Terrain image generation system and method

Abstract

To express each of a large-scale terrain to a small-scale terrain in detail by one drawing.SOLUTION: A normal vector having information on the direction and inclination of a shadow with respect to a stereoscopic plane estimated from a point cloud is calculated (S601) based on point cloud data in an XXZ format extracted from point cloud data classified only by board height or point cloud data in a digital elevation model (DEM), illuminances of the point cloud are calculated by a PCV method based on the normal vector (S603), and hueing is performed with reference to the illuminances (S605).SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a system and method for generating terrain images based on point cloud data. [Background technology]

[0002] There is a known technology for creating a 3D model of an object based on point cloud data (see, for example, Patent Document 1). Point cloud data, which is made up of a collection of points, can capture the unevenness of the terrain in more detail than when using DEM (digital elevation model) data made up of a grid.

[0003] Furthermore, since point cloud data is the source data for generating DEM data, if a topographical representation map is created directly from point cloud data without going through DEM data, it can be created in a short amount of time, making it particularly useful for interpreting the topography of disaster sites where urgency is required. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2010-266419 Summary of the Invention [Problem to be solved by the invention]

[0005] Topographical representation maps created based on point cloud data contain a large amount of information and have high resolution, making it possible to capture the detailed unevenness of the terrain. However, large-scale topography (faults, deep landslides, geology, geological structures) and small-scale topography (geological collapses, springs, artificial alterations) have different data resolutions, making it difficult to clearly represent each on the same map.

[0006] Therefore, it is necessary to prepare a map showing the large-scale topography and a map showing the small-scale topography in order to interpret the topography of the disaster site.

[0007] The present invention has been made in light of the above circumstances, and aims to represent in detail on a single drawing both large-scale and small-scale topography. [Means for solving the problem]

[0008] The terrain image generation system of the present invention is a terrain image generation system that generates a terrain image based on point cloud data, and is equipped with a normal vector calculation means that calculates the normal vector of each point from the point cloud data, an illuminance calculation means that calculates the illuminance of the point cloud based on the calculated normal vector, and a hue processing means that performs a hue processing that assigns a hue according to the contrast indicating the terrain undulations by referring to the calculated illuminance.

[0009] According to the above configuration, the illuminance distribution calculated based on the point cloud data is expressed in a high-density pointillist drawing format, regardless of the resolution of the point cloud data. Therefore, even when drawing a high-resolution topographical map, it is possible to continuously express small-scale to large-scale topographical features on a single topographical map.

[0010] In the above configuration, the coloring process assigns two hues that are complementary to each other on the color wheel.

[0011] According to the above configuration, the uneven shape of the terrain can be clearly expressed by allocating two hues that are complementary to each other on the color wheel according to the contrast in accordance with the illuminance of the point cloud.

[0012] In the above configuration, the hue processing is performed by allocating yellow to bright contrast areas and blue to dark contrast areas.

[0013] According to the above configuration, by performing a colorization process using blue and yellow, which are complementary colors on the color wheel, depending on the illuminance of the point cloud, it is possible to obtain data that can clearly represent uneven shapes from small to large scale terrain with high resolution, regardless of the resolution of the point cloud data.

[0014] The above-mentioned configuration further comprises a mixing means for mixing one or more images selected from a shaded relief map, a topographical relief index map, and a monochrome gradient map with the colorized image.

[0015] According to the above configuration, by mixing one or more maps selected from a shaded relief map, a topographical relief index map, and a black-and-white gradient map with the color-processed data, a plan view with a more emphasized shading effect can be obtained.

[0016] The terrain image generation method of the present invention is a terrain image generation method for generating a terrain image based on point cloud data, and includes the steps of calculating the normal vector of each point from the point cloud data, calculating the illuminance of the point cloud based on the calculated normal vector, and assigning hues that are complementary to each other on the color wheel to bright and dark contrast areas that indicate the terrain undulations, respectively, by referring to the calculated illuminance.

[0017] According to the above configuration, the illuminance distribution calculated based on the point cloud data is expressed using high-density pointillism, regardless of the resolution of the point cloud data. Therefore, even when drawing a high-resolution topographical map, it is possible to continuously depict small to large-scale topography on a single topographical map. Furthermore, by assigning two complementary hues on the color wheel according to the contrast depending on the illuminance of the point cloud, the uneven shape of the topography can be clearly expressed. [Effects of the Invention]

[0018] According to the present invention, the illuminance distribution calculated based on point cloud data is expressed using high-density pointillism, regardless of the resolution of the point cloud data. Therefore, even when drawing a high-resolution topographical map, it is possible to continuously express small-scale to large-scale topographical features on a single topographical map. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a diagram showing a schematic configuration of a topographical image generating system according to the present invention; [Figure 2] FIG. 2 is a diagram for explaining the flow of processing in the stereoscopic image generation system according to the present invention. BEST MODE FOR CARRYING OUT THE INVENTION

[0020] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0021] 1 is a block diagram showing the schematic configuration of a stereoscopic image generation system according to the present invention. The stereoscopic image generation system 100 includes a point cloud data processing unit 200, a two-dimensional data processing unit 300, and a three-dimensional data processing unit 400.

[0022] The point cloud data processing unit 200 includes a normal vector calculation unit 201 that calculates the normal vector of each point having information on the direction and inclination of shading relative to the three-dimensional surface estimated from the point cloud, an illuminance calculation unit 203 that calculates the illuminance of the point cloud using the PCV method based on the normal vector of the point cloud calculated by the normal vector calculation unit 201, and a hue processing unit 205 that performs hue conversion by referring to the illuminance.

[0023] The point cloud data handled by the point cloud data processing unit 200 is point cloud data classified only by ground height or point cloud data in XXZ format extracted from digital elevation model (DEM) data. The point cloud data processing unit 200 uses software such as "CloudCompare" that can process point cloud data, and calculates the illuminance of the point cloud using its extension tool "qPCV (Portion_de_Ciel Visible)."

[0024] The illuminance of a point cloud indicates the incident energy when light from a light source is shone on a small surface relative to the normal of each point in the point cloud. The illuminance distribution calculated based on point cloud data is expressed as a high-density pointillist drawing, regardless of the resolution of the point cloud data. Therefore, even when drawing a high-resolution topographical map, it is possible to continuously depict small-scale to large-scale topography on a single topographical map.

[0025] The colorization process uses blue and yellow, which are complementary colors on the color wheel, according to the illuminance of the point cloud, allocating yellow to bright areas of contrast and blue to dark areas of contrast, and adjusting the color range according to the contrast. Based on high-density pointillist point cloud data, and by allocating the complementary colors blue and yellow according to the brightness of the contrast, uneven shapes of landforms from small to large scale can be clearly expressed.

[0026] The two-dimensional data processing unit 300 performs processing for outputting a plan view based on the image data generated by the point cloud data processing unit 200.

[0027] The two-dimensional data processing unit 300 includes an illuminance raster image generation unit 301 that generates an illuminance raster image whose color tone is the scalar value of illuminance, an image mixing ratio setting unit 303 that adjusts the color tone of each illuminance raster image and sets an image mixing ratio for overlaying with various images, and a mixed image generation unit 305 that outputs an image obtained by mixing the illuminance raster image and various images based on the mixing ratio set by the image mixing ratio setting unit 303.

[0028] The various images to be mixed in the mixed image generator 305 are images created by the image generator 500 (described later), and are, for example, one or more images selected from a shaded relief map, a topographic relief index map, and a monochrome gradient map. The various image generator 500 generates the various images to be mixed with the illuminance raster image directly from point cloud data or based on a digital elevation model generated from point cloud data.

[0029] The three-dimensional data processing unit 400 performs processing based on the image data generated by the point cloud data processing unit 200 to output a bird's-eye view of a three-dimensional space by pointillism drawing and a ground surface map of a three-dimensional design model.

[0030] The three-dimensional data processing unit 400 includes a normal scalar calculation unit 401 that calculates normal scalar values ​​of XYZ components from point cloud data to which normals have been added in order to generate monochrome point cloud data, and an illuminance / normal scalar mixing unit 403 that mixes the illuminance calculated by the illuminance calculation unit 203 and the monochrome of the normal scalar calculated by the normal scalar calculation unit 401 in order to generate point cloud data including RGB attributes.

[0031] The image creation unit 500 includes a point cloud / DEM conversion unit 501 that converts the image data of the point cloud generated by the point cloud data processing unit 200 into digital elevation model (DEM) data, a shaded relief map creation unit 503 that creates a shaded relief map to emphasize the three-dimensional appearance of the terrain based on the DEM data, a topographical ruggedness index creation unit 505 that creates a topographical ruggedness index map based on the DEM data according to a topographical ruggedness index that represents the difference in elevation between a specific point and its surrounding points, and a black-and-white gradation map creation unit 507 that creates a black-and-white gradation map with black and white color schemes based on the DEM data.

