarity value calculates completion notice；Line similarity calculation module, will have a Similarity value to obtain line phase using arithmetic mean method
Like angle value, set up " line Similarity value tables of data resultLine ", and result is stored in data storage device and issued the user with
Line Similarity value calculates completion notice；Comprehensive similarity computing module, the wired Similarity value of institute is obtained using arithmetic mean method
Comprehensive similarity value, set up " comprehensive similarity Value Data table resultCom ", and by result be stored in data storage device and to
Family sends comprehensive similarity value and calculates completion notice；After the completion of above-mentioned flow, data storage device and data output device
The calculating process numerical value and result value of a similarity, line similarity and comprehensive similarity are carried out certainly according to user's request respectively
Dynamic storage and Formatting Output, mainly including nodes X, Y-axis coordinate, the unit distance values of road alignment, point similarity value calculation,
Line similarity value calculation or comprehensive similarity calculated value, and set up " dataOut ".
It is road perspective view model schematic shown in Fig. 3.Including schematic perspective view (3-a) and induction elemental map (Fig. 3-b).
Demand is evaluated according to user first and sets up road perspective effect figure；Perspective effect figure is carried out using Adobe illustrator
Pattern vectorization processing, forms schematic perspective view；The stripping of non-induced element is carried out for schematic perspective view, emphasis retains road alignment
Key element obtains induction elemental map.Based on inducing elemental map, the quantitative analysis and differentiation of similarity are carried out.
In order to which more intuitively explanation carries out the flow and result of road alignment similarity evaluation with the device, with identical
Illustrate the specific implementation process evaluated under alignment condition exemplified by surface road A and road tunnel B inducing effect.Particular content is such as
Under：
1st, the perspective graph model based on foundation, referring to table 1.Under same ratio chi and viewing angles, surface road A is extracted
X, Y-axis coordinate with induction line on the left of road tunnel B and right side induction line.
2nd, arrange parameter.
Five parameters are set：The unit distance values Δ d of road alignment, determine basic induction line, determine contrast start node,
The weight of X-axis similarity and the weight of Y-axis similarity, referring to table 2.
3rd, Similarity Measure is put
Referring to table 3, by taking left side induction line as an example, illustrate the calculating process of some similarities.With (0,35.454) for first
Node is contrasted, according to unit distance values 1, obtains and the X of all contrast nodes, Y-axis coordinate on line is induced on the left of surface road A.
Referring to table 4, all contrast nodes obtain straight line most short distance as research object using on surface road A left sides induction line
From corresponding X, Y-axis coordinate on lower road tunnel B left sides induction line, and judge whether the coordinate induces on the left of road tunnel B
In the linear range of line, if in linear scope, according to the Similarity value of the two-dimentional Similarity Measure node；Otherwise, should
The Similarity value of node is 0.
It is identical with the calculating process for calculating point similarity on left side induction line referring to table 5, calculate and obtain right side induction line
Point Similarity value.
4th, line Similarity Measure
Referring to table 6, left side, which is induced on line, has a Similarity value to take arithmetic average, obtains the similar of left side induction line
Angle value；Similarly, the Similarity value of right side induction line is obtained.
