JP2022039005A - Map data generation device and map data generation method - Google Patents

Abstract

To provide a map data generation device that can acquire a proper road sign by a simple method.SOLUTION: A map data generation device comprises: a data acquisition unit that acquires measurement data on a reflection point group of a road surface scanned from a laser scanner; a reflection intensity image generation unit that generates a reflection intensity image according to reflection intensity of the measurement data on the reflection point group acquired by the data acquisition unit; an area extraction unit that extracts a candidate image area of a road sign of a road surface from the reflection intensity image; a candidate image generation unit that generates a candidate image of the road sign according to the extracted candidate image area; a road sign identification unit that determines whether the candidate image is the road sign on the basis of the candidate image generated by the candidate image generation unit and a template image of the road sign stored in a storage unit; and a map generation unit that sets vector data corresponding to the template image when it is determined by the road sign identification unit that the candidate image is the road sign, and generates map data.SELECTED DRAWING: Figure 6

Description

The present disclosure relates to a map data generator that generates map data based on point cloud data.

As a technique for measuring the shape of a feature, there is a laser measurement technique for acquiring three-dimensional point cloud data representing the shape of the feature using a laser scanner. For example, the Mobile Mapping System (MMS) can acquire 3D point cloud data and visible images while driving on the road using a laser scanner and camera mounted on the vehicle, and can be used for map creation and navigation. It is used for spatial data collection.

It is also being studied to detect road markings using the measurement data of the laser reflection point cloud acquired by MMS. Specifically, the road markings are detected by utilizing the difference in laser reflection intensity due to the difference in color and material between the road markings on the road surface and other parts (asphalt, concrete, etc.). Unlike visible images, the measurement data of the laser reflection point cloud is not easily affected by the lighting conditions.

In this regard, for example, in Japanese Patent Application Laid-Open No. 2016-142604, a plurality of relational expressions expressing the relationship between the position of each point included in the local region and the feature amount are obtained by function approximation based on the point group data. The gradient intensity and the gradient direction of the feature amount are calculated from the relational expression for each of the local regions, and the gradient intensity image showing the gradient intensity of each point included in the predetermined region and the gradient of each point included in the predetermined region. A gradient direction image showing the direction is generated, a plurality of line segments are extracted from the gradient intensity image and the gradient direction image, a line segment representing a feature is detected from the plurality of line segments, and the detected line segment is detected. Discloses a method of generating map data representing a feature in.

Further, in Japanese Patent Application Laid-Open No. 2018-173749, the reflection point cloud belonging to the point cloud cluster is classified according to the threshold value determined for each point cloud cluster, and the reflection point cloud having a laser reflection intensity equal to or higher than the threshold value is indicated as one road sign. The method of extracting as a road marking candidate point cloud corresponding to is disclosed.

Japanese Unexamined Patent Publication No. 2016-142604 Japanese Unexamined Patent Publication No. 2018-173749

On the other hand, in the conventional method, since the road marking based on the point cloud data is extracted as it is, complicated work such as manually correcting the map data is required. In addition, it was difficult to guarantee the quality of the map due to variations in work.

An object of the present disclosure is to provide a map data generation device and a map data generation method capable of acquiring appropriate road markings by a simple method.

The map data generator of the present disclosure has a data acquisition unit that acquires measurement data of a group of reflection points on a road surface scanned from a laser scanner, and a reflection intensity according to the reflection intensity of the measurement data of the reflection point group acquired by the data acquisition unit. A reflection intensity image generation unit that generates an image, an area extraction unit that extracts a road marking candidate image area from the reflection intensity image, and a candidate that generates a road marking candidate image according to the extracted candidate image area. A road marking identification unit that determines whether or not the candidate image is a road marking based on the image generation unit, the candidate image generated by the candidate image generation unit, and the template image of the road marking stored in the storage unit. It also includes a map generation unit that sets vector data corresponding to the template image and generates map data when the road marking identification unit determines that the candidate image is a road marking.

