CN110619134A - Integrated detection and restoration method for solving problem of flying spot and spot density of road network data - Google Patents

Abstract

The invention relates to the technical field of highway informatization, and discloses an integrated detection and restoration method for solving the problems of flying spots and spot density of road network data, which comprises the following steps: preparing data, unifying data formats, carrying out primary detection on the data, cleaning the data and warehousing the data; long straight line detection and repair; short straight line, flying point detection and preliminary processing, respectively calculating the distance between all adjacent nodes in the road section and the angle of included angles formed by all adjacent three nodes, judging as a short straight line if the distance is less than a threshold, only keeping the previous point if the distance is 0, judging as a flying point if the angle is less than the threshold, and directly deleting the flying point if the angle is too small; the short straight line and the flying spot are integrally processed, the short straight line needs to be comprehensively solved with the flying spot, and the problem of the short straight line is preferentially solved for the condition that the short straight line cannot be considered; elimination of the residual flying spot problem. The method obviously improves the efficiency of detecting and repairing the road network data line shape, and has higher accuracy and normalization.

Description

Integrated detection and restoration method for solving problem of flying spot and spot density of road network data

Technical Field

The invention relates to the technical field of highway informatization, in particular to an integrated detection and restoration method for solving the problems of flying points and point density of road network data.

Background

The construction of the rural highway is an important pillar and a key pivot of the village pleasure strategy. In order to accurately master the development condition, construction requirement and construction process of rural roads, from 2005, the organization of the transportation department develops special investigation work of national highway accessibility conditions of rural areas, and utilizes the GPS technology to acquire highway accessibility conditions of rural areas (towns) and villages built nationwide and acquire spatial data and attribute data of rural roads. In 2007, the transportation department established an annual updating system of the rural highway basic data and the electronic map, and required that each province reports the rural highway basic data and the annual updating data to the transportation department every year, and the transportation department then performs quality inspection on the data in a unified manner. Because the GPS signals have drift and many uncontrollable factors exist in the manual acquisition process, the GPS track data inevitably has some linear problems, wherein the most typical problems include flying spot, long straight line and short straight line problems. In order to practically ensure the quality of the rural road basic data, the linear problem existing in the road network must be detected and repaired. At present, the detection of the problems of point density and flying points has been completely automated, but the problem repair mainly depends on manual editing, the efficiency is low, and the quality is difficult to ensure.

Disclosure of Invention

The invention aims to provide an integrated detection and restoration method for solving the problem of flying point and point density of road network data, which has the characteristics of clear thought, clear steps, precise algorithm, small linear change and high operation efficiency and aims to solve the problems in the background technology.

In order to achieve the purpose, the invention provides the following technical scheme:

the integrated detection and restoration method for solving the problem of flying spot and spot density of the road network data comprises the following steps:

s1: data preparation, including uniform data format, performing preliminary inspection on the data, automatically detecting road data with serious problems, performing manual editing, cleaning the basically qualified data of the preliminary inspection, and warehousing;

s2: detecting and repairing the long straight line, calculating the distance between adjacent nodes in the road section, and judging the long straight line if the distance is greater than a threshold value; the repairing method is that points are taken as newly added nodes of the road section from the starting point of the long straight line along the long straight line according to a fixed distance until the end point of the long straight line, and if the distance of the last section is smaller than the threshold value of the short straight line, the section is merged into the last section;

s3: respectively calculating the distance between two adjacent nodes in a road section, judging as a short straight line if the distance is less than a threshold value, only keeping the previous node if the distance is 0, calculating the angle of an included angle formed by three adjacent nodes in the road section, judging as a flying point if the angle is less than the threshold value, deleting the flying point if the angle is too small, detecting the flying point again, and carrying out iterative processing until the flying point with too small angle does not exist;

s4: the short straight line and the flying point are integrally repaired, a node which meets the conditions of the flying point and the short straight line simultaneously is searched as a node of the road section along the advancing direction of the road section within a certain distance range, and if the conditions cannot be considered, the node which meets the conditions of the short straight line is searched as the node of the road section;

s5: and detecting and repairing the residual flying spots, namely respectively taking points with certain distances from the flying spots on two sides of the residual flying spots, replacing the flying spots with two newly obtained nodes, detecting the flying spots again after the processing is finished, and performing iterative processing until no flying spots exist in the road section.

