CN106599044A - Recognition and processing method for road network target information - Google Patents

Abstract

The embodiment of the present invention discloses a method for identifying and processing road network target information. The processing method includes: A. searching for deletion points on the arc segment of the spatial data road network; B. searching for the deletion points found in step A Delete; C, find the projection points and fitting points on the arc segment of the spatial data road network; D, cluster the found projection points and fitting points into each specified processing unit; E, group the Projection points and fitting points are respectively subjected to projection processing and fitting processing. From the above, the embodiment of the present invention is beneficial to repair and accurately reflect the spatial relationship between entities in the road network of the reservoir body, and form a logically more complete topology structure to meet the requirements of automatic cartography synthesis.

Description

A method for identifying and processing road network target information

technical field

The invention relates to the field of geographic information system cartography synthesis, in particular to a method for identifying and processing road network target information for automatic computer cartography synthesis.

Background technique

The map is a tool for people to understand and transform the world. Its appearance has greatly improved people's understanding of objectively existing things and their surrounding environment. GIS is based on digitized maps. With the continuous development of the economy, people have higher and higher requirements for the richness of geographic information spatial data at different scales, so the database often needs to store geographic information spatial data of various scales as richly as possible, such as: national basic geographic information system. On the basis of the existing basic geographic database, how to quickly and faithfully derive any small-scale map database from the large-scale map database has become a research hotspot in modern cartography.

The road network is the most basic geographical element in map expression, an important part of the national spatial information infrastructure framework data, and a spatial division line used to divide other artificial geographical elements on the map. General map elements. At present, the expression of any form of map depends on the road network, which is the most important type of linear elements in the database. The quality of automatic cartographic synthesis depends largely on the automatic synthesis of road networks. Synthesis is very necessary and critical.

Contents of the invention

In view of this, the main purpose of the present invention is to propose a method for identifying and processing road network target information, aiming at the problem of spatial relationship errors in the road network data of the existing urban large-scale map database, by analyzing the characteristics of the road network topological relationship Target information is identified and processed to repair and accurately reflect the spatial relationship between road network entities in the database, and form a more complete topology to meet the needs of automatic cartography synthesis.

An embodiment of the present invention provides a method for identifying road network target information, including the following steps:

A. Establish an arc segment R tree for the arc segment of the spatial data road network;

B. Establishing an array for each paired arc segment in the pre-intersected arc segment tree found;

C, for any pair of arcs in the array and any component node on the first arc of any pair of arcs, perform the following steps:

c1, judging whether a component node on the first arc segment in the paired arc segment is on the second arc segment in the paired arc segment;

c2, if not, then the line segment that every two adjacent points on this second arc segment is formed line segment R tree is established;

c3. Search to obtain the line segment within the specified first threshold range outside the outer box of the current component node;

c4, taking the component node on the first arc as the center point, and making a buffer circle with the specified second threshold as the radius;

c5. Determine whether the buffer circle intersects the line segment obtained by the search; if so, record the intersecting line segment, the center point, the first arc segment and the second arc segment;

c6. Identify the type of the center point according to the relationship between the distance between the center point and the two ends of the intersecting line segment and the specified third threshold.

From the above, the automatic identification of road network target information is realized through the above method.

Preferably, said step c6 includes:

When it is judged that the distance between the center point and one of the two ends of the intersecting line segment is less than the third threshold, and the point of the two ends is located on the first arc segment, and when the distance between the center point and the two ends When the distance between the fitting point after fitting of any point and the next point of the center point in the first arc segment is less than the third threshold, the center point is a deletion point.

From the above, the recognition of the deleted points in the road network is realized.

Preferably, said step c6 includes:

When it is judged that the distances between the center point and the two ends of the intersecting line segment are greater than the third threshold, the center point is a projected point.

From the above, the recognition of projected points in the road network is realized.

Preferably, said step c6 includes:

or

When the distances between the center point and the two ends of the intersecting line segment are less than the third threshold;

Then the center point is the fitting point.