Preferably, the region extraction unit extracts a candidate image region of the road marking of the road surface whose reflection intensity value is higher than a predetermined value from the reflection intensity image.

Preferably, the candidate image generation unit generates a surrounding image in a predetermined area range as a candidate image for road marking according to the extracted candidate image area.

Preferably, the road display identification unit determines whether or not the candidate image is a road sign by template matching between the candidate image and the template image of the road sign.

The map data generation method of the present disclosure includes a step of acquiring measurement data of a group of reflection points on a road surface scanned from a laser scanner, and a step of generating a reflection intensity image according to the reflection intensity of the measurement data of the acquired reflection point group. , A step of extracting a candidate image area of a road sign on a road surface from a reflection intensity image, a step of generating a candidate image of a road sign according to the extracted candidate image area, a step of generating a candidate image of a road sign, and a storage unit. A step to determine whether the candidate image is a road sign based on the stored road sign template image, and vector data corresponding to the template image when the candidate image is determined to be a road sign. It is provided with a step of setting and generating map data.

The map data generation device and the map data generation method of the present disclosure can acquire appropriate road markings by a simple method.

It is a figure explaining the structure of the map data generation apparatus 1 according to an embodiment. It is a figure explaining the outline of the process of the map data generation apparatus 1 based on an embodiment. It is a figure explaining the template image data based on an embodiment. It is a figure explaining the processing flow of the map data generation apparatus 1 based on an embodiment. It is a flow diagram explaining the generation of the candidate image of the candidate image generation unit 14 according to an embodiment. It is a figure explaining the specific example of the process of the map data generation apparatus 1 according to an embodiment. It is a figure explaining the grouping process of the candidate image generation part 14 of the map data generation apparatus 1 according to an embodiment. It is a figure explaining the identification of the road marking of the map data generation apparatus 1 according to an embodiment. It is a figure explaining the adjustment of the template image according to an embodiment.

The embodiments will be described in detail with reference to the drawings. The same or corresponding parts in the drawings are designated by the same reference numerals, and the description thereof will not be repeated.

The map data generation device is a device that generates map data including the road sign based on the measurement data of the reflection point cloud acquired by the laser scanner from the target space for detecting the road sign. The measurement data of the reflection point cloud is acquired by, for example, MMS. In MMS, as an example, a laser scanner mounted on an automobile irradiates a laser diagonally downward or diagonally upward from the upper part of a vehicle body, and detects laser reflected light from an object. The optical axis of the laser is scanned laterally, and laser pulses are emitted at every minute angle within the scanning angle range. The distance is measured based on the time from the laser emission to the reception of the reflected light, and at that time, the laser emission direction, time, and the position / posture of the vehicle body are measured. From these measurement data, point cloud data representing the three-dimensional coordinates of the point where the laser pulse is reflected in the target space can be obtained. The laser scanner also measures the intensity of reflected light (reflection intensity).

Road markings are markings drawn on the road surface and are legal markings that display regulations or instructions regarding road traffic, but the road markings detected by the map data generator are not limited to legal markings. Further, the road marking is basically a paved surface provided on the ground surface paved with concrete, asphalt, etc., but the paved surface does not have to be a road. Further, the road surface in this example may include a paved surface such as a parking lot, in addition to a road including a roadway and a sidewalk.

FIG. 1 is a diagram illustrating a configuration of a map data generation device 1 according to an embodiment. With reference to FIG. 1, the map data generation device 1 includes an arithmetic processing unit 10, a storage device 20, an input device 30, and an output device 40. As the arithmetic processing unit 10, it is possible to make dedicated hardware for performing various arithmetic processing, but the arithmetic processing unit 10 is constructed by using a computer and a program executed on the computer.

The CPU (Central Processing Unit) of the computer constitutes the arithmetic processing unit 10, and for example, various functional blocks of the arithmetic processing apparatus 10 are realized by executing a program stored in the storage device 20.

