KR101079359B1 - A system for generating digital map using an aerial photograph and aerial light detection of ranging data - Google Patents

Abstract

PURPOSE: A system for generating a digital map using an aerial photograph and an aerial light detection of ranging data is provided to obtain spatial information through a filtering of areal laser measurement. CONSTITUTION: A data classifying unit(310) divides input data into a ground data and a non-ground data. A ground data acquisition unit(320) classifies a ground data of a flat region and slope region. A non-ground data acquisition unit(330) creates a building data from the classified non-ground data. A 3D spatial information generating unit(340) creates 3D spatial information using a road data and a building data.

Description

A digital map construction system using aerial photography and aerial laser survey data

The present invention relates to a system and method for constructing a digital map using aerial photography and aerial laser survey data.

As the modern society becomes more complex and advanced, the importance of geospatial information among various knowledge information is increasing day by day for efficient use and management of national space. The field of using spatial information is a global industry of interest, and Internet-based map services and 3D geographic information services are already provided by Google and Microsoft. In addition, the three-dimensional geographic information (GI) software industry market has been segmented and developed by function, and the diversity and expertise of the field of using geographic information is applied to the field of application of geographic information system (GIS) -based technology. Contributed to creation.

In order to generate more realistic and accurate geospatial information using more accurate analysis tools such as aerial photography and aerial laser survey data, which are the basic data of geographic information, analysis techniques using aerial photography are increasing. In addition, aerial photographs provide rich texture information on the ground surface, but have disadvantages such as shadows and ups and downs, and aerial laser survey data provide highly accurate three-dimensional terrain coordinates in the form of points, but do not provide rich texture information. . Thus, in order to provide more accurate and realistic spatial information, it is necessary to complement or fuse the disadvantages of aerial photography or aerial laser survey data.

In the present invention, it is intended to produce a higher quality digital map using aerial laser survey data.

In addition, the present invention is to effectively classify the ground data and non-surface data through the filtering of the air laser survey data.

In addition, the present invention, to reflect the inclination of the ground data to obtain more accurate road data.

In addition, the present invention is to obtain more accurate building data using the center of gravity of the non-surface data.

In addition, the present invention, to obtain more clear image information through the image post-processing process.

The present invention also provides a method that can fuse the advantages of aerial photography and aerial laser survey data.

In addition, the present invention is to provide a more accurate and realistic spatial information by using a combination of aerial photography and aerial laser survey data.

The present invention provides an aerial laser survey data receiving unit for receiving aerial laser survey data through a web server; A first filtering process is performed on the aerial laser survey data to be classified into ground data and non-ground data, and the first filtering process includes the following equations (R _ij -R _{l _ min} > T, where R _ij is _Points to any cell in a specific region, R _{l_min} indicates the minimum height value in the specific region at the l < th & gt _; repetition number, T denotes the first threshold value, and the first threshold value is represented by the equation T = sM _l. Wherein s represents the maximum inclination of the target area and M _l represents the window size of the l < th > repetition frequency; A second filtering unit which extracts point data corresponding to a building from the classified non-planar information using a second filtering method; A polygon generator which generates a polygon from point data corresponding to the extracted building by using an irregular triangle network model; The center of gravity of the polygon is calculated from point data corresponding to the building, and the outline of the polygon is called a function (x (t), y (t)), and the area of the polygon is A. expression

,

,

A weight center point obtaining unit calculating the weight center point based on the weight center point; And a digital map generator for acquiring an external facial expression element of the aerial image using the center of gravity and generating a digital map based on the external facial expression element. Provide a build system.

Also, in the present invention, the first filtering process obtains a minimum value among height values of a point within a specific area and sets only points having a height value larger than the minimum value as a comparison target value, and then the comparison target value is pre-set. If larger than the set first threshold value, it is classified as non-ground data, and if the comparison target value is smaller than or equal to the predetermined first threshold value, the data is classified as ground data.