From the above, the identification of fitting points in the road network is realized.

Based on the above identification method of road network target information, an embodiment of the present invention also provides a processing method of road network target information, including the following steps:

D. Find the deletion point on the arc segment of the spatial data road network;

E, the deletion point found in step A is deleted;

F. Find projection points and fitting points on the arc segment of the spatial data road network;

G. Clustering the found projection points and fitting points;

H. Perform projection processing and fitting processing on the clustered projection points and fitting points respectively.

From the above, the processing of the road network target information is realized through the above method.

Preferably, said step G includes:

Use any found projection point or fitting point as the center of the circle, and draw a buffer circle with the specified fourth threshold as the radius;

The projection points and fitting points within the fourth threshold range are clustered into designated processing units.

From the above, the clustering of projection points and fitting points is realized.

Preferably, said step H includes:

H1, sequentially determine whether there are projection points in each processing unit;

H2, when judging that there are projection points, further judge whether the number of projection points is 1;

H3. When it is judged that the number of projection points is 1, perform projection processing on the projection point;

H4, when judging that the number of projected points is not 1, further judge whether the line segments of projected arcs are the same;

H5. When it is judged that the line segments of the projection arc are the same, each projection point is fitted to obtain the fitting point, and then the fitting point is projected to the projection arc;

H6. When it is judged that the line segments of the projected arcs are not the same, each projected point is subjected to fitting processing to obtain the fitted point, and then the fitted point is found to be a projected line segment closest to the fitted point. projection processing;

H7. When it is judged in step H1 that there is no projection point in the processing unit, further judge whether there is a fitting point; when it is judged that there is a fitting point, perform fitting processing.

From the above, the projection processing and fitting processing of the clustered projection points and fitting points are realized.

To sum up, this application helps to repair and accurately reflect the spatial relationship between road network entities in the database through the identification and processing of target information, that is, through the identification and processing of deleted points, projected points, and fitted points. Form a more complete topology to meet the needs of automatic cartographic synthesis.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present invention. For those skilled in the art, other drawings can also be obtained according to these drawings without any creative effort.

Fig. 1 is the gap and fragment type figure that exist in the road network data of reservoir body of the present invention;

Fig. 2 is a classification diagram of road network target information of the present invention;

Fig. 3 is a step diagram of the road network target information identification method of the present invention;

Fig. 4 is a step diagram of the method for processing road network target information of the present invention;

Fig. 5 is a step diagram of the method for clustering and processing road network target information of the present invention;

Fig. 6 is a comparison chart of the experimental test results of the present invention.

detailed description

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

To accomplish the task of automatic cartographic generalization, the key lies in the consistency of spatial data. Topological relationship is an important content of spatial data consistency. This invention takes the current urban large-scale map database as the research object, and finds that there are some common problems in the topological relationship of database data. For example, although the planar objects between different layers are visually It meets the needs of cartography, but there are actual gaps and fragments, but they cannot be automatically identified and repaired in the existing technology, and it is difficult to meet the needs of automatic cartography synthesis when considering the spatial constraints of elements.

In order to overcome the deficiencies in the prior art, the embodiment of the present application aims at the phenomenon that there are gaps and fragments in the topological relationship of the road network entities in the urban large-scale database. A method for object classification and identification and corresponding processing, which realizes the automatic restoration of spatial relationships and effectively ensures the topological integrity of road network spatial data.

There are gaps and fragments in the topological relationship of the spatial entity in the urban large-scale database described in the present invention. As shown in FIG. 1, the gaps and fragments are roughly divided into three types, intersecting, separating and interweaving. in:

Description 1. Topological attributes. When the scale of the map changes, the size and shape of the map will change accordingly, and some properties of the map graphics will also change accordingly, such as the length, area, angle and relative distance between map graphics. However, certain graphic properties will not change, such as the adjacency, inclusion, intersection, and geometric types of elements (such as point, line, and surface types) of map graphics. These properties that remain unchanged in the continuous change of the graph are called topological properties.