Further, S1 imports the road network data stored in the shape format into an Oracle or SQL Server database through ArcSDE for convenience of organization and management of the data.

Further, the angle of the flying spot deletion in step 3 is less than 20 °.

Further, in S4, for each segment of polyline, a new point sequence array is established, a newly selected node is added to the array, and after the point selection is completed, a new polyline is formed according to the array, and the original polyline is replaced.

Further, firstly, adding a first node point1 of polyline into the array as a first node, and then, selecting a second node of the array, wherein the selection of the second node is completely according to the distance from the first node, and the specific selection mode is as follows:

1) if the distance between the next node point2 of polyline and point1 is greater than the short straight line threshold, directly adding point2 into array as a second node, and if the distance between point2 and point1 is less than or equal to the short straight line threshold, turning to the next step;

2) and recording the next node of point2 on polyline as point3, if the distance between point1 and point3 is greater than the short straight line threshold, searching a point meeting the condition on the line segment formed by point2 and point3, and if the distance between point1 and point3 is less than the short straight line threshold, taking point3 as a new point2, and continuing to switch to the steps.

Further, the third and the following nodes of array are selected until the penultimate node of polyline, and the specific selection mode of the nodes is as follows:

1) selecting a penultimate point of array as point1, a penultimate point of array as point2 and a next node of polyline as point3, if the distance between point3 and point2 is greater than a short straight line threshold value and the angle of an included angle formed by point1, point2 and point3 is greater than an angle threshold value, adding point3 into the array, and if point3 does not meet the condition, turning to the next step;

2) defining the next point of point3 on polyline as point4, and a line segment formed by point3 and point4 as polylineTmp, starting from point3 on the polylineTmp, selecting point add according to the step length of 0.1 meter, if the point add meets the conditions of flying points and short straight lines, adding the point add into array, if no point simultaneously meeting the conditions of flying points and short straight lines is available up to point4, and the distance between point4 and point2 is less than raidus, defining point4 as new point3, repeating the step, and if the distance between point add and point2 is greater than ius, turning to the next step;

3) starting from the last point added to array, the next node of polyline is designated point 3. If the distance between point3 and point2 is greater than the short straight line threshold, adding point3 into array, otherwise, starting from point3, selecting point add with the step length of 0.1 meter along the advancing direction of polyline, and if the distance between point add and point2 is greater than the short straight line threshold, adding point add into array;

4) when the last point of polyline is selected, the point is marked as point2, the last point of array is marked as point1, if the distance between point2 and point1 is smaller than the short straight line threshold value, point1 is removed from array, then point2 is added into array, if the distance between point2 and point1 is larger than the short straight line threshold value, point2 is directly added into array, and a new line shape formed by array replaces the original line shape.

Compared with the prior art, the invention has the beneficial effects that: the invention provides an integrated detection and restoration method for solving the problem of flying spot and spot density of road network data, which automatically detects road section data with serious linear problem or format problem and carries out manual editing, then detects and restores long straight line problem in road section, then carries out short straight line and flying spot detection and preliminary restoration, then selects nodes of road section on the premise of comprehensively considering the problems of flying spot and short straight line, preferentially considers the problem of solving short straight line if the problems of flying spot and short straight line are not considered, ensures that the problem of short straight line is completely solved in the step, and finally selects points with fixed distances from both sides to flying spot for residual flying spots, replaces the original flying spot as a new node of road section, and carries out iterative processing until the flying spots are completely eliminated. The invention can process the linear problems of point density, flying points and the like in the road network according to the relative requirements of traffic management departments on the road linear, obviously improves the linear detection and repair efficiency of the road network data by utilizing the spatial information technology, and has higher accuracy and normalization.