Further, in the present invention, in the weight center point obtaining unit, the height value of the center of gravity of the polygon constructed from the point data corresponding to the building, interpolates the height value of the point data corresponding to the building from an irregular triangular network model It is characterized by obtaining.

When constructing two-dimensional or three-dimensional spatial information through embodiments of the present invention, it is possible to support improved intuition and decision making by providing a more accurate and realistic virtual environment. Previous studies related to road modeling have been expensive due to acquiring sensor data while driving all roads, but in the present invention, more accurate three-dimensional spatial information is obtained by acquiring spatial information through filtering of air laser survey data. You can build a system.

In addition, by providing a way to fuse the advantages of aerial photography and aerial laser survey data, high quality digital maps can be produced.

1 is a schematic block diagram of a three-dimensional spatial information construction system to which the present invention is applied.
2 is a schematic block diagram of an aerial laser survey data acquisition unit 211 for acquiring aerial laser survey data according to an embodiment to which the present invention is applied.
3 is a schematic block diagram of a data processing unit 300 according to an embodiment to which the present invention is applied.
4 is a schematic block diagram of a data classifying unit 310 according to an embodiment to which the present invention is applied.
5 is a schematic block diagram of a ground data acquisition unit 320 according to an embodiment to which the present invention is applied.
FIG. 6 illustrates an embodiment to which the present invention is applied to explain a method for generating a 3D road model from ground data.
7 is a schematic block diagram of the non-planar data acquisition unit 330 as an embodiment to which the present invention is applied.
FIG. 8 is a flowchart for extracting ground data and non-ground data from an aerial laser survey data according to an embodiment to which the present invention is applied.

Hereinafter, the configuration and operation of the embodiments of the present invention with reference to the accompanying drawings, the configuration and operation of the present invention described by the drawings will be described as one embodiment, whereby the technical spirit of the present invention And its core composition and operation are not limited.

In addition, the terms used in the present invention are as widely used as possible now

Phosphorus terms were selected, but specific cases will be described using terms arbitrarily selected by the applicant. In such a case, since the meaning is clearly described in the detailed description of the part, it should not be interpreted simply by the name of the term used in the description of the present invention, and it should be understood that the meaning of the term should be interpreted. . In particular, in the present specification, data or information is a term that encompasses values, parameters, coefficients, elements, and the like, and in some cases, the meaning is interpreted differently. Can be.

Spatial Information System (Spatial Information System) is an information system that integrates and processes geographic data occupying spatial location and related attribute data to efficiently collect, store, update, process, analyze, and output various types of spatial information. It can be defined as the collective organization of hardware, software, geographic data, and human resources used for the purpose. According to the use of the spatial information system, various applications of spatial information have been facilitated. In addition, it is possible to generate three-dimensional spatial information about the target area by using information on a predetermined level of facilities, aerial photographs, and digital topographic maps. It became possible. Therefore, the embodiments for generating more efficient three-dimensional spatial information will be described.

1 is a schematic block diagram of a three-dimensional spatial information construction system to which the present invention is applied.

Referring to FIG. 1, the 3D spatial information building system 100 includes a data receiving unit 200, a data processing unit 300, a database unit 400, and an image display unit 500.

The data receiver 200 includes an aerial laser survey data receiver 210 and an aerial photo data receiver 220. Here, the aviation laser survey data means information related to the point where the aviation laser survey system is mounted on the aircraft to scan a laser pulse on the ground surface and is reflected using the arrival time of the reflected laser wave. For example, the position coordinate value by the laser survey, the rotation error information of the X-axis or Y-axis generated during the aerial laser survey, the coordinate value of the entry point of the aerial laser survey data and the distance information of the outermost ground survey coordinate value of the building, There may be scale correction factors, constant height difference information between ground reference points and aerial laser survey data. The aerial photo data means two-dimensional coordinate information and attribute information that can be obtained from the aerial photo. For example, ground coordinates are calculated for internal facial expression information, a variable used when calculating the physical coordinate system of a camera, and mutual expression information, a variable used when generating a stereoscopic model, from a pixel coordinate system of an aerial photograph. When the absolute facial expression information, the mutual facial expression, and the absolute facial expression work are completed, the external facial expression element, which is a constant of the equation used to calculate a ground coordinate value corresponding to an arbitrary position on the aerial photograph, may be included. have.