Description 2. Topological elements. There are many kinds of entities in geographic space, and their shapes are also varied, but they are generally represented by three types of features on the map, namely, point features, line features and area features. In a two-dimensional plane, they can correspond to three basic graphic elements, namely nodes, arcs and regions, and these three graphic elements are called topological elements.

Note 2.1 Nodes include isolated points, endpoints of arcs, connection points of arcs, interior points of polygons, boundary points of polygons, etc.

Note 2.2 An arc segment refers to an ordered line segment between two nodes, and the two nodes of an arc segment can be different nodes or the same node.

Note 2.3 Area refers to a polygonal area surrounded by multiple closed arc chains, which can be represented by polygonal graphics.

As shown in Figure 2, it is a schematic diagram of the classification of spatial entity target information. In the figure, the identified target information is divided into three categories. The three types of target information are specifically: the first type of target information is the deletion point, the second type of target information is the projection point, and the third type of target information is the fitting point. in:

Description 1. Delete point. As shown in Figure 2(a), arc A and arc B intersect at point O, the distance from point P on arc A to arc B is less than the threshold, and the distance between point P and point O is less than the threshold, then The point P is called the deletion point.

Description 2. Projection point. As shown in Figure 2(b), arcs A and B do not intersect, the distance between point P on arc A and arc B is less than the threshold, but the distance between P and point O and point O' of arc B is greater than the threshold , then the point P is called the projection point.

Description 3. Fitting points. As shown in Figure 2(c), arcs A and B do not intersect, and the distance between point P on arc A and point O on arc B is less than the threshold, then point P and point O are called fitting points .

Description 4. Basic processing model. The basic processing model is divided into three types. The first one is to delete the point P directly, such as the point P in Figure 2(a); the second one is to project the point P to the projection point P' of the arc segment. , insert the projection point P' on the arc, and move the point P to the projection point P', as shown in Figure 2(b). Fitting process, fit a new point P' or select a point P' with the largest weight, and then move all points to P', such as point P and point O in Figure 2(c).

(1) Projection point calculation formula

First, calculate the coefficient k of the straight line: Let the starting point and end point of the straight line be A(x ₁ , y ₁ ), B(x ₂ , y ₂ ) respectively, a point outside the straight line be C(x ₀ , y ₀ ), and the vertical foot be D ; and let k=|AD|/|AB|. but And because of so, so Bring in the coordinates, that is,

Then find the projection point D coordinates (x, y):

X＝X ₁ +k*(X ₂ *X ₁ ) (2)

y=y ₁ +k*(y ₂ *y ₁ ) (3)

(2) Fitting point calculation formula

Wherein, ( _xi , y _i ) (i=0, 1, 2, ..., n) are coordinates of all fitting origins, and n is the number of fitting origins.

As shown in Figure 3, it is a step diagram of the road network target information identification method of the present invention. In order to quickly find the target information, the R tree of the arc minimum circumscribed rectangular frame is established, and the paired arc segments that may be the target information are found. Then make a detailed judgment on the points on the paired arcs to improve the efficiency of the algorithm. Among them, the minimum circumscribed rectangle refers to the maximum range of several two-dimensional shapes (such as points, straight lines, polygons) represented by two-dimensional coordinates, that is, the maximum abscissa, minimum abscissa, maximum The ordinate, minimum ordinate defines the bounding rectangle. The specific steps of road network target recognition are as follows:

S301. Establish an arc segment tree. Specifically, an arc segment R-tree of the smallest circumscribed rectangular frame of the arc segment arc of all the spatial data is established.

S302. The arc segments that may intersect in all the arc segments arc in the arc segment R tree are paired to obtain each paired arc segment, specifically: the outer box of each arc segment arc in all arc segments arc is outward Expand the range of a threshold, search in the created arc segment R tree, pair with the searched arc segments that may intersect with it, and put each paired arc segment into the array that the arc segments may intersect.

S303. Select a pair of arc segments arcA and arcB from the array.

S304. Select a point P on the arcA of the pair of arcs.

