CN101833665B - Method for extracting roads from remote sensing map image - Google Patents
Method for extracting roads from remote sensing map image Download PDFInfo
- Publication number
- CN101833665B CN101833665B CN 201010186151 CN201010186151A CN101833665B CN 101833665 B CN101833665 B CN 101833665B CN 201010186151 CN201010186151 CN 201010186151 CN 201010186151 A CN201010186151 A CN 201010186151A CN 101833665 B CN101833665 B CN 101833665B
- Authority
- CN
- China
- Prior art keywords
- road
- highway section
- line segment
- remote sensing
- roads
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Processing Or Creating Images (AREA)
- Instructional Devices (AREA)
- Navigation (AREA)
Abstract
The invention relates to a method for extracting roads from remote sensing map images in the field of remote sensing technique. The method comprises the following steps of: marking two edges of the roads; automatically generating road segments by using a projection method according to the marked edges; linking the road segments into the roads; adding road names; and saving every road in a data format by taking an intersection as the center. The method for extracting the roads from the remote sensing map image has the advantages of easy implementation, simple operation, high accuracy and the like, and is a fast and conveniently extracting mode for the roads of a remote sensing map.
Description
Technical field
What the present invention relates to is the method for the information extraction in a kind of remote sensing technology field, is specifically related to a kind of method for extracting roads of remote sensing map image.
Background technology
Numerical map is that the cartography factor and the discrete data that have definite coordinate and attribute in the GIS-T system are summarized and orderly set at the discernible storage medium of computing machine, have the computing machine identifiability, can measure the characteristics such as the property calculated, analyticity, the property transmitted, in field extensive application such as auto-navigation system and intelligent transportation systems (ITS).Development along with the GPS technology, Aki Okuno and Wenhuan Shi have proposed a kind of digital map construction method based on gps data, the method adopts vehicle GPS image data information and generates based on this bitmap, further uses the technology such as expansion, refinement bitmap to be processed the topological structure that obtains city road network again.Realized the quick renewal of numerical map based on the map constructing method of gps data, but its geography information of extracting is limited, has only obtained road network structure, and can't obtain the road name of road, the information such as have a lot of social connections, thereby and be not suitable for the application of intelligent transportation system etc.For addressing this problem, the present invention proposes a kind of semi-automatic road information extracting method for high-resolution remote sensing image, road name that can the quick obtaining road, the information such as have a lot of social connections, the road information that obtains by GPS is replenished and revises, and perfect road information is to satisfy the application demand of intelligent transportation system.Simultaneously this method employing is the semi-automatic mode of guidance manually, has overcome the problem of full-automatic map generating mode degree of accuracy deficiency, is particularly useful for the city road network of road conditions complexity, can realize quick, the accurately renewal of numerical map.
Existing literature search is found, at present for the road extraction mode in the high-resolution remote sensing image mainly contain based on rim detection and Schema-based identification.But because road conditions is complicated, house, trees, vehicle etc. are the interference to detecting all, so that discrimination, stability all can not meet the demands.The people such as Aaron K.Shackelford only have 82% in 2003 articles " Fully automatedroad network extraction from high-resolution satellite multispectral imagery " of delivering at IEEEInternational on Geoscience and Remote Sensing Symposium for the extraction integrated degree of urban road, and accuracy is 71%.The article that Wenzhong Shi and Changqing Zhu delivered at IEEE Transactions on Geoscience and Remote Sensing in 2002 " The Line SegmentMatch Method for Extracting Road Network From High-Resolution Satellite Images " proposed based on characteristics of image, for the priori of road net and the line match algorithm of Related Mathematical Models.It ranges up to about 90% the discrimination of urban road, but several parameters need to be set when detecting, and does not detect and have a lot of social connections.
Summary of the invention
The object of the invention is to overcome deficiency of the prior art, a kind of method for extracting roads of remote sensing map image is provided, realized the road information in the remote sensing map is extracted quickly and easily.
The present invention realizes by following mode, the present invention includes following steps:
Step 3, the highway section is connected into road;
Step 4, the road name of adding;
Step 5, the data layout of each road centered by the intersection stored.
Marking road section described in the step 1 refers to: the both sides of coming marking road section with line segment.Because road exists crooked, every road represents with a plurality of straight highway sections, and every complete road has defined and terminated in the intersection, and complete road comprises a plurality of highway sections.
Described highway section, the distance between its two edge lines section is the width in this highway section; The center line of two edge lines is the position, highway section.