The aerial laser survey data receiving unit 210 receives aerial laser survey data from the aerial laser survey data obtaining unit 211, and the aerial photograph data receiving unit 220 receives aerial photograph data, user input values, and the like. In this case, the data may be transmitted through a web server or a mobile communication network. Although not shown in FIG. 1, the aerial laser survey data acquisition unit 211 may be included in the 3D spatial information construction system 100 to which the present invention is applied. The aerial laser survey data acquisition unit 211 will be described in more detail with reference to FIG. 2.

The data processing unit 300 receives at least one of aerial photographic data or aerial laser survey data from the data receiving unit 200 and compares the input data with the ground data in order to generate more accurate and efficient spatial information. 3D spatial information is generated through image processing on the classified ground data and non-ground data. Referring to FIG. 3, the data processor 300 includes a data classifier 310, a ground data acquirer 320, a non-ground data acquirer 330, and a three-dimensional spatial information generator 340. .

According to an embodiment of the present invention, the data classifying unit 310 of the data processing unit 300 may perform image processing according to each data characteristic when using the data classification unit 310 according to the characteristics of the input data. Can be obtained. For example, the received data may be classified into ground data and non-ground data, which may be used for digital map production. Here, the ground data may include road data, and the non-ground data may include building data, plant group data, and the like.

When aviation laser survey data is input from the data receiver 200, the data processing unit 300 may apply the filtering process after converting the aviation laser survey data into raster data of a regular grid of 0.4 m intervals. have. At this time, the search area is set first considering the average elevation of the target area and the regular grid spacing, and then the error is removed by searching for excessively high or low points compared to the surrounding points to improve the accuracy when classifying the data. Can be. Here, excessively high or low points may be determined by a predetermined reference value. By applying a filtering process to the error-resolved aerial laser survey data, the ground data and the non-ground data can be classified. A detailed description thereof will be made with reference to FIG. 4.

In another embodiment of the present invention, the ground data acquisition unit 320 of the data processing unit 300 distinguishes the ground data of the flat area from the ground data of the inclined area by using the calculation method proposed by the present invention. By doing so, more accurate ground data can be obtained. The road data may be generated from the ground data. This will be described in more detail with reference to FIG. 5.

In another embodiment of the present invention, the non-planar data acquisition unit 330 of the data processing unit 300 may generate building data from the classified non-planar data. For example, contour lines may be extracted from the non-planar data and selected as boundary data of the building. In addition, the selected boundary data may be linearized and converted into a polygon to obtain building data about a target building. By using the calculation method proposed by the present invention, more accurate building data can be obtained. This will be described in more detail with reference to FIG. 6.

The 3D spatial information generator 340 may generate more accurate 3D spatial information by using road data and building data acquired from the data processing unit 300.

The database unit 140 stores aerial laser survey data and aerial photograph data, and provides information necessary when the data processing unit 300 generates two-dimensional or three-dimensional spatial information. Then, the two-dimensional or three-dimensional spatial information generated from the data processing unit 300 is stored.

In addition, the image display unit 500 outputs two-dimensional or three-dimensional spatial information obtained from the data processing unit 300. For example, the image display unit 500 may output a result of two-dimensional or three-dimensional numerical mapping and provide it to the user.

In the present invention, the numerical mapping refers to an operation of measuring a feature of a survey aerial photograph or satellite image as numerical data by using an analytical drawing system and recording it on a computer. The aerial photographing system is a digital photograph used for a spatial information system. A device that reads aerial photographs taken using surveying equipment. In addition, a digital map means a drawing or a digital map in which information of a paper map is digitized into a set of geometric figure elements or pixels in the form of points, lines, and planes. Two-dimensional or three-dimensional spatial information obtained by the present invention can be used for the numerical diagram.