S305. Determine whether the current point P is on arcB; if yes, execute S306; if not, execute S312.

S306. If the current node P is on arcB, it indicates that arcA intersects arcB, and the arcA and arcB are identified as the intersecting result, and the processing of the point ends;

S307 , judging whether there is a P point on the current arc segment arcA that has not been judged and processed in S305 ; if yes, go to S308 ; if not, go to S309 .

S308, select an unprocessed point P on the current arc segment arcA, and return to execute S305;

S309 , judging whether each paired arc segment has a paired arc segment that has not undergone the selection process in S303 ; if yes, execute S310 ; if not, execute 311 .

S310. Select a pair of arc segments not selected in S304, and return to execute S304.

S311, the processing ends.

S312. When it is judged in step S305 that the current point P is not on arcB, perform arc recognition. Build an R tree for the outer box of the line segment formed by every two adjacent points on the arcB of the current paired arc segment.

S313 , expand the outer frame of the point P on the current arcA to a threshold range to search, and judge whether the line segment can be searched within the threshold range. If yes, execute S314; if not, return to execute S307.

S314, taking the current point P as the center point, and using the specified first threshold as the radius to make a buffer circle;

S315, judging whether the buffer circle intersects the line segment searched in S313; if not, execute S316; if yes, execute S317.

S316, do not perform record processing.

S317, record the intersecting line segment, the center point P, the current arcA and the current arcB;

S318, further identifying and distinguishing which type of the three types of target information the center point P belongs to.

Specifically, through the relationship between the distances PA, PB and the threshold DistanceEpsilon between the point P and the endpoints A and B of the line segment AB to identify and distinguish which of the three types of target information the point P belongs to:

1. When PA is less than DistanceEpsilon and point A is on arcA and the distance between the fitting point of P and A and the next point of point P in arcA (in the direction of AP) is less than DistanceEpsilon, then point P is the deletion point;

When PB is less than DistanceEpsilon and point B is on arcA and the distance between the fitting point of P and B and the next point of point P in arcA (in the direction of BP) is less than DistanceEpsilon, then point P is the deletion point;

2. When PA is greater than DistanceEpsilon and PB is greater than DistanceEpsilon, point P is the projection point;

3. When PA is less than or equal to DistanceEpsilon and PB is greater than DistanceEpsilon, then point P is the fitting point;

When PA is greater than DistanceEpsilon and PB is less than or equal to DistanceEpsilon, point P is the fitting point;

When PA is less than DistanceEpsilon and PB is less than DistanceEpsilon, point P is the fitting point.

Return to execute S307. The recognition process is completed until all points P of all paired arc segments are determined.

As shown in Figure 4, based on the above road network target information identification method, the present invention also provides a road network target information processing method, the specific steps are as follows:

Step 401, find the deletion point. For the identification of spatial data target information, all arcs are searched for the above-mentioned deletion points and recorded;

In step 402, delete points are processed. The processing method for the deleted point is to directly delete the deleted point on the arc segment; repeat 401 until the first type of point cannot be found, that is, the deleted point;

Step 403, find projection points and fitting points. For the identification of spatial data target information, search for the above-mentioned projection points and fitting points for all arc segments, and record them if there are; if not, the algorithm ends and the processing is completed;

Step 404, clustering the projection points and fitting points. Put the projected and fitted points within the threshold range into one processing unit.

Step 405, classification judgment. Judging whether there are projection points in all processing units, if there are projection points, then execute step 406; if there is no projection point, then execute step 411.

Step 406, judging the number of projection points in the processing unit, when it is judged that the number of projection points is 1, execute step 407, and perform projection processing; when it is judged that the number of projection points is not 1, execute step 408.

Step 407, perform projection processing.

Step 408, judging whether the line segments of the projected arcs are the same; when judging that the line segments of the projected arcs are the same, go to step 409; when judging that the line segments of the projected arcs are not the same, go to step 410.

In step 409, the projection processing is performed after fitting the multiple projection points.