Drawings

FIG. 1 is a flow chart of the present invention;

FIG. 2 is a schematic diagram of the flying spot calculation process of the present invention;

FIG. 3 is a schematic diagram of a second point selection process for a road segment according to the present invention;

FIG. 4 is a schematic illustration of a selection process for a third and subsequent point of a road segment in accordance with the present invention;

fig. 5 is a schematic diagram of the elimination process of the residual flying spot in the road section of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Referring to fig. 1, an integrated detection and restoration method for solving the problem of flying spot and spot density of road network data includes the following steps:

step 1: preparing data:

the main content of data preparation comprises a uniform data format, the data is preliminarily checked, road data with serious problems (such as the same coordinates of all nodes in a road section and data with topological problems of self-intersection, self-overlapping and the like in a line shape) is automatically detected and manually edited, and the basically qualified data of the preliminary check is cleaned and put in a warehouse; firstly, the road network data is preliminarily checked, road data with obvious problems is removed, for example, coordinates of all nodes in a road section are the same, topology checking is carried out on the data, and road section data with topology errors such as self-intersection, self-overlapping and the like is found out.

Then, for convenience of organization and management of data, road network data stored in a SHAPE format is imported into databases such as an Oracle database and an SQL Server through ArcSDE, after the data is imported, attribute information of the road network is stored in corresponding attribute fields in the databases, and spatial information is stored in SHAPE fields. For convenience of calculation, the SHAPE type is set as Geometry, and the spatial coordinate system selects a geographic coordinate system, such as: WGS84 coordinate system or CGCS2000 coordinate system.

Step 2: detecting and repairing the long straight line problem:

too low a point density will be reflected by a long straight line problem, i.e. the distance between two adjacent nodes is larger than a certain threshold. The long straight line detection method comprises the steps of sequentially traversing all nodes in a road section, calculating Euclidean distances between adjacent nodes, and marking the distance as a long straight line if the distance exceeds a threshold value. In the distance calculation process, firstly, a space coordinate system of the road network data needs to be converted into a projection coordinate system from a geographical coordinate system, and for a WGS84 coordinate system, a universal transverse axis mercator projection with a 6-degree band is generally adopted; for the CGCS2000 coordinate system, a gaussian-gram projection of the 6 degree band is typically used.

Calculating the distance between adjacent nodes in the road section, and if the distance is greater than a threshold value, judging the road section to be a long straight line; the repairing method is that points are taken as newly added nodes of the road section from the starting point of the long straight line along the long straight line according to a fixed distance until the end point of the long straight line, and if the distance of the last section is smaller than the threshold value of the short straight line, the section is merged into the last section; through the step, the problem of long straight lines can be solved, and a new problem of short straight lines can not be generated.

And step 3: short straight line and flying spot detection and preliminary repair:

the detection of short straight lines and flying spots also requires the transformation of the spatial coordinate system of the data from the geographical coordinate system to the projection coordinate system. For the projected road section, firstly calculating the distance between all adjacent nodes in the road section, if the distance is less than a specified threshold value, judging the road section as a short straight line, if the distance is 0 m, only reserving the previous point, and deleting the next point to eliminate the short straight line with the length of 0 m; and then calculating included angle angles formed by all three adjacent nodes in the road section, if the angle is smaller than a certain threshold value, judging the flying point, if the angle is too small (smaller than 20 degrees), deleting the flying point, detecting the flying point again, and carrying out iterative processing until the flying point with too small angle does not exist.

Wherein, 1. angle calculation formula

In the detection process of the flying point, the angle of an included angle formed by three adjacent nodes in the road section, namely the size of ≈ ACB in fig. 2, is actually calculated, and the angle calculation adopts the cosine theorem, namely:

where a is the length of the side AC, b is the length of the side BC, and c is the length of the side AC.

2. Optimization of distance and angle calculation efficiency

Usually, for each road segment, the distance between all two adjacent nodes and the angle of the included angle formed by all three adjacent nodes are calculated, if there are n nodes in a road segment, the total number of distance calculation and angle calculation needs (n-1) times and (n-2) times, and it involves time-consuming operations such as squaring, dividing, etc., and there are a lot of repeated calculations, and the calculation efficiency is low. In order to improve the calculation efficiency, a numpy library of python is used, multiple times of numerical calculation are converted into single matrix operation, the batch calculation of the distance and the angle is realized, and the calculation efficiency can be improved by more than one hundred times due to the characteristics of numpy in the aspects of memory access modes, CPU cache, vectorization instructions and the like.