2 is a schematic block diagram of an aerial laser survey data acquisition unit 211 for acquiring aerial laser survey data according to an embodiment to which the present invention is applied.

The aerial laser survey data acquisition unit 211 may include a laser scanning unit 212, a reference position acquisition unit 213, a variable data acquisition unit 214, and a data correction unit 215. The data corrector 215 may include a rotation amount corrector 215-1, a scale corrector 215-2, and a height corrector 215-3.

The laser scanning unit 212 may obtain a distance between the ground object and the aircraft. In other words, it is possible to measure the distance from the launch point of the laser wave to the topography and the features of the reflected ground. For example, by scanning the laser pulse to the ground at a constant altitude and recording the reflected time it is possible to determine the three-dimensional position of the reflection point. The laser scanning unit 212 may include, for example, a fixed laser scanner having a scan angle of 0 degrees, or may include a laser scanner having a high point density. The reference position obtaining unit 213 may obtain absolute position information of the moving object. For example, reference position information of the laser sensor may be acquired by using a global positioning system (GPS) receiver. Here, the laser sensor may be included in the laser scanning unit 212. The variable data acquisition unit 214 may acquire rotational elements, speed, and position information of the moving object. For example, the position information of the movable body may be obtained by using the rotation amount of the movable body and the measured acceleration that change every moment.

Data obtained from the laser scanning unit 212, the reference position obtaining unit 213, and the variable data obtaining unit 214 may be used to produce a numerical map. In this case, the obtained data may be classified and used according to data characteristics in the data classifying unit 310 of FIG. 1 to produce a more accurate and efficient numerical map. This may be performed by the data classifying unit 350. For example, the data may be classified into ground data and non-ground data for numerical mapping.

On the other hand, since the obtained data includes a mechanical error or a stochastic error, it is necessary to correct this. Therefore, the data correction unit 215 may correct the data for accuracy, efficiency, and freshness of the obtained data. The data corrector 215 may include a rotation amount corrector 215-1, a scale corrector 215-2, and a height corrector 215-3.

The rotation amount correction unit 215-1 may correct an error of the rotation amount of the Y axis generated during the aerial laser survey. For example, the error value of the Y-axis rotation amount may be corrected by measuring the distance between the coordinate value of the entry point of the aerial laser survey data and the outermost ground survey coordinate value of the building. At this time, the sign of the correction amount of the rotation amount error value may vary depending on the flight direction. For example, when the aviation laser survey data is located above the outline of the building in the same direction as the flight direction, the correction amount may have a positive value.

The X-axis direction of the laser scanning unit 212 may be the same as the traveling direction of the plane. Therefore, the correction can be performed by setting the roof surface of the building having a flat surface in a certain section in which the relief displacement of the height value is large in the terrain and the boundary surface as in the building. For example, the height difference between the outer point of the building actually measured with respect to the building in the flight direction and the points obtained through laser surveying may be calculated and corrected.

In addition, the rotation amount correction unit 215-1 may correct an error in the rotation amount of the X-axis generated during the aerial laser survey. When the laser pulse is scanned from side to side orthogonally to the flight direction at a constant scan width, the positional information may vary greatly according to the change in the amount of rotation of the X axis. Therefore, in the present invention, the aerial laser survey may be performed along the transverse direction of the building so that the rooftop surface of the building can be surveyed to correct the change in the rotation amount of the X axis.

In addition, the scale correction unit 215-2 may correct the scale of the distance measurement generated during the aerial laser survey. In the present invention, scale correction may be performed in a form orthogonal to the ground reference point arrangement. To do this, it is necessary to survey the ground reference point in advance within the range of the scan width of the aerial laser survey on a wide and flat surface.

In addition, the height correction unit 215-3 can correct the difference between the constant height value between the ground reference point and the aerial laser survey data. In the present invention, the aerial laser survey data may be corrected to have the same height value as the ground reference point. For example, one may fly along the ground reference point arrangement to investigate a constant displacement along the Z axis. In this case, a more accurate difference in height value can be measured.

The corrected data can be composed of more accurate input data for the construction of three-dimensional spatial information, thereby producing a more precise and high-quality digital map.

3 is a schematic block diagram of a data processing unit 300 according to an embodiment to which the present invention is applied.

Referring to FIG. 3, the data processing unit 300 includes a data classifier 310, a ground data acquirer 320, a non-surface data acquirer 330, and a three-dimensional spatial information generator 340. do.

The data classifying unit 310 may classify the data into ground data and non-ground data by removing an error from data received from the aerial laser survey data receiving unit 210 or the aerial photograph data receiving unit 220 and applying a filtering process. have.

The ground data acquisition unit 320 acquires more accurate ground data by distinguishing ground data of a flat area from ground data of an inclined area, and setting a variable for continuity search of the ground data by reflecting the slope of the inclined area. can do. And road data may be generated from the obtained ground data.

The non-planar data acquisition unit 330 may generate building data from the classified non-planar data. For example, contour lines may be extracted from the non-planar data and selected as boundary data of the building. In addition, the selected boundary data may be linearized and converted into a polygon to obtain building data about a target building.

The 3D spatial information generator 340 may generate 3D spatial information using the obtained road data and building data.

The data classifying unit 310, the ground data obtaining unit 320, and the non-grounding data obtaining unit 330 will be described in more detail below with reference to FIGS. 4 to 6.

4 is a schematic block diagram of a data classifying unit 310 according to an embodiment to which the present invention is applied.

The data classifier 310 may include an error remover 311, a first filtering unit 312, and a data extractor 313.

The error removing unit 311 may first set the search area in consideration of the average elevation of the target area and the normal grid spacing. After setting the search area, the data having a large error can be removed by searching for and removing excessively high or low points in comparison with surrounding points in the search area. For example, points having coordinate values out of a predetermined error value range based on the coordinate values of the surrounding points may be removed when classifying data. Through this process, more accurate data can be obtained by removing data having a large error value in advance.

The first filtering unit 312 may classify the data into ground data and non-ground data by performing a first filtering process on the data from which the error is removed. For example, a method of first searching for a minimum value among height values of a point within a specific area, setting only points having a height value larger than the minimum value as a comparison target value, and comparing the comparison target value with a preset threshold value. Can be used. Points where the comparison target value is greater than the threshold may be classified as non-ground data, and points where the comparison target value is less than or equal to the threshold may be classified as ground data. In this case, Equation 1 may be used.

Here, R _ij indicates an arbitrary cell in the specific region, and R _{l _ min} indicates the minimum height value in the specific region in the iteration number lth. In addition, T represents a threshold value, where the threshold value may be a value measured by an experiment or may mean the minimum value. The threshold value T may be determined by Equation 2 below.

Here, s represents the maximum inclination of the target area, and M _l represents the window size of the first iteration number.

By performing the iterative operation through the above process, the data extracting unit 313 can gradually classify and extract the ground data and the non-ground data.

5 is a schematic block diagram of a ground data acquisition unit 320 according to an embodiment to which the present invention is applied.

The ground data acquirer 320 may include a variable calculator 321, a threshold value comparer 322, an inclination determiner 323, and a data determiner 324.

The variable calculator 321 calculates a variable used to search for an area in violation of the continuity of the points constituting the ground among the ground data classified by the data classifying unit 310. For example, the data forming the ground in an area divided into grids of a predetermined size have a similar distribution of height values as compared to other data in the grid. In the present invention, experiments were carried out using real data to determine the allowable height difference in a flat area, and when the ground data were composed of data having a height difference within at least 8%, the most desirable result was obtained. Could. Therefore, the height variable H may be obtained by dividing the difference between the height value of the highest point and the lowest point in the lattice by the grid size as shown in Equation 3 below.

Here, H represents the height variable, Max (h) represents the height value of the highest point in the grid, Min (h) represents the height value of the lowest point, L represents the size of the grid.

The threshold comparison unit 322 may acquire more precise ground data by comparing the obtained height variable H with a predetermined threshold value. For example, when applying an optimal threshold value of 8% according to the experimental results, if H> 0.08 in Equation 3, it is determined that the grid is not suitable to classify as ground data, and if H ≤ 0.08, the grid is ground It can be judged to be suitable for classification as data. If it is determined that the data in the grid is suitable to be classified as the ground data, the threshold comparison unit 322 transmits the determined data to the data determination unit 324. On the other hand, if it is determined that the data in the grid is not suitable to classify the ground data, the data in the grid is transmitted to the slope determination unit 323.

The inclination determination unit 323 calculates the inclination of the data determined to be inappropriate and transmits the calculated inclination to the variable calculator 321. The variable calculator 321 may calculate the modified height variable mod_H by reflecting the inclination input from the inclination determination unit 323. The modified height variable mod_H may be obtained by Equation 4 below.

Here, mod_H represents a modified height variable, H represents an already obtained height variable, θ represents an inclination angle, and C represents an inclination factor. At this time, the inclination factor (C) may be a value in consideration of the inclination of the general actual slope topography, for example, may have a value of 0.7 according to this experiment. By modifying the inclination of the inclined topography by Equation 4, it is possible to obtain a modified variable mod_H. The modified variable mod_H thus obtained is transmitted to the threshold comparator 322 again, and according to the comparison result of the threshold comparator 322, a loop of the previous process or a data decision unit is finally performed. 324 is sent.

The data determining unit 324 determines only the data finally received as the ground data by receiving only the data determined to be suitable as the ground data through the threshold comparison unit 322. The determined ground data is transmitted to the 3D spatial information generator 340 to generate a more accurate 3D road model.

FIG. 6 illustrates an embodiment to which the present invention is applied to explain a method for generating a 3D road model from ground data.

First, road boundary information may be extracted from numerical map data stored in the database unit 400. Based on the road boundary information, points corresponding to a road boundary may be extracted from the classified ground data. For example, after dividing an arbitrary polygon into a convex polygon, it may be determined whether a point in the road area exists inside one of the divided convex polygons. According to the determination result, adjacent points of neighboring points may be set for each of the extracted points in the road area, and initial areas may be generated based on the points. The region may be expanded while including the surrounding points from the region which is likely to be a real meaningful plane among the generated initial regions. The extended region may include coefficients of the plane equation, points included in the region, and boundary information thereof.

As such, surface areas may be created for each point within the road boundary, and the surface areas may be grouped into a single road surface group based on edge lengths and height differences between the surface areas. For example, the criteria for grouping into road surface populations can be set to connectivity and relative protrusion between two adjacent surface areas. Here, the connectivity is to find the edges adjacent to each other in the two regions, and to calculate the degree of connectivity between them, the closer the distance between the edges, the longer the length of each edge, and the point of the region containing the edge The denser the density, the higher the connectivity can be defined. The connectivity may be defined as in Equation 5 below.

Where θ _c (S ₁ , S ₂ ) is S ₁ and S ₂ Indicates connectivity between S ₁ and S ₂ Each represents a surface area. θ _c (S ₁ / S ₂ ) is S ₂ S ₁ based on surface area The degree of connection with the surface area is shown.

In addition, the protrusion may be calculated based on a vertical height difference between two horizontally adjacent regions, and may be set based on finding adjacent edges of two regions and their vertical height differences. The protrusion may be defined as in Equation 6 below.

As in Equation 6, the protrusion may be defined as the sum of protrusions with all adjacent regions S _k , which is calculated by considering the length of the edge closest to the region and the height difference between the region and the other region. Can be. Where θ _elev Represents the protruding from all regions, L (e _i ) represents the length of the edge closest to the region, and D (e _{i , 1} , e _{j , k} , z) is between the region and the other region. Indicates the height difference.

The three-dimensional road model may be generated using the points forming the grouped road surface group and the road boundary information. In this case, since the boundary of the actual road surface group is irregular, the boundary data may be smoothly corrected by using the road boundary of the digital map.

Referring to FIG. 6, the black dots in FIG. 6 represent road surface points in aerial laser survey data. Among them, the outermost points connected by thick dotted lines may be referred to as road boundary points 401 that can be known from aerial laser survey data. Then, the aerial laser survey data road boundary point 401 is repaired to the road boundary line 402 on the numerical map. In this case, the generated intersection may be referred to as a newly generated road boundary point 403. The horizontal position of the new road surface boundary point may be calculated by calculating the position of the newly generated road boundary point. The aviation laser survey data points existing within a certain distance on the horizontal position about the calculated intersection can be retrieved (404) from which the plane coefficient can be calculated. In addition, new three-dimensional road boundary data may be generated by substituting the horizontal coordinates of the intersection points into a plane equation to calculate the height values. In this case, the generated 3D road boundary data may be used to generate a 3D road model by a digital elevation model (DEM), a triangulated irregular network (TIN), a contour data representation method, or the like.

7 is a schematic block diagram of the non-planar data acquisition unit 330 as an embodiment to which the present invention is applied.

The non-surface data acquirer 330 may include a second filter 331, a polygon generator 332, and a weight center acquirer 333.

The second filtering unit 331 may extract building data from the non-planar data received from the data classifying unit 310. For example, a local maximum filtering method may be applied as the second filtering process. The local maximum filtering method is a method of dividing building data in a vector domain, and can be used as it is without interpolating aerial laser survey data having a point format of a point data.

In addition, the polygon generator 332 may generate a unit polygon from the extracted building data by using an irregular triangular network (TIN) model. Area conditions can then be used to isolate more accurate building data. For example, by selecting the minimum building area within the target area and removing smaller polygons, only more accurate building data can be obtained. In this case, the minimum building area may be based on the minimum area building on the 1/1000 numerical map.

The center of gravity acquisition unit 333 may acquire an external expression element of the aerial image by using a center of gravity that is a point representing a polygon composed of point data among the acquired building data. For example, if there is an arbitrary polygon and the outline of the arbitrary polygon is expressed as a function of (x (t), y (t)), the area A of the polygon surrounded by the outline is same.

At this time, x (t) is approximated to 0.5 (x (t + 1) + x (t)) and y (t) to 0.5 (y (t + 1) + y (t)), and x '( If we approximate t) as (x (t + 1)-x (t)) and y '(t) as (y (t + 1)-y (t)), the equation 5 for the coordinate function It can be written as Equation 8 below by discrete coordinate pairs.

At this time, the coordinate of the center of gravity of the polygon can be obtained by the following equation (9).

 If Equation 9 is rearranged in the form of coordinate pairs, the coordinate of the center of gravity is expressed as Equation 10 below.

On the other hand, the aerial image is rich in texture information, it is easy to visually determine the roof surface of the building. Therefore, after acquiring the image coordinates for the corner point of the building manually, the center of gravity can be obtained by calculating the area. The height value of the center of gravity of the building information obtained from the aerial laser survey data may be obtained by interpolating the height values of the point data constituting the building from an irregular triangular network model.

As described above, the non-surface data acquisition unit 330 may use a center of gravity point to find a common object between aerial photography and aerial laser survey data. In addition, an external facial expression element of the aerial photograph is calculated using the center of gravity as reference information, and is transmitted to a digital map generator (not shown) or a 3D spatial information generator 340, and the digital map generator generates the external facial expression. A digital map is generated based on the element, and the 3D spatial information generator 340 generates a 3D building model using the external facial expression element.

As another embodiment to which the present invention is applied, the image display unit 500 may display a clearer image by applying an image processing function in an image post-processing process.

An embodiment of applying an image processing function in an image post-processing process will be described. For example, a method of image processing in consideration of inverse filtering in the YUV domain will be described. Since each pixel has all RGB values, it is possible to perform image processing on a luminance channel by converting from an RGB domain to a YUV or YCbCr domain. For example, a case of processing an image by converting to a YUV channel will be described. The high frequency component of the Y channel can be obtained by filtering with a high pass filter in the Y channel. The relational expressions are shown in Equations 11 and 12.