Specifically, multiple projection points are fitted to a new point P', and the new point P' is projected onto the line segment, and a point P" is inserted in the middle of the projected line segment, and all processing units are updated about the arc segment of the insertion point Index information; and return to step 403.

Step 410, after performing fitting processing on multiple projection points, find a projection line segment closest to the fitting point for projection processing, specifically, fit multiple projection points to a new point P', and search for the new point P' P' is the closest line segment, and project the new point P' to the closest line segment, and insert the point P" in the middle of the projected line segment, update the index information of all processing units on the insertion point arc segment; and return to execute Step 403.

Step 411, judging whether there are fitting points in each processing unit. If yes, execute step 413 . If not, return to step 403.

Step 412, perform fitting processing on all the fitting points in the processing unit to fit a new point P", or select a point P" with the largest weight, then move all points to P", and return to step 403.

As shown in FIG. 5 , for the above step 404 , the projection points and fitting points are clustered. The embodiment of the present invention also provides a target information clustering algorithm. According to the spatial relationship of the target information, the target information is clustered into each processing unit. The specific steps are as follows (because the deletion point does not need to be clustered, the target information clustering only considers projected and fitted points):

Step 501. Select a point P that has been identified in S301-318 (the point P in this embodiment refers to the projection point and the fitting point).

Step 502, create a new processing unit including the current point P.

Step 503 , expand the outer bounding box of the current point P by a threshold range, and put the projected points and/or fitted points within the range into the R tree.

Step 504, judge one by one whether other points P identified in S301-318 are in the R-tree, if not, execute step 505; if yes, execute step 506.

Step 505 , select a point P not in the R-tree; return to step 502 .

Step 506, put the point P in the R-tree determined in step 504 into the newly created processing unit in step 502.

Step 507, judging whether the identified points P in S301-318 are all put into the newly created processing unit in step 502. If yes, go to step 508 ; if not, go back to step 504 .

Step 508, the clustering process is completed.

As shown in Figure 6, it is a comparison chart of experimental test results of the present invention. This algorithm has done a large number of tests on the urban road network data of 650 square kilometers in Chengdu. The given experimental threshold is 0.1 meters. The processing results obtained are as follows:

The lines with larger width in Fig. 6(a) and Fig. 6(b) represent the topological arc of the road before processing, and the lines with smaller width represent the topological arc of the road after processing;

Figure 6(c) and Figure 6(d) are the junction of the road network intersection and the road surface, Figure 6(c) is the original data, and Figure 6(d) is the processed data.

Figure 6(e) and Figure 6(f) are road network intersections, Figure 6(e) is the topology of the original data, and Figure 6(f) is the processed topology after deleting redundant points.

The experimental test environment is a single PC, the Windows version is Windows XP, the system type is 32-bit operating system, the CPU is Intel Core2 Quad Q8400, the main frequency is 2.66GHz, the memory (RAM) is 3.25GB, and the total size of the hard disk is 60GB (solid state), the test data selects the basic data of 1:500 in the downtown area of Chengdu. Taking the road network data of 650 square kilometers as an example, in terms of efficiency: the total time for topology construction, topology preprocessing and topology reconstruction is 179 seconds, of which the topology preprocessing takes 123 seconds; in terms of accuracy: 13049 target information are identified, The recognition accuracy rate is 100%, see Table 1.

Table 1 Comparison table of road surface topology preprocessing results

The experimental results show that it is very effective to use topological relations to express and repair the spatial relations of spatial data. Based on the identification and processing model, the method of using topological relations to process spatial data satisfies the data consistency requirements of automatic cartography synthesis in terms of accuracy. ; In terms of efficiency, it meets the requirements of actual production practice.

In summary, the present invention has the advantage of realizing the automatic identification and repair of gaps and fragments existing in the existing urban large-scale map database road network entity spatial relationship, and meeting the comprehensive needs of road network automatic mapping when taking into account the spatial constraint relationship of elements . It overcomes that the existing road network synthesis methods are mostly based on road geometric features, semantic features, and attribute features for road simplification, and there is no comprehensive, systematic, and consistent processing method for road network topology features, especially wrong targets, which will affect the road network. The overall grasp of the road network structure in the synthesis makes the road network synthesis method unable to efficiently apply to the defects of multi-scale road cartography.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the scope of the present invention. within the scope of protection.