[0032] As described above, the three-dimensional image generation system 100 generates point cloud data expressed using high-density pointillism in the point cloud data processing unit 200, and then performs colorization processing using blue and yellow, which are complementary colors on the color wheel, based on the generated point cloud data and depending on the illuminance of the point cloud, thereby obtaining data that can clearly express uneven shapes from small-scale to large-scale terrain with high resolution, regardless of the resolution of the point cloud data.

[0033] The data processed by the point cloud data processing unit 200 is subjected to two-dimensional processing for outputting a plan view, and three-dimensional processing for outputting a bird's-eye view, etc. In the two-dimensional processing for outputting a plan view, by further mixing one or more maps selected from a shaded relief map, a topographical relief index map, and a black-and-white gradient map, a plan view with a more emphasized shading effect can be obtained.

[0034] The processing flow in the 3D image generation system 100 will be described below with reference to Fig. 2. Based on point cloud data classified only by ground height or point cloud data in XXZ format extracted from digital elevation model (DEM) data, the computer calculates the normal vector of each point, which contains information on the direction and inclination of shading relative to the 3D surface estimated from the point cloud (S601), calculates the illuminance of the point cloud using the PCV method based on the normal vector of the point cloud (S603), and performs hue generation with reference to the illuminance (S605).

[0035] When the computer outputs two-dimensional data based on the data processed in step S605 (S607:A), it generates an illuminance raster image whose color tone is the scalar value of illuminance (S609), adjusts the color tone of each illuminance raster image, and sets an image mixing ratio for overlaying it with various images such as shaded relief maps (S611), and outputs an image obtained by mixing the illuminance raster image with various images such as shaded relief maps, topographical relief index maps, and monochrome gradient maps generated in step S619 based on the set mixing ratio (S613). While checking the mixed image, it repeats the processes from step S611 to step S613 until sufficient shading and lighting effects are obtained, and when it determines that sufficient shading and lighting effects are obtained (S615:Yes), it outputs the mixed image as two-dimensional data for a floor plan (S617).

[0036] When the computer outputs a bird's-eye view of a three-dimensional space using pointillism painting or a ground surface map of a three-dimensional design model based on the data processed in step S605 (S607:B), it calculates normal scalar values ​​of the XYZ components from the point cloud data to which normals have been added in order to generate monochrome point cloud data (S621), and mixes the illuminance calculated in step S603 with the monochrome tone of the normal scalar calculated in step S621 to generate point cloud data including RGB attributes (S623). While checking the mixed image, it repeats the processes from step S621 to step S623 until a sufficient shading effect and lighting effect are obtained, and when it determines that a sufficient shading effect and lighting effect are obtained (S625:Yes), it outputs the result as three-dimensional data for a bird's-eye view, ground surface, etc. (S623). [Explanation of symbols]

[0037] 100 Stereoscopic Image Generation System 200 Point cloud data processing section 201 Normal vector calculation unit 203 Illuminance calculation unit 205 Colorization processing section 300 2D data processing section 301 Illuminance raster image generation unit 303 Image mixing ratio setting section 305 Mixed image generation unit 400 3D data processing section 401 Normal scalar calculation unit 403 Illuminance / Normal Scalar Mixer 500 Image Creation Department 501 Point Cloud / DEM Conversion Unit 503 Shaded Relief Map Creation Department 505 Topographical Relief Index Creation Department 507 Black and White Gradient Coloring Creation Department

Claims

1. A terrain image generation system that generates a terrain image based on point cloud data, a normal vector calculation means for calculating a normal vector of each point from the point cloud data; an illuminance calculation means for calculating the illuminance of the point cloud based on the calculated normal vector; a coloring processing means for performing a coloring process to assign a hue in accordance with a contrast showing the undulations of the terrain by referring to the calculated illuminance; A terrain image generation system comprising:

2. 2. The topographical image generating system according to claim 1, wherein the colorization process assigns two hues that are complementary to each other on the color wheel.

3. 2. The topographical image generating system according to claim 1, wherein the coloring process is performed by allocating yellow to bright contrast areas and blue to dark contrast areas.

4. 2. The topographical image generating system according to claim 1, further comprising a mixing means for mixing the colorized image with one or more images selected from a shaded relief map, a topographical relief index map, and a black-and-white gradient map.

5. A terrain image generation method for generating a terrain image based on point cloud data, comprising: calculating a normal vector of each point from the point cloud data; calculating the illumination of the point cloud based on the calculated normal vector; a step of allocating hues that are complementary to each other on the color wheel to bright and dark contrast areas that represent the undulations of the terrain, with reference to the calculated illuminance; A method for generating a terrain image comprising:

Images