Here, Y '(i, j) refers to a high frequency component obtained at the (i, j) position of the Y channel. HF _y (i, j) represents the high frequency component in the Y channel, and H _HPF is a high pass filter. When using α, β, and γ, high frequency components can be adjusted according to regions. As a result, a more natural image can be obtained, and deterioration due to unwanted noise can be prevented. The image enhancement may be performed by using the high frequency component obtained from the Y channel as shown in Equation 13.

는 개선된 Y 채널 값을 나타내고, E{Y(i,j)}는 획득한 영상 Y채널 영상의 평균을 의미한다.here

Denotes an improved Y channel value, and E {Y (i, j)} denotes an average of the acquired image Y channel image.

Image enhancement is performed only on the Y channel and not on the U and V channels. The improved Y channel value and the original U and V channel values are inversely transformed into the RGB domain to finally obtain an improved image.

FIG. 8 is a flowchart for extracting ground data and non-ground data from an aerial laser survey data according to an embodiment to which the present invention is applied.

First, the aviation laser survey data receiving unit 210 receives aviation laser survey data through a web server (S810).

The error removing unit 311 sets a search area of the received aerial laser survey data and removes error data exceeding a preset error value range based on coordinate values of surrounding points in the search area (S820).

The first filtering unit 312 performs a first filtering process on the aviation laser survey data from which the error data has been removed, and classifies it into ground data and non-ground data, but the first filtering process is first performed within a specific area. A minimum value among the height values is obtained and only points having a height value larger than the minimum value are set as a comparison target value (S830). In operation S840, the comparison target value is compared with a predetermined first threshold value. If the comparison target value is larger than the predetermined first threshold value, the data is classified as non-ground data (S850). If the comparison target value is smaller than or equal to the predetermined first threshold value, the data is classified as ground data (S860). In this case, Equation 1 [R _ij -R _{l _ min} > T] (where R _ij indicates an arbitrary cell in a specific region, and R _{l _ min} denotes the minimum height value in the specific region in the l th iteration number). And T represents a first threshold value, wherein the first threshold value is determined by Equation 2 [T = sM _{l ]} , where s represents a maximum slope of the target area and M _l represents a repeat count l. Second window size).

The data extraction unit 313 extracts the ground data and the non-ground data classified as described above, and generates three-dimensional spatial information using the ground data and the non-ground data.

As mentioned above, preferred embodiments of the present invention are disclosed for purposes of illustration, and those skilled in the art can improve and change various other embodiments within the spirit and technical scope of the present invention disclosed in the appended claims below. , Replacement or addition would be possible.

Claims (3)

An aviation laser survey data receiving unit for receiving aviation laser survey data through a web server;
A first filtering process is performed on the aerial laser survey data and classified into ground data and non-ground data. The first filtering process includes the following equations (R _ij -R _{l_min} > T, where R _ij is a specific region). _Refers to any cell within, R _{l_min} refers to the minimum height value in a particular region at the l < th & gt _; iteration, and T represents the first threshold value, which is determined by the equation T = sM _l Where s represents the maximum inclination of the target area and M _l represents the window size of the l < th > repetition frequency.
A second filtering unit which extracts point data corresponding to a building from the classified non-planar information using a second filtering method;
A polygon generator which generates a polygon from point data corresponding to the extracted building by using an irregular triangle network model;
The center of gravity of the polygon is calculated from the point data corresponding to the building, and the outline of the polygon is called a function called (x (t), y (t)), where t represents a real parameter. If the area of the polygon is A,

,

A weight center point obtaining unit calculating the weight center point based on the weight center point; And
A digital map generator for acquiring an external facial expression element of the aerial image using the center of gravity and generating a digital map based on the external facial expression element
Digital map construction system using aerial photography and aerial laser survey data, characterized in that it comprises a. delete delete

Images