And 4, step 4: short straight line and flying spot integrated repair:

the problem of long straight lines has been completely eliminated in step 2, leaving only the problems of short straight lines and flying spots. This step eliminates all short straight line problems, but leaves some flying spot problems. The solution idea of the invention is to search nodes satisfying both the flying point and the short straight line conditions as nodes of a road section along the advancing direction of the road section within a certain distance range, and if the nodes satisfying the short straight line conditions cannot be considered, the nodes satisfying the short straight line conditions are searched as nodes of the road section;

and establishing a new point sequence array for each section of polyline, adding a newly selected node into the array, forming a new polyline according to the array after point selection is finished, and replacing the original polyline. First, the polyline's first node point1 is added to the array as the first node. Then, a second node of the array is selected. The second node is selected based solely on the distance from the first node. The specific selection mode is as follows:

1) if the distance between the next node point2 of polyline and point1 is greater than the short straight line threshold, directly adding point2 into array as a second node, and if the distance between point2 and point1 is less than or equal to the short straight line threshold, turning to the next step;

2) the next node of point2 on polyline is recorded as point3, if the distance between point1 and point3 is greater than the short straight line threshold, a point meeting the condition is searched on the line segment formed by point2 and point3, and the distance between a point and point1 is inevitably equal to the short straight line threshold. As shown in fig. 3, the length of L3 is a short straight line threshold, the length of L2 can be calculated according to the cosine formula, the length of L1 and the distance between A, B can be calculated according to the pythagorean theorem, and the length between point2 and a can be calculated to determine the coordinate of point a, and point a is placed in array. And if the distance between the point1 and the point3 is smaller than the short straight line threshold, taking the point3 as a new point2, and continuing to move to the step.

Then, the third and later nodes of array are selected until the penultimate node of polyline. And searching a new node which meets the conditions of the short straight line and the flying point simultaneously in a certain range along the road section, and adding the array, wherein the problem of the short straight line is preferentially considered under the condition that the new node cannot meet the conditions of the short straight line and the flying point simultaneously. The maximum search range is set as radius, the numerical value of radius cannot be too small or too large, if radius is too small, the modification which can be generated on the linear shape is very limited, and the problem of flying spots is difficult to solve; if radius is too large, the problem of flying spots is easier to solve, but the variation generated by the line shape may be too large, so that the detail information of the road section is lost, and the integrity of the information of the road section is damaged, especially in mountainous areas, the situation that a sharp curve exists in the road section is likely to be the same as the actual situation, and the line shape is more cautiously modified. The default value of radius is generally 10 meters, and for roads and expressways in flat terrain areas, radius can be set slightly larger, and for roads and rural highways in complicated terrain areas, radius should be relatively smaller. The specific selection mode of the nodes is as follows:

1) and selecting the point from the second last point of array as point1, the point from the second last point of array as point2 and the next node of polyline as point3, and adding point3 into the array if the distance between point3 and point2 is greater than the short straight line threshold and the angle of the included angle formed by point1, point2 and point3 is greater than the angle threshold. If point3 does not satisfy the condition, proceed to the next step.

2) As shown in FIG. 4, the next point of point3 on polyline is defined as point4, and the line segment formed by point3 and point4 is defined as polyline Tmp, on polyline Tmp, from point3, point add is selected according to the step size of 0.1 meter, and if the point add meets the conditions of flying point and short straight line, the point add is added into array. If no point satisfying both the flying point and the short straight line condition is obtained up to point4 and the distance between point4 and point2 is smaller than raidus, point4 is determined as a new point3 and step 2 is repeated. If the distance between pointAdd and point2 is greater than radius, proceed to the next step.

3) Starting from the last point added to array, the next node of polyline is designated point 3. If the distance between point3 and point2 is greater than the short straight line threshold, point3 is added to array. Otherwise, starting from point3, selecting point pointAdd in a step size of 0.1 meter along the advancing direction of polyline, and adding the point add to array if the distance between the point add and point2 is greater than the short straight line threshold.