The arithmetic processing device 10 includes a data acquisition unit 11, a reflection intensity image generation unit 12, a region extraction unit 13, a candidate image generation unit 14, a road marking identification unit 15, and a map generation unit 16.

The storage device 20 is composed of a hard disk or the like built in the computer. The storage device 20 causes the arithmetic processing device 10 to function as a data acquisition unit 11, a reflection intensity image generation unit 12, a region extraction unit 13, a candidate image generation unit 14, a road marking identification unit 15, and a map generation unit 16. Stores programs for this purpose, other programs, and various data required for processing. For example, the storage device 20 stores laser measurement data as data to be processed. Further, the storage device 20 stores the template vector of the road marking in advance.

The laser measurement data is measurement data of a group of reflection points acquired by MMS traveling on a road, and includes information on three-dimensional coordinates and reflection intensity of each reflection point. The Cartesian coordinate system XYZ is defined as the three-dimensional coordinate system in the target space, the XY plane is defined as the horizontal plane, and the Z axis is defined as the vertical axis whose positive direction is upward.

The input device 30 is a keyboard, a mouse, or the like, and is used for the user to perform an operation. Further, the input device 30 may include an interface device for inputting data from another system.

The output device 40 is a display, a printer, or the like, and is used for showing a road surface or a road marking to a user by screen display, printing, or the like.

The data acquisition unit 11 acquires the laser measurement data stored in the storage device 20.

The reflection intensity image generation unit 12 generates a reflection intensity image according to the laser reflection intensity of the reflection point cloud.

The area extraction unit 13 extracts a candidate image area for road markings on the road surface from the reflection intensity image. Specifically, the region extraction unit 13 extracts a candidate image region for road markings on a road surface having a reflection intensity value higher than a predetermined value from the reflection intensity image. At this point, clustering is performed on the distribution of the region where the reflection intensity value is higher than the predetermined value from the reflection intensity image, and the candidate image region of the road marking on the road surface is extracted.

The candidate image generation unit 14 generates a candidate image of the road marking on the road surface according to the extracted candidate image area.

The road marking identification unit 15 determines whether or not the candidate image is a road marking based on the generated candidate image and the template image of the road marking on the road surface. The template image is generated by rasterizing the template vector stored in the storage device 20.

When the candidate image is determined to be a road marking, the map generation unit 16 sets vector data corresponding to the template image and generates map data.

FIG. 2 is a diagram illustrating an outline of processing of the map data generation device 1 based on the embodiment. As shown in FIG. 2A, as an example, a reflection intensity image of laser measurement data, which is data to be processed, is shown. FIG. 2B is a diagram illustrating a case where a candidate image region is extracted from a reflection intensity image. Then, the candidate image of the arrow is generated according to the candidate image area. Then, template matching is performed in which the template image is rotated and expanded / contracted and compared with the candidate image. FIG. 2C shows map data in which vector data corresponding to the template image is set when it is determined to be a road marking. In this example, the map data is generated by setting the vector data corresponding to the template image instead of the candidate image. That is, instead of creating the map data as it is or by shaping the candidate image, the map data is created using the vector data corresponding to the pre-formatted template image. This makes it possible to create appropriate map data. That is, even if the shape of the candidate image of the road marking is deformed due to dirt or defects, it is possible to generate map data in an appropriately shaped state.

FIG. 3 is a diagram illustrating a template image based on the embodiment. As shown in FIG. 3, in this example, a case where template images T1 to T8 based on eight template vectors stored in the storage device 20 are provided is shown. A "white line" template image, a "character" template image, a "speed display" template image, an "arrow" template image, and the like are shown. This example is an example, and template images based on various templates are provided according to the type and use of the road. The template image is generated by rasterizing the template vector stored in the storage device 20.

FIG. 4 is a diagram illustrating a processing flow of the map data generation device 1 based on the embodiment. As shown in FIG. 4, the map data generation device 1 acquires the data (step S2). Specifically, the data acquisition unit 11 acquires the laser measurement data stored in the storage device 20. Next, the map data generation device 1 generates a reflection intensity image (step S4). Specifically, the reflection intensity image generation unit 12 generates a reflection intensity image according to the laser reflection intensity of the reflection point cloud. Next, the map data generation device 1 extracts a candidate image area (step S8). After the image editing process, the area extraction unit 13 extracts a candidate image area for road markings on the road surface from the reflection intensity image. Next, the map data generation device 1 generates a candidate image (step S10). The candidate image generation unit 14 generates a candidate image of the road marking on the road surface according to the extracted candidate image area. Next, the map data generation device 1 identifies whether or not the candidate image is a road marking (step S12). The road marking identification unit 15 discriminates whether or not the road marking is based on the generated candidate image and the template image of the road marking on the road surface. Next, the map data generation device 1 determines whether or not it is a road marking (step S14). The road marking identification unit 15 outputs the identification result to the map generation unit 16. Next, in step S14, when the map data generation device 1 determines that it is a road marking (YES in step S14), vector data is set (step S16). The map generation unit 16 sets vector data corresponding to the template image of the road sign identified by the road sign identification unit 15 at the same position as the candidate image of the road sign as map data. Next, the map data generation device 1 determines whether or not all the candidate image areas have been extracted (step S18). The area extraction unit 13 determines whether or not all the candidate image areas have been extracted. If it is determined in step S18 that the map data generation device 1 has extracted all the candidate image areas (YES in step S18), the process ends (end). On the other hand, if it is determined in step S18 that the map data generation device 1 has not extracted all the candidate image areas (NO in step S18), another candidate candidate image area is extracted (step S20). When the region extraction unit 13 determines that all the candidate image regions have not been extracted, the region extraction unit 13 extracts another candidate image region in the same manner as described in step S8. Then, the process returns to step S10.

FIG. 5 is a flow chart illustrating extraction of a candidate image region of the region extraction unit 13 according to the embodiment. FIG. 6 is a diagram illustrating a specific example of processing of the area extraction unit 13 of the map data generation device 1 according to the embodiment.

As shown in FIG. 5, the region extraction unit 13 executes a histogram smoothing process as an image editing process (step S20). FIG. 6A shows a reflection intensity image of the laser measurement data. FIG. 6B shows a case where a histogram smoothing process is executed on a reflection intensity image as an image editing process of the region extraction unit 13.

Next, the area extraction unit 13 executes a binarization process as an image editing process (step S22). In FIG. 6C, as the image editing process of the region extraction unit 13, the reflection intensity image obtained by performing the histogram smoothing process is binarized based on whether or not the reflection intensity is equal to or higher than a predetermined threshold value. Is shown when you run.

Next, the area extraction unit 13 executes a noise reduction process as an image editing process (step S24). FIG. 6D shows a case where the noise reduction process is executed as the image editing process of the area extraction unit 13. Specifically, the region extraction unit 13 executes an expansion treatment on the peripheral region of the pixels of the reflection intensity image, and then performs an erosion treatment. On the contrary, after the erosion treatment is executed, the expansion treatment is executed. It is possible to remove noise by executing the process.

Next, the area extraction unit 13 executes the grouping process (step S26). The region extraction unit 13 removes noise from the reflection intensity image by the image editing process, and extracts a region having a reflection intensity value higher than a predetermined value. That is, a plurality of areas constituting the road marking on the road surface are extracted. On the other hand, the characters on the road markings are not connected and the areas forming the characters are separated. The region extraction unit 13 performs clustering related to the distribution of the region and executes the grouping process. Specifically, the region extraction unit 13 executes a grouping process in which those having a short distance are grouped into the same group based on the distance between the extracted plurality of regions. As a result, even areas that form separated characters can be processed as the same group.

FIG. 7 is a diagram illustrating a grouping process of the area extraction unit 13 of the map data generation device 1 according to the embodiment. FIG. 7A shows a case where, for example, an area of a candidate point cloud for displaying characters such as “place” is extracted. The parts that are not concatenated may not be judged as the same character. FIG. 7B shows a case where regions having a short distance are connected and processed as the same group.

With reference to FIG. 5 again, the area extraction unit 13 then executes the area determination process (step S28). As an area determination process, the area of pixels in each grouped area is calculated. If the calculated area is smaller or larger than the predetermined area range, the possibility of road marking is low, so the process of removing the group from the target of road marking is executed.

Then, proceed to the next process (return).

The candidate image generation unit 14 generates a candidate image of the road marking on the road surface according to the candidate image area extracted by the area extraction unit 13. Specifically, the candidate image generation unit 14 executes a crop process to generate a surrounding image as a candidate image for road markings, leaving a target portion and erasing the other portions. As a result, the candidate image generation unit 14 can generate a peripheral image in a predetermined area range as a candidate image for road markings.

FIG. 8 is a diagram illustrating identification of road markings of the map data generation device 1 according to the embodiment. FIG. 8A shows a case where a region for generating a candidate image is extracted. Then, as shown in FIG. 8B, a target candidate image is generated. Specifically, a candidate image is generated in which the target portion is left and the other portions are erased. FIG. 8C shows a template matching method with the white line template image described with reference to FIG. Specifically, the template image is slightly rotated and translated. Then, the candidate image and the template image are compared to search for the position / angle at which the degree of similarity (normalized intercorrelation coefficient as an example) is maximized. A candidate image is identified as a road marking if the normalized intercorrelation coefficient between the template image and the candidate image is greater than or equal to the threshold. In this example, it is identified as a road marking in the white line part. The template matching method is not limited to the method using the normalized cross-correlation coefficient, but is not limited to the method using the normalized cross-correlation coefficient, SSD (Sum of Squared Difference), SAD (Sum of Absolute Difference), ZNCC (Zero means Normalized Cross Correlation), SSDA (Sequential). Similarity Detection Algorithm) may be used to evaluate the degree of similarity so that the candidate image can be identified as a road sign.

FIG. 9 is a diagram illustrating the adjustment of the template image according to the embodiment. FIG. 9A shows the generated candidate image. FIG. 9B shows an example in which the angle of the candidate image is adjusted according to the template image of the white line. In FIG. 9C, the candidate image and the template image are compared, the length and width are compared, and the template image is expanded and contracted in at least one of the length direction and the width direction to adjust the size. It is shown. The above-mentioned template matching is performed by rotating and translating the stretched template image.

The road marking identification unit 15 identifies the candidate image as a road marking when the normalized intercorrelation coefficient between the template image and the candidate image is equal to or greater than the threshold value. In this example, the road marking identification unit 15 identifies the candidate image as the road marking in the white line portion.

The map generation unit 16 sets vector data corresponding to the template image of the road sign identified by the road sign identification unit 15 at the same position as the candidate image of the road sign as map data.

(Action)
The data acquisition unit 11 of the map data generation device 1 acquires the measurement data of the reflection point cloud of the road surface scanned from the laser scanner. The reflection intensity image generation unit 12 generates a reflection intensity image according to the reflection intensity of the measurement data of the reflection point cloud acquired by the data acquisition unit 11. The area extraction unit 13 extracts a candidate image area for road markings on the road surface from the reflection intensity image. The candidate image generation unit 14 generates a candidate image for road marking according to the extracted candidate image area. The road marking identification unit 15 determines whether or not the candidate image is a road marking based on the candidate image generated by the candidate image generation unit 14 and the template image of the road marking. When the road sign identification unit 15 determines that the candidate image is a road sign, the map generation unit 16 sets vector data corresponding to the template image and generates map data. The map data generation device 1 generates map data based on the measurement data of the reflection point cloud of the road surface scanned from the laser scanner by the processing. That is, it is possible to obtain an appropriate road marking by a simple method. At this point, vector data is used instead of the candidate image to generate map data. Conventionally, since the candidate image is generated based on the point cloud data, for example, when the locus around the candidate image is traced, even a straight line image such as a white line is not exactly a straight line. In addition, if the road marking is dirty or missing, the condition is reflected and the condition becomes inappropriate as map data. Therefore, it is possible to obtain appropriate road markings by generating map data using vector data. In addition, complicated work such as manually correcting the map data is not required, and the work does not vary, and the quality of the map can be guaranteed. Furthermore, it is possible to efficiently process map data by modifying and editing the data.