6th, comprehensive similarity is calculated
Left side induction line similarity is taken into arithmetic average with right side induction line similarity, comprehensive similarity value is obtained：
0.8732。
Table 1
Table 2
Table 3
Table 4
Table 5
Table 6
Claims (2)
1. a kind of road alignment method for evaluating similarity based on perspective view principle, its step is as follows：
Step one：Set up road perspective graph model；Based on road three-dimensional scene models, according to highway layout size, in identical
Under engineer's scale and viewing angles, the perspective view for studying road is built；
Step 2：Alignment elements is extracted；Road perspective graph model based on structure, X, Y-axis coordinate to road alignment are counted
Calculate and extract；Signified road alignment is the linear concept of broad sense herein, is linear, the abbreviation induction line being made up of induction element；
Wherein, induction element provides inducing action including the street lamp in the range of roadmarking, path space, greening installation for driver
Element；
Step 3：Parameter setting；Set parameter includes the unit distance values of road alignment, chooses some road alignment conduct
The basic weight for inducing line, the start node of contrast, the weight of X-axis similarity and Y-axis similarity being taken for basic induction line selection；
Wherein, the selection principle of basic induction line is：Choose the maximum conduct of Y value and induce line substantially；
Step 4：Similarity Measure；
（1）Point Similarity Measure；Using start node as the first of similarity comparison node, according to the unit distance values of setting,
With reference to the road alignment X, Y-axis coordinate initially obtained, calculated respectively based on linear interpolation method and obtain inducing all right on line substantially
X, Y-axis coordinate than node；For a certain node on basic induction line, it, to the air line distance for comparing induction line, will be this
Node and the minimum value for being compared all nodal distances on induction line；If the node apart from minimum value is comparing the linear model of induction line
In enclosing, then calculate to obtain and compare the X of the node, Y-axis coordinate on induction line, and a phase is obtained according to two-dimentional calculating formula of similarity
Like degree；Otherwise, the node is not present, it is impossible to obtain X, Y coordinate, then the point similarity of the node is 0；
（2）Line Similarity Measure；After all point Similarity values are obtained, calculated using arithmetic mean method and obtain line similarity
Value；
（3）Comprehensive similarity is calculated；Obtaining after wired Similarity value, calculated using arithmetic mean method and obtain comprehensive similarity
Value, the assessment for the linear similarity in two road or a plurality of road alignment similarity.
2. a kind of road alignment similarity evaluation device based on perspective view principle, it is characterised in that including：Alignment elements is extracted
Module, parameter setting module, data import modul, point similarity calculation module, line similarity calculation module, comprehensive similarity meter
Calculate module, data memory module and data outputting module；
Alignment elements extraction module, the road perspective view based on structure extracts road alignment under same ratio, same view angle
X, Y-axis coordinate；
Parameter setting module, sets the unit distance values of road alignment according to user's request, chooses some road alignment as base
This induction line, for basic induction line selection take the start node of contrast, the weight of X-axis similarity and the power of Y-axis similarity are set
Weight；
Data import modul, imports the X of road alignment, Y-axis coordinate, substantially the unit distance values of road alignment, induction line, basic
Induce the weight of start node, the weight of X-axis similarity and the Y-axis similarity of line；
Point similarity calculation module, on the basis of inducing the start node of line substantially, X, Y-axis with reference to linear elements recognition are sat
Mark, according to the unit distance values of the road alignment of setting, calculates and obtains the institute X in need for contrasting node, Y-axis on basic induction line
Coordinate；Based on this, for the individual node of basic induction line, find and compare on induction line with the presence or absence of air line distance most short section
Point, and the X of node, Y-axis coordinate on induction line are compared in acquisition, and a point similarity is calculated according to two-dimentional similarity calculating method；
Line similarity calculation module, result based on all node similarities calculates the similar of two induction lines or a plurality of induction line
Degree；
Comprehensive similarity computing module, the result based on all induction line similarities calculates the synthesis of two road or a plurality of road
Similarity；
Data memory module, realizes that each data processing stage data are preserved；
Data outputting module, according to user's request output node X, Y-axis coordinate, the unit distance values of road alignment, point similarity
Calculated value, line similarity value calculation or comprehensive similarity calculated value；
Above-mentioned module is mutually related according to the processing direction of evaluating data stream；Wherein, alignment elements extraction module is to carry out