Sciagraphy described in the step 2 generates the highway section automatically, refers to: to any two line segments successively following automatic formation condition judge if any which bar be not inconsistent, then no longer down judge; If all meet, think that then these two line segments think two edges in certain highway section:
1.: two line segments are approximate parallel, and the difference of two line segment slopes is in threshold range;
2.: it is very high that the projection of a line segment on another line segment overlaps degree;
3.: distance is less than the breadth extreme of road between two line segments.
Described breadth extreme can be set, and also can calculate by the line segment in the program.The latter's algorithm is: all distances that meet between two line segments of above-mentioned requirements are sorted, and getting intermediate value is normal road width, supposes the widest for normal value 5 times of road; The existing ultimate range of getting less than 5 times of normal values is the breadth extreme of road.
The present invention is because the road curved way of remote sensing map image adopts broken line form to describe.Described in the step 3 highway section being connected into road, is to utilize the data structure of double linked list to utilize pointer to link in the highway section, and the rear employing broken line data layout that all is disposed carries out the road storage.
The present invention need carry out huge search because the map datum amount is large when data such as carrying out GIS are processed, so design centered by the intersection, the mode that the road that each is related with the intersection is stored is together carried out the data storage, is convenient to subsequent calculations.
The present invention has realization of being easy to, simple to operate, and the accuracy advantages of higher can provide to the road in the remote sensing map a kind of quickly and easily extracting mode.
Description of drawings
Fig. 1, the inventive method process flow diagram.
Schematic diagram behind Fig. 2, the mark edge.
Map behind Fig. 3, the generation highway section.
Fig. 4, each condition diagram when automatically generating the highway section.
Fig. 5, the highway section is connected into road and marks the road name.
Fig. 6, schematic diagram when converting the box type stores to.
Fig. 7, whole structure figure.
Embodiment
Below in conjunction with drawings and Examples technical scheme of the present invention is described in further detail.Following examples are implemented under take technical solution of the present invention as prerequisite, have provided detailed embodiment and process, but protection scope of the present invention is not limited to following embodiment.
Embodiment
The present embodiment selects * economizes * the high-definition remote sensing map image in city, and its process flow diagram comprises the steps: as shown in Figure 1
As shown in Figure 2, the marking road section edge refers to come with line segment the both sides of marking road section.
Because road exists crooked, every road represents with a plurality of straight highway sections, and every complete road has defined and terminated in the intersection, and complete road comprises a plurality of highway sections.Shown a specific highway section among Fig. 3.
Described highway section, the distance between its two edge lines section is the width in this highway section; The center line of two edge lines is the position, highway section.
According to analysis, following three conditions are satisfied in the highway section.Any two line segments are carried out following automatic formation condition successively judge if any any bar be not inconsistent, then no longer down judge; If all meet, think that then these two line segments think two edges in certain highway section:
1.: two line segments are approximate parallel, and namely the difference of two line segment slopes is in threshold range, and shown in Fig. 4 (a), this example adopts threshold value 0.2rad;
2.: the projection on another line segment of line segment overlaps degree greater than certain threshold value, overlaps level calculating method shown in Fig. 4 (b), namely overlaps the projected length of degree=line segment a on line segment b/line segment b length.This example adopts threshold value 75%;
3.: distance is less than the breadth extreme of road between two line segments.
Breadth extreme can by artificial setting, also can calculate by the line segment in the program.The employing latter in this example.Specific algorithm is: to existing each line segment, finds out the line segment that satisfies with it above-mentioned two requirements and asks distance, and the minimum value of recording distance, if the line segment that does not meet, then distance is composed higher limit.Have 594 line segments in this example, thus 594 distances obtained, to non-distance apart from higher limit sort (totally 406), getting intermediate value is normal highway section width, this example intermediate value is 14 pixel values, amounts to 16.8 meters of actual ranges, supposes the widest for normal value 5 times in highway section; The existing breadth extreme of getting less than 5 times of normal values is the breadth extreme in highway section, and obtaining ultimate range in this example is 65 pixel values, amounts to 78 meters of actual ranges.
Shown in Fig. 4 (c), because two line segments are not really parallel, so the distance between them adopts a line segment two-end-point to the mean value of another line segment distance.
After obtaining two edges, highway section, with the width of the distance between these two line segments as this highway section, with their center line as the position, highway section.Ask the method for road axis to be: with the direction of a line segment along its normal toward another line segment (can by calculate apart from the time distance sign judge) half of distance between mobile two line segments.
All line segments all correctly mate becomes the highway section, amounts to 297 highway sections.
Step 3, the highway section is connected into road:
Starting point and terminal point for each highway section, with starting point and the endpoint calculation distance in all highway section, if with the distance of a plurality of beginning or ends less than given threshold value (adopt 25 pixel values in this example, amount to into actual range 30m), think that then this point is the point of intersection, does not process; If only with the distance of a point less than above-mentioned threshold value, think that then these two highway sections link to each other at these two some places, utilize the data structure of double linked list to utilize pointer to link in the highway section.For finding the point that connects the highway section no longer to participate in the calculating of next time.
The rear data layout that utilizes doubly linked list to convert broken line to that all is disposed carries out the road storage, every road initialization road " road name " by name.Obtain altogether 143 roads.
Road after the highway section connects as shown in Figure 5.Adopt successively 4 kinds of different color cycle to draw when drawing road, whether wrongly be convenient to observe.
Step 4, the road name of adding:
By every Road name initial value that step 3 obtains, click road name ejection dialog box and can make amendment.
Step 5, the data layout of each road centered by the intersection stored:
Because the map datum amount is larger, when processing, data such as carrying out GIS need carry out huge search, so the data structure of design polylinebox form is stored.Shown as shown in Figure 6 a polylinebox.As shown in FIG., each polylinebox comprises an intersection, every road being associated of intersection therewith, a minimum rectangular region that comprises these roads.Mate first rectangular region when searching for like this, then carry out the road search, accelerate search speed.
Be that the specific algorithm of polylinebox form is as follows with the road format conversion:
Starting point and terminal point for every road, starting point and endpoint calculation distance with all road, if there is the distance of two points (to adopt 25 pixel values in this example less than given threshold value, amount to into actual range 30m), then get the mean value of these two points and be the intersection point in these two road, this intersection point and existing crossroad stomion are calculated distance, if distance is arranged less than above-mentioned threshold value, think that then this intersection point overlaps with existing crossroad stomion, gives up this intersection point; Otherwise think that this is a new crossroad stomion, is stored as a new polylinebox.For this new polylinebox, look for all and its crossroad stomion less than the road of above-mentioned threshold value, and find out the up and down maximum value of all flex points and end points on these road broken lines, as the position of road place rectangular region.
Comprise altogether 76 polylinebox structures in this example.Fig. 7 is the whole structure figure after processing.For beautiful interface, do not show the rectangular region among each polylinebox.
Claims (3)
1. the method for extracting roads of a remote sensing map image is characterized in that, comprises the steps:
Step 1, edge on both sides of the road, mark road specifically refer to come with line segment the both sides of marking road section, because that road exists is crooked, every road represents with a plurality of straight highway sections, and every complete road has defined and terminated in the intersection, and complete road comprises a plurality of highway sections;
Step 2, according to the edge of mark adopting sciagraphy automatically to generate the highway section;
In the step 2, described highway section, the distance between its two edge lines section is the width in this highway section; The direction of a line segment along its normal toward another line segment moved half of distance between two line segments, namely obtains road axis, with their center line as the highway section;
In the step 2, described sciagraphy generates the highway section automatically, refers to: any two line segments are carried out following automatic formation condition successively judge if any which bar be not inconsistent, then no longer down judge; If all meet, think that then these two line segments are two edges in certain highway section:
1.: two line segments are approximate parallel, and the difference of two line segment slopes is in threshold range;
2.: the projection of a line segment on another line segment overlaps degree greater than setting threshold;
3.: distance is less than the breadth extreme of road between two line segments;
The length of the projected length of described coincidence degree=line segment on another line segment/another line segment, Threshold is 75%;
Step 3, the highway section is connected into road;
Step 4, the road name of adding;
Step 5, the data layout of each road centered by the intersection stored;
In the step 5, described data layout centered by the intersection is the storage organization of polylinebox, comprises an intersection and the minimum rectangular area of all roads of being associated of intersection therewith.
2. the method for extracting roads of remote sensing map image according to claim 1, it is characterized in that, described breadth extreme is: all distances that meet between two line segments of above-mentioned requirements are sorted, and getting intermediate value is normal road width, supposes the widest for normal value 5 times of road; The existing ultimate range of getting less than 5 times of normal values is the breadth extreme of road.
3. the method for extracting roads of remote sensing map image according to claim 1, it is characterized in that, described in the step 3 highway section is connected into road, utilize the data structure of double linked list to utilize pointer to link in the highway section, the data layout of the rear employing broken line that all is disposed carries out the road storage.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010186151 CN101833665B (en) | 2010-05-28 | 2010-05-28 | Method for extracting roads from remote sensing map image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010186151 CN101833665B (en) | 2010-05-28 | 2010-05-28 | Method for extracting roads from remote sensing map image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101833665A CN101833665A (en) | 2010-09-15 |
| CN101833665B true CN101833665B (en) | 2013-02-27 |
Family
ID=42717728
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010186151 Expired - Fee Related CN101833665B (en) | 2010-05-28 | 2010-05-28 | Method for extracting roads from remote sensing map image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101833665B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103258203B (en) * | 2013-05-20 | 2016-08-17 | 武汉大学 | The center line of road extraction method of remote sensing image |
| CN105184270B (en) * | 2015-09-18 | 2019-06-18 | 深圳市数字城市工程研究中心 | A kind of road information Remotely sensed acquisition method based on Pulse Coupled Neural Network method |
| CN108537169B (en) * | 2018-04-09 | 2022-01-25 | 吉林大学 | High-resolution remote sensing image road extraction method based on center line and road width detection algorithm |
| CN110619258B (en) * | 2018-10-26 | 2022-02-15 | 国交空间信息技术（北京）有限公司 | Road track checking method based on high-resolution remote sensing image |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006113580A2 (en) * | 2005-04-15 | 2006-10-26 | Mississippi State University | Linear correspondence assessment |
-
2010
- 2010-05-28 CN CN 201010186151 patent/CN101833665B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006113580A2 (en) * | 2005-04-15 | 2006-10-26 | Mississippi State University | Linear correspondence assessment |
Non-Patent Citations (1)
| Title |
|---|
| 彭静等.城市航空图像中多等级道路的线特征提取.《第十二届图象图形学学术会议》.2005,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101833665A (en) | 2010-09-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110146910B (en) | Positioning method and device based on data fusion of GPS and laser radar | |
| CN110008872B (en) | Road network extraction method combining vehicle track and remote sensing image | |
| CN104866670B (en) | Based on road network topology change automatic testing method and the system of GPS space-time track | |
| CN101464158A (en) | Automatic generation method for road network grid digital map based on GPS positioning | |
| CN111427904B (en) | High-precision map data updating method and device and electronic equipment | |
| CN104062671B (en) | The GNSS Floating Car map-matching method of curvature limitation and device | |
| CN104866500A (en) | Method and device for displaying pictures in classified manner | |
| CN108204819B (en) | Map data automatic detection method and device and hybrid navigation system | |
| CN103162664B (en) | Altitude figures acquisition methods and device, navigator | |
| CN113628291B (en) | Multi-shape target grid data vectorization method based on boundary extraction and combination | |
| CN104422451A (en) | Road recognition method and road recognition apparatus | |
| CN104281746A (en) | Method for generating traffic safety road characteristic graphs on basis of point-cloud | |
| WO2021051346A1 (en) | Three-dimensional vehicle lane line determination method, device, and electronic apparatus | |
| CN101833665B (en) | Method for extracting roads from remote sensing map image | |
| Elias et al. | Automatic generation and application of landmarks in navigation data sets | |
| CN108959414B (en) | Method for extracting missing road from navigation data points based on machine learning | |
| CN109740274A (en) | High-precision road-map generation method and device | |
| Hinz et al. | Road extraction focussing on urban areas | |
| CN112802041A (en) | Ground laser point cloud building contour line extraction method based on improved Alpha Shapes algorithm | |
| CN114509065B (en) | Map construction method, system, vehicle terminal, server and storage medium | |
| CN106500594A (en) | Fusion reflected intensity and the railroad track method for semi-automatically detecting of geometric properties | |
| CN114859374A (en) | Newly-built railway cross measurement method based on unmanned aerial vehicle laser point cloud and image fusion | |
| CN117523403A (en) | Method, system, equipment and medium for detecting spot change of residence map | |
| CN115909241A (en) | Lane line detection method, system, electronic device and storage medium | |
| CN115031744A (en) | Cognitive map positioning method and system based on sparse point cloud-texture information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130227 Termination date: 20150528 |
|
| EXPY | Termination of patent right or utility model |