Claims (7)

1. a kind of recognition methodss of road network target information, it is characterised in that comprise the following steps：

A, the segmental arc of spatial data road network is set up into segmental arc R tree；

B, each the pre- intersecting paired segmental arc in the segmental arc tree for finding is set up into array；

C, for the array in arbitrary paired segmental arc and arbitrary paired segmental arc the first segmental arc on arbitrary composition node, hold Row following steps：

Whether the composition node in c1, the first segmental arc for judging in paired segmental arc is in the second segmental arc in the paired segmental arc；

C2, if it is not, then by each two in second segmental arc it is adjacent point composition line segment set up line segment R trees；

C3, search obtain the line segment in the range of the outside specified first threshold of the outsourcing frame of current composition node；

C4, using the composition node in first segmental arc as centre point, to specify Second Threshold to make buffering circle as radius；

C5, judge that the buffering circle searches for whether the line segment that obtains intersects with described；If so, then record intersecting line segment, centre point, First segmental arc and second segmental arc；

C6, intersected with this according to the centre point line segment two-end-point distance and specify the 3rd threshold value relation, recognize the center of circle The type of point.

2. method according to claim 1, it is characterised in that step c6 includes：

The distance of any in the two-end-point that the centre point intersects line segment with this is judged is less than the 3rd threshold value, and the two-end-point In the point be located in first segmental arc, and the match point after any point in the centre point and the two-end-point is fitted and the circle When the distance of the next point of the first segmental arc is less than three threshold values, then the centre point is deletion point to the heart o'clock.

3. method according to claim 1, it is characterised in that step c6 includes：

When judging that the centre point intersects the distance of the two-end-point of line segment with this and is both greater than three threshold values, then the centre point is to throw Shadow point.

4. method according to claim 1, it is characterised in that step c6 includes：

The distance of any in the two-end-point that the centre point intersects line segment with this is judged is less than the 3rd threshold value, and intersects with this When the distance of another point in the two-end-point of line segment is more than three threshold values；Or

When the distance that the centre point intersects the two-end-point of line segment with this is both less than three threshold values；

Then the centre point is match point.

5. a kind of processing method of the road network target information of the recognition methodss based on described in any one of claim 1-4, it is special Levy and be, comprise the following steps：

D, the deletion point searched in the segmental arc of spatial data road network；

E, by the deletion point deletion found in step A；

F, the subpoint and match point searched in the segmental arc of spatial data road network；

G, the subpoint for finding and match point are clustered；

H, the subpoint and match point after cluster is carried out into respectively projection process and process of fitting treatment.

6. method according to claim 5, it is characterised in that step G includes：

It is the center of circle by the arbitrary subpoint for finding or match point, to specify the 4th threshold value to draw buffering circle as radius；

By the subpoint in the 4th threshold range and match point cluster to designated treatment unit.

7. method according to claim 6, it is characterised in that step H includes：

H1, judge whether there is subpoint in each processing unit successively；

H2, when judge have subpoint when, whether the number for determining whether subpoint is 1；

H3, when the number for judging subpoint is for 1, the subpoint is carried out into projection process；

H4, when judging the number of subpoint not for 1, whether the line segment for determining whether to project segmental arc is same；

H5, when judge project segmental arc line segment for same when, each subpoint is done and obtain after process of fitting treatment match point, then general The match point to the projection segmental arc does projection process；

H6, when judge project segmental arc line segment not for same when, each subpoint is done and obtain after process of fitting treatment match point, then The match point is done into projection process to for finding apart from the nearest Projection Line Segment of the match point；

H7, when there is no subpoint during processing unit is judged in step H1, further determined whether match point；When judgement has plan During chalaza, process is fitted.