The first module evaluated, it provides data basis for next stage Similarity Measure, and it has flowed directly into data import modul；
In parameter setting module, the parameter according to necessary to user's request sets Similarity Measure, it has flowed directly into data importing
Module；In data import modul, the data for Similarity Measure are imported, its data flow point Similarity Measure
Module；In a similarity calculation module, based on two-dimentional similarity calculating method, the Similarity value of each node is obtained, and deposit
Enter data storage device, be to carry out the basis that line Similarity Measure and comprehensive similarity are calculated next stage, and flow directly into line
Similarity calculation module；In online similarity calculation module, based on the similarity value calculation of each node, line similarity is integrated into
Value, and data storage device is stored in, and flowed directly into comprehensive similarity computing module；In comprehensive similarity computing module,
Comprehensive similarity value is obtained, preservation is realized respectively via data memory module and data outputting module and exports.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410165776.9A CN103914600B (en) | 2014-04-23 | 2014-04-23 | A kind of road alignment method for evaluating similarity and device based on perspective view principle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410165776.9A CN103914600B (en) | 2014-04-23 | 2014-04-23 | A kind of road alignment method for evaluating similarity and device based on perspective view principle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103914600A CN103914600A (en) | 2014-07-09 |
| CN103914600B true CN103914600B (en) | 2017-07-18 |
Family
ID=51040276
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410165776.9A Active CN103914600B (en) | 2014-04-23 | 2014-04-23 | A kind of road alignment method for evaluating similarity and device based on perspective view principle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103914600B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112530183B (en) * | 2020-10-27 | 2022-09-06 | 连云港杰瑞电子有限公司 | City trunk similarity measurement method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101922939B (en) * | 2009-06-11 | 2013-09-18 | 高德信息技术有限公司 | Map matching method and device in navigation process |
| JP5062498B2 (en) * | 2010-03-31 | 2012-10-31 | アイシン・エィ・ダブリュ株式会社 | Reference data generation system and position positioning system for landscape matching |
-
2014
- 2014-04-23 CN CN201410165776.9A patent/CN103914600B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN103914600A (en) | 2014-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105404938B (en) | A kind of line walking method for optimizing route shortening the patrolling transmission line time | |
| CN105205864B (en) | Geologic structure face threedimensional model method for automatic modeling and system based on multi-source data | |
| CN103065361B (en) | Three-dimensional island sand table implementation method | |
| CN104008252A (en) | Method for automatically generating plane-section drawing of underground cable duct lines | |
| CN107330139A (en) | Collision checking method based on BIM technology | |
| CN106407540A (en) | Underground pipeline auxiliary planning verification method based on three-dimensional GIS technology | |
| CN103970837B (en) | Discontinuous DEM classified manufacturing method based on urban land and vertical planning | |
| CN105355042A (en) | Road network extraction method based on taxi GPS | |
| US20170186225A1 (en) | Method and apparatus for generating a composite indexable linear data structure to permit selection of map elements based on linear elements | |
| CN111382472A (en) | Method and device for predicting shield-induced proximity structure deformation by random forest fusion SVM (support vector machine) | |
| CN107944203A (en) | A kind of visual architectural design method of wind-velocity streamline | |
| CN108959598A (en) | The method that CAD data based on GIS imported into SWMM | |
| CN107507188A (en) | A kind of method and device based on machine learning extraction image information | |
| CN108664705A (en) | A method of the simulation complicated landform roughness of ground surface based on OpenFOAM | |
| CN107895370A (en) | A kind of saddle point extracting method based on contour | |
| CN103530700B (en) | Urban distribution network saturation loading Comprehensive Prediction Method | |
| CN103426126A (en) | Generation method for segmental contact topological relationship of feeder line group of regional power grid | |
| CN107341598A (en) | Semantic and geometry calculations incorporated complex building specification accordance detection means and method | |
| CN103914600B (en) | A kind of road alignment method for evaluating similarity and device based on perspective view principle | |
| CN109299301A (en) | A kind of method for searching three-dimension model based on distribution of shapes and curvature | |
| CN103207923A (en) | Computer aided design method for vertical design of road plane intersection | |
| CN106202799A (en) | A kind of multipoint hoisting computational methods | |
| CN111199064A (en) | Method for generating three-dimensional center line of subway rail surface | |
| CN107507213A (en) | A kind of method and device based on end point extraction image information | |
| CN107169080A (en) | A kind of geospatial analysis system being combined based on GIS and spatial database |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |