JP4614444B2 - Automatic generation method of road network data - Google Patents

Description

  The present invention relates to a technology for automatically generating road network data of a road center line from various graphic data related to a road among digital map data having a layer structure by a computer.

  Road network data used for car navigation or the like is composed of a polygonal line vector figure indicating the center position of a road and a point vector figure indicating a road intersection. In general, a broken line indicating the center position is referred to as a link, and a point indicating an intersection is referred to as a node. When generating these data, image data such as aerial photographs and topographic maps are displayed on a graphic display, and a road network is manually input such as digitizing.

  In recent years, it has become easier to obtain vectorized digital map data, and data on facilities on roads has been input with high accuracy. These digital map data generally have a layer structure, and can be extracted and used for each type.

  Examples of the figure related to the road include a block boundary line and a side groove indicating the outside of the road, and a lane boundary paint indicating the sidewalk boundary. Also, there are stop lines, pedestrian crossing paint, center lines, etc. that exist inside the road. In order to generate a road network indicating the center line of a road from such graphic data, a method of finding an outer vector of a pair of roads and generating a center line vector so as to take the middle of the vectors.

JP 2004-246554 A

  In the conventional method, each object exists independently, and there is a problem that it is difficult to find the correspondence because of the complexity of the intersection shape. In other words, since it is difficult to pair the road outside line, there are few places where the road network can be automatically generated, and there are many corrections after the automatic generation, so the cost required for road network generation is high and it takes time. there were.

  An object of the present invention is to provide a technique for automatically generating road network data with higher accuracy at a lower cost.

  The present invention is characterized in that high-accuracy road network data is automatically generated over the entire area by using a road as one image line and generating a center line by thinning the road.

  First, the figure of the layer which comprises the outer side of a road, such as a boundary line of a block, a gutter, and a curb, is converted into a binary image from the digital map data already vectorized. Then, expansion / contraction processing is performed using a width corresponding to one lane of the road as a guideline to generate image data with the road as one image line. Similarly, the graphic data of a layer that forms the vicinity of an intersection such as a pedestrian crossing or a stop line is converted into a binarized image, and image data that fills the vicinity of the intersection is created by expansion / contraction processing. The two previously created image data are combined (OR operation), and the white areas created in the road in this image data are painted on the road surface such as stop lines, pedestrian crossings, and regulatory information inside the road. Compared with a binary image generated from graphic data such as the above, a solid image is generated for overlapping portions.

  Through the above process, image data having a road as one image line is generated, and this is thinned to generate vector data of the road center line. After thinning, the generated vector polygonal line is shaped by performing a correction process using road characteristics to obtain final road network data.

  According to the present invention, road network data can be automatically generated using graphic data relating to roads included in digital map data. This can save labor and reduce costs for creating road network data.

  Hereinafter, an automatic road network generation system according to an embodiment of the present invention will be specifically described with reference to the drawings. FIG. 1 is a functional block diagram showing a configuration example of a road network automatic generation system according to this embodiment. As shown in FIG. 1, a road network automatic generation system according to the present embodiment includes a central processing unit 1 that performs input / output of road network data, a display device 3 with a keyboard device 2 having character and graphic screens, And a digital map data file 4 that can be referred to from the processing device 1.

  The central processing unit 1 includes a road data extraction processing unit 5, a vector image conversion processing unit 6, a road embedding processing unit 7, a thinning processing unit 8, and a shape correction processing unit 9. The digital map data file 4 is stored in a CD drive or a storage device, and is read into the central processing unit 1 to read the road data extraction processing unit 5, vector image conversion processing unit 6, road filling process of the central processing unit 1. The road network data automatically generated by the unit 7, the thinning processing unit 8, and the shape correction processing unit 9 is displayed on the display device 3.

  FIG. 2 is a flowchart showing an outline of the procedure of the road network automatic generation process executed in the central processing unit 1 in the road network automatic generation system according to the present embodiment. As shown in FIGS. 1 and 2, it corresponds to the road data extraction processing 5, vector image conversion processing unit 6, road embedding processing unit 7, thinning processing unit 8, and shape correction processing unit 9 of the central processing unit 1, respectively. The processing is performed in the order of the road data extraction step ST1, the vector image conversion step ST2, the road filling step ST3, the thinning step ST4, and the shape correction step ST5. The road filling process ST3 further includes a road outside processing process ST31, an intersection processing process ST32, and a filling process ST33.

  The process in each process will be described in detail.

(Road data extraction process ST1)
In the road data extraction step, graphic data of a layer related to a road is extracted from digital map data having a layer structure as shown in FIG. Graphic data is held in a vector format such as polyline or polygon. As shown in FIG. 3, the map data includes layers such as road inner structures, road outer lines, sidewalks, pedestrian crossings, house frames, house names, contour lines, place names, landmarks, and railways. Among them, a layer related to the road (in this case, five layers from the first layer to the fifth layer among the above layers) is extracted.

(Vector image conversion step ST2)
In the vector image conversion step, an image memory area that can be mapped is created based on graphic data (indicated by graphic L1 in FIG. 4A) in a layer related to the road. First, a region including at least a part of the graphic data L1 is mapped in units of pixels based on the respective graphic data L1 (L1 + Px1), and the graphic data L1 is erased, whereby a binary image memory as shown in FIG. Data is created (Px1). The binary image memory data (graphic data separated for each layer) formed by the above processing is used for the first classification (FIG. 5A) that constitutes the outside of the road, and the first that constitutes the vicinity of the intersection of the road. Two types of images (FIG. 6A) and a third category constituting the inside of the road (FIG. 7B) can be divided into three types of images.

  As the first classification constituting the outside of the road, there are road outside lines (may be block lines), gutters, curbs, separation bands, and the like. The second classification constituting the vicinity of the intersection of the road includes lane boundary paint such as a pedestrian crossing and a stop line. The third classification constituting the inside of the road includes lane boundary paint such as a center line, a lane boundary line, and various regulatory signs.

(Road outside processing step ST31)
In the road outside processing step, the image in FIG. 5A relating to the first classification that forms the outside of the road is subjected to image expansion processing with a size corresponding to a general road lane segment. Thus, if the road corresponds to one lane, the road is formed as one thick image line (for example, L2 and L3). This state is shown in FIG. On the other hand, a contraction process is applied by the same size as the expanded size to generate an image memory in a state as shown in FIG. In this state, if the road is equivalent to one lane, the inside is filled, but the thick road is not filled. Therefore, processing for dealing with this is performed below. First, the first process is a process of painting the vicinity of an intersection, which is performed in (intersection process step ST32).

(Intersection processing step ST32)
In the road intersection processing step, the image of FIG. 6A relating to the second classification constituting the intersection is subjected to an expansion process with a size sufficient to fill all the regions (for example, Ac1) estimated to be the intersection (FIG. 6). (See (b)). This is contracted again to obtain the state shown in FIG. 6C, and is combined with the image generated in step ST31 (corresponding to FIG. 5C: FIG. 6D) (or processing). The image shown in 6 (e) is generated. As a result, the vicinity of the intersection is also filled, and only a large number of white areas where the thick road is divided at each intersection remain. In order to fill this white area, the process of (filling process step ST33) is performed.

(Infill processing step ST33)
In the filling process, the third classification that constitutes the inside of the road in the closed region in the image shown in FIG. A process of filling all the white areas overlapping with the image shown in FIG. Thereby, as shown in FIG.7 (c), it will be in the state by which all the insides of the road were filled. It should be noted that the white area filling process in consideration of overlap is a white area separated by the black pixels shown in FIG. 7A to be filled and a white area where the black pixels shown in FIG. This corresponds to the step of painting all the area.

(Thinning process step ST4)
In the thinning process step, the thinning process is performed on the image shown in FIG. 8A in which the entire inside of the road is filled by using a method in which the width of the image line is uniformly narrowed from both sides. As shown in (b), vector data for only the center line of the road is created.