The method described in each of the above embodiments is a storage of a magnetic disk (hard disk, etc.), an optical disk (CDROM, DVD, etc.), a magneto-optical disk (MO), a semiconductor memory, etc., as a program that can be executed by a computer. It can also be stored in a medium and distributed. Further, the storage medium may be in any form as long as it is a storage medium that can store a program and can be read by a computer.

Further, in order to realize the above embodiment, an OS (operating system) running on the computer based on instructions of a program installed on the computer from a storage medium, MW (middleware) such as database management software and network software, and the like. You may execute a part of each process of. Further, the storage medium in each embodiment is not limited to a medium independent of the computer, but also includes a storage medium in which a program transmitted by a LAN, the Internet, or the like is downloaded and stored or temporarily stored. Further, the storage medium is not limited to one, and the case where the processing in each of the above embodiments is executed from a plurality of media is also included in the storage medium in the present invention, and the medium configuration may be any configuration. The computer in each embodiment executes each process in each of the above embodiments based on the program stored in the storage medium, and is composed of one device such as a personal computer and a network of a plurality of devices. Any configuration such as a connected system may be used.

It should be considered that the embodiments disclosed this time are exemplary in all respects and not restrictive. The scope of the present disclosure is shown by the scope of claims, not the description described above, and is intended to include all modifications within the meaning and scope equivalent to the scope of claims.

1 Map data generation device, 10 Arithmetic processing device, 11 Data acquisition unit, 12 Reflection intensity image generation unit, 13 Area extraction unit, 14 Candidate image generation unit, 15 Road marking identification unit, 16 Map generation unit, 20 Storage device, 30 Input device, 40 output device.

Claims (5)

A data acquisition unit that acquires measurement data of a group of reflection points on the road surface scanned from a laser scanner, and a data acquisition unit.
A reflection intensity image generation unit that generates a reflection intensity image according to the reflection intensity of the measurement data of the reflection point cloud acquired by the data acquisition unit, and a reflection intensity image generation unit.
A region extraction unit that extracts a candidate image region for road markings on the road surface from the reflection intensity image, and a region extraction unit.
A candidate image generation unit that generates a candidate image of the road marking according to the extracted candidate image area, and a candidate image generation unit.
A road marking identification unit that determines whether or not the candidate image is a road marking based on the candidate image generated by the candidate image generation unit and the template image of the road marking.
A map data generation device including a map generation unit that sets vector data corresponding to the template image and generates map data when the road marking identification unit determines that the candidate image is a road marking. The map data generation device according to claim 1, wherein the region extraction unit extracts a candidate image region of a road marking on the road surface whose reflection intensity value is higher than a predetermined value from the reflection intensity image. The map data generation device according to claim 1, wherein the candidate image generation unit generates a peripheral image in a predetermined area range as a candidate image for the road marking according to the extracted candidate image area. The map data generation device according to claim 1, wherein the road display identification unit determines whether or not the candidate image is a road sign by template matching between the candidate image and the template image of the road sign. The step of acquiring the measurement data of the reflection point cloud of the road surface scanned from the laser scanner, and
A step to generate a reflection intensity image according to the reflection intensity of the measured data of the acquired reflection point cloud, and
A step of extracting a candidate image area of a road marking on the road surface from the reflection intensity image, and
A step of generating a candidate image of the road marking according to the extracted candidate image area, and
A step of determining whether or not the candidate image is a road marking based on the generated candidate image and the template image of the road marking, and
A map data generation method comprising a step of setting vector data corresponding to the template image and generating map data when the candidate image is determined to be a road sign.

Images