And finally, when the last point of the polyline is selected, the point is marked as point2, the last point of the array is marked as point1, if the distance between point2 and point1 is smaller than a short straight line threshold value, point1 is removed from the array, and then point2 is added into the array. If the distance between point2 and point1 is greater than the short straight line threshold, point2 is added directly to the array. And replacing the original linear shape with the new linear shape formed by array.

And 5: detecting and repairing residual flying spots:

and the residual flying spots need to be eliminated in a point encryption mode, and for the residual flying spots, the specific encryption mode is that points with certain distances from the flying spots on two sides are respectively taken, the newly obtained two nodes are used for replacing the flying spots, the line shape is modified, the flying spot detection is carried out again after the processing is finished, and the iterative processing is carried out until no flying spot exists in the road section. The specific operation is as shown in fig. 5, the angle BAC is less than the threshold value, the point a is a flying point, the points E, D are respectively taken on AB and AC, so that the distances of AE and AD are equal to a fixed length, and the flying point can be eliminated by replacing the original point a with D, E points.

The method provided by the invention is explained by taking road basic data of a certain province as an example: and selecting road network data of a certain province in China to perform linear detection and repair experiments. The road network comprises 14000 road sections, covers about 20 ten thousand square kilometers, has the horizontal and vertical coordinate span of over 500 kilometers, covers most of the situations in the actual road network management, and is helpful for accurately evaluating the practicability and reliability of the proposed algorithm. The threshold value for the short straight line is set to 3 meters, the threshold value for the long straight line is set to 200 meters, and the threshold value for the flying spot is set to 130 °. Before and after the restoration, the line shape detection is respectively carried out on the road network data, and the detection results are shown in the following table 1:

TABLE 1

	Number of short straight line problems	Number of long straight line problems	Number of flying spot problems	Total number of linear problems
					Road network data before restoration	80896	1526	2203	84625
Repaired road network data	2348	0	0	2348
					Proportion of repair problem	97.1％	100％	100％	97.2％

Before and after the restoration, the lengths of the road sections are respectively calculated and the length changes of the road sections are counted, and the statistical results are shown in table 2:

TABLE 2

Percentage of change in length	Number of road sections	Proportion of number of road sections to total number
			>5％	2	0.01％
1％‐3％	40	0.29％
			0.1％‐1％	522	3.75％
<0.1％	13370	95.95％

Through analysis of experimental results, the method provided by the invention can completely solve the problems of flying spots and long straight lines on the premise of no obvious change of the line shape, the number of the short straight line problems is obviously reduced, and the total number of the line shape problems is reduced by 97%.

In the experimental part, the whole province road network data of a certain province in China is selected for linear detection and restoration, and the experimental result shows that the method can quickly detect the linear problem of the road network and restore the linear problem to the maximum extent, and has higher practicability and reliability.

In summary, the integrated detection and repair method for solving the problem of flying spot and point density of road network data provided by the invention automatically detects and manually edits road segment data with serious linear problem or format problem, then detects and repairs long straight line problem in road segment, then detects and primarily repairs short straight line and flying spot, then selects nodes of road segment on the premise of comprehensively considering flying spot and short straight line problem, if the short straight line problem cannot be solved, the short straight line problem is ensured to be completely solved in this step, and finally selects points with fixed distance from flying spot on both sides to replace original flying spot as new nodes of road segment, and iterates processing until flying spots are completely eliminated. The invention can process the linear problems of point density, flying points and the like in the road network according to the relative requirements of traffic management departments on the road linear, obviously improves the linear detection and repair efficiency of the road network data by utilizing the spatial information technology, and has higher accuracy and normalization.

The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be able to cover the technical solutions and the inventive concepts of the present invention within the technical scope of the present invention.

Claims (6)

1. The integrated detection and restoration method for solving the problems of flying points and point density of road network data is characterized by comprising the following steps of:

s1: data preparation, including uniform data format, performing preliminary inspection on the data, automatically detecting road data with serious problems, performing manual editing, cleaning the basically qualified data of the preliminary inspection, and warehousing;

s2: detecting and repairing the long straight line, calculating the distance between adjacent nodes in the road section, and judging the long straight line if the distance is greater than a threshold value; the repairing method is that points are taken as newly added nodes of the road section from the starting point of the long straight line along the long straight line according to a fixed distance until the end point of the long straight line, and if the distance of the last section is smaller than the threshold value of the short straight line, the section is merged into the last section;

s3: respectively calculating the distance between two adjacent nodes in a road section, judging as a short straight line if the distance is less than a threshold value, only keeping the previous node if the distance is 0, calculating the angle of an included angle formed by three adjacent nodes in the road section, judging as a flying point if the angle is less than the threshold value, deleting the flying point if the angle is too small, detecting the flying point again, and carrying out iterative processing until the flying point with too small angle does not exist;

s4: the short straight line and the flying point are integrally repaired, a node which meets the conditions of the flying point and the short straight line simultaneously is searched as a node of the road section along the advancing direction of the road section within a certain distance range, and if the conditions cannot be considered, the node which meets the conditions of the short straight line is searched as the node of the road section;

s5: and detecting and repairing the residual flying spots, namely respectively taking points with certain distances from the flying spots on two sides of the residual flying spots, replacing the flying spots with two newly obtained nodes, detecting the flying spots again after the processing is finished, and performing iterative processing until no flying spots exist in the road section.

2. The method for integrally detecting and repairing flying spot and spot density problems of road network data according to claim 1, wherein S1 is configured to import road network data stored in a shappefile format into an Oracle or SQL Server database through ArcSDE for convenience of organization and management of the data.

3. The method for integrally detecting and repairing flying spots and point densities of road network data according to claim 1, wherein the flying spot deletion angle in step 3 is less than 20 °.

4. The method for integrally detecting and repairing flying spot and spot density problems of road network data according to claim 1, wherein a new spot sequence array is established for each polyline in S4, a newly selected node is added to the array, and after the point selection is finished, a new polyline is formed according to the array to replace the original polyline.

5. The method for integrally detecting and repairing flying spot and spot density problems of road network data according to claim 4, wherein a first node point1 of polyline is added to array as a first node, and then a second node of array is selected, wherein the second node is selected completely according to a distance from the first node, and the specific selection method is as follows:

1) if the distance between the next node point2 of polyline and point1 is greater than the short straight line threshold, directly adding point2 into array as a second node, and if the distance between point2 and point1 is less than or equal to the short straight line threshold, turning to the next step;

2) and recording the next node of point2 on polyline as point3, if the distance between point1 and point3 is greater than the short straight line threshold, searching a point meeting the condition on the line segment formed by point2 and point3, and if the distance between point1 and point3 is less than the short straight line threshold, taking point3 as a new point2, and continuing to switch to the steps.

6. The method as claimed in claim 4, wherein the third and subsequent nodes of array are selected to the penultimate node of polyline, and the specific selection of the nodes is as follows:

1) selecting a penultimate point of array as point1, a penultimate point of array as point2 and a next node of polyline as point3, if the distance between point3 and point2 is greater than a short straight line threshold value and the angle of an included angle formed by point1, point2 and point3 is greater than an angle threshold value, adding point3 into the array, and if point3 does not meet the condition, turning to the next step;

2) defining the next point of point3 on polyline as point4, and a line segment formed by point3 and point4 as polylineTmp, starting from point3 on the polylineTmp, selecting point add according to the step length of 0.1 meter, if the point add meets the conditions of flying points and short straight lines, adding the point add into array, if no point simultaneously meeting the conditions of flying points and short straight lines is available up to point4, and the distance between point4 and point2 is less than raidus, defining point4 as new point3, repeating the step, and if the distance between point add and point2 is greater than ius, turning to the next step;

3) starting from the last point added to array, the next node of polyline is designated point 3. If the distance between point3 and point2 is greater than the short straight line threshold, adding point3 into array, otherwise, starting from point3, selecting point add with the step length of 0.1 meter along the advancing direction of polyline, and if the distance between point add and point2 is greater than the short straight line threshold, adding point add into array;

4) when the last point of polyline is selected, the point is marked as point2, the last point of array is marked as point1, if the distance between point2 and point1 is smaller than the short straight line threshold value, point1 is removed from array, then point2 is added into array, if the distance between point2 and point1 is larger than the short straight line threshold value, point2 is directly added into array, and a new line shape formed by array replaces the original line shape.