(Shape correction step ST5)
In the shape correction processing step, the road centerline vector data created in the thinning step ST4 is subjected to shape correction processing in consideration of the following road characteristics, and the data is shaped.
1) It is rare that the line segment constituting the road vector is largely changed in angle, and the tendency is stronger as it is closer to the connecting portion of the road.
2) In many cases, a link (a line connecting nodes connected to a road) can be approximated by a straight line. For example, in FIG. 8B, the link 1-2 connecting the node 1 and the node 2 has a plurality of short graphics s between the nodes 1-2, but these graphics s are removed.
3) The links are often connected at right angles.

  As a result of the above processing steps, vector road network data as shown in FIG. 8C can be finally created.

  Note that a program for causing a computer to execute the above method or a recording medium storing the program is also included in the scope of the present invention.

  INDUSTRIAL APPLICABILITY The present invention can be used for creating graphic data related to roads used in car navigation devices and the like.

It is a block diagram which shows the example of schematic structure of the road network automatic generation system by one embodiment of this invention. It is a flowchart figure which shows the flow of the road network automatic generation process by this Embodiment. It is a figure which shows the example of the layer structure of digital map data. It is a figure which shows one structural example of the image memory produced | generated by vector image conversion. It is a figure which shows the result along the flow of the image memory process of a road outer side process process. It is a figure which shows the result along the flow of the image memory process of an intersection process process. It is a figure which shows the result along the flow of the image memory process of a filling process. It is a figure which shows the result along the flow of the image memory process of a thinning and shape correction process process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Central processing unit, 2 ... Keyboard apparatus, 3 ... Display apparatus, 4 ... Digital map data file, 5 ... Road data extraction process part, 6 ... Vector image conversion process part, 7 ... Road embedding process part, 8 ... Thin line Processing unit, 9... Shape correction processing unit.

Claims (3)

A road data extraction processing unit that extracts graphic data related to roads from map data;
Binary image data is created based on the graphic data, and the binary image data is divided into a first classification that forms the outside of the road, a second classification that forms the vicinity of the intersection of the road, and the inside of the road. An image conversion processing unit that is divided into three types of images;
A road embedding processing unit that fills the inside of the road based on the first to third classifications;
An apparatus for automatically generating a road network figure, comprising: a thinning processing unit that performs processing for narrowing the width of a road subjected to filling processing to demarcate a center line of the road. The road embedding processing unit
The image belonging to the first category is subjected to expansion processing of the image in the size of the road lanes min, and the road outside the processing unit to add a shrinkage treatment in expanded size and the same size,
Intersection processing that performs expansion processing and contraction processing on the image belonging to the second classification in a size that fills the area estimated to be an intersection, and performs OR operation processing on the image generated by the road outside processing unit And
For the white area inside the road, an embedding processing unit that fills an area overlapping with the image belonging to the third classification;
2. The thinning processing unit according to claim 1, further comprising: a thinning processing unit that generates vector data including a center line of a road by narrowing a width of an image line from both sides of the image subjected to the filling process. Automatic generation device for road network graphics. A method of automatically creating a road network figure by a computer,
A first classification that binarizes the map data stored in the digital map database in a layer structure and configures the outside of the road, a second classification that configures the vicinity of the intersection of the road, and a third that configures the inside of the road An image conversion processing step for performing processing by an image conversion processing unit that is divided into three types of images:
The image belonging to the first category is subjected to expansion processing of the image in a common road lane of magnitude, performs expanded size and the same size in a contracted handle added Ru road outside process, the first for images that belong to two categories, the expansion in size that region suspected image memory and the intersection is filled performs erosion processing, performs an OR operation processing between image memory that is more generated on the road outside processing A filling process step of performing processing by a road filling processing unit that fills a region overlapping with an image belonging to the third classification for a white area inside the road;
A thinning processing step for performing processing by the thinning processing unit that creates vector data including the center line of the road by narrowing the width of the image line from both sides of the image subjected to the embedding processing. A method for automatically generating road network figures.

Images