CN109031234B - A Fast Method for Obtaining 3D Isosurface of Radar Reflectance Data - Google Patents

Abstract

The invention provides a method for obtaining a three-dimensional isosurface of radar reflectivity data, which includes the following steps: normalizing radar three-dimensional scanning data; constructing a hexahedral grid of radar data points; and calculating a three-dimensional isosurface of radar reflectivity data. The invention can quickly obtain the three-dimensional isosurface of the radar reflectivity data, so that the radar product data can be well applied to the three-dimensional scene, and can better provide effective decision support for disaster reduction and disaster prevention of meteorological disasters, and has good social and economic benefits. At the same time, its ability to quickly obtain the shape of rainfall cloud clusters can shorten the response time of relevant departments to disasters.

Description

A Fast Method for Obtaining 3D Isosurface of Radar Reflectance Data

Technical field:

The invention relates to a method for quickly obtaining a three-dimensional isosurface of radar reflectivity data, belonging to the technical field of meteorological surveying and mapping.

Background technique:

The three-dimensional isosurface of the radar reflectivity data can express the physical characteristics of the strong echo region of the rainfall cloud very intuitively, which is of great significance for the judgment of the rainy weather. At present, the acquisition of the 3D isosurface of the conventional radar reflectivity data is based on the 3D regular grid. In this process, the arrangement of the three-dimensional regular grid is very different from the radar volume scanning method. Therefore, it is necessary to perform a large number of interpolation calculations through the existing radar detection data of each layer to obtain each three-dimensional regular small grid. The corresponding data value, and then based on this, the corresponding three-dimensional isosurface data is calculated. This method, on the one hand, needs to consume a lot of computing overhead and increase the computing time, on the other hand, it will introduce redundant data to a certain extent, increase the data volume of product data, and bring unnecessary obstacles to applications such as data transmission. . Therefore, the existing methods for calculating the isosurface of radar reflectivity data are difficult to apply to real-time 3D weather monitoring services.

Invention content:

The present invention provides a method for obtaining a three-dimensional isosurface of radar reflectivity data, and the specific technical scheme is as follows:

A method for quickly acquiring a three-dimensional isosurface of radar reflectivity data, comprising the following steps:

Normalization of radar 3D scanning data; hexahedral grid construction of radar data points; 3D isosurface calculation of radar reflectivity data; among which,

1. Normalization of radar 3D scan data: Normalize each scan layer of radar 3D scan data to a standard layer with 360 scan line data, and obtain radar scans with upper and lower corresponding data points in each scan layer data;

2. Hexahedral mesh construction of radar data points:

2.1) For the radar scan data obtained by the above normalization, take the adjacent jth and j+1th scan lines on any scan layer L and take the point coordinates of the adjacent ith and i+1th data points respectively. , a total of four points are connected as vertices to form a quadrilateral;

Similarly, take the point coordinates of the ith and i+1th data points of the adjacent jth and j+1th scan lines on the L+1 scan layer, and a total of four points are connected as vertices to form a quadrilateral;

The point coordinates taken by the Lth and L+1th scan layers are connected according to the upper and lower relationship, and a hexahedron is constructed; where L represents the number of scan layers, j represents the azimuth angle corresponding to the scan line, and j takes 0≤j≤ An integer of 359, i represents any data point on the scan line;

2.2) When the number of data points on the L+1th layer in step 2.1 is less than the number of data points on the Lth layer, that is, when the jth and j+1th scan lines in the Lth layer are obtained. Four data points, when the jth and j+1th scan lines in the L+1th layer lack the upper and lower corresponding data points, the last two scan lines corresponding to the L+1th layer are repeatedly taken. data points (used as four data points so that they can correspond to the four data points of the L layer), and the corresponding connection of the obtained data points is constructed into a hexahedron. In this case, because the L+1 layer The data points in are reused, and the constructed hexahedron is essentially a degenerate hexahedron;

2.3) Repeat steps 2.1 and 2.2, and perform the above-mentioned hexahedral construction for all data points on each adjacent scanning layer in turn, and obtain a hexahedral grid constructed from all radar reflectivity data;

3. Calculation of three-dimensional isosurface of radar reflectivity data:

The hexahedral mesh obtained in step 2 is calculated by the MC algorithm to obtain the three-dimensional isosurface data of the radar reflectivity data.

A further design of the present invention is:

In step 3, the calculation of the MC algorithm for the hexahedral mesh includes the following specific steps:

3.1) For the constructed hexahedron mesh, choose any hexahedron, and perform MC algorithm processing according to a preset reflectivity value d to detect whether each edge of the current hexahedron contains a point with a reflectivity value d, if so, then Connect the points with the reflectance value d on each edge into several triangular patches, and save these generated triangular patches; if not, process the next hexahedron.

3.2) Repeat step 3.1 to complete the calculation of the MC algorithm for each hexahedron in the hexahedral grid, and all the saved triangular patches constitute the three-dimensional isosurface data of the radar reflectivity data.

The specific process of the normalization is as follows: in each scanning layer, taking the radar antenna position as the center, constructing the first scan line with an azimuth angle of 0 degrees, constructing a scan line for each integer azimuth angle, and so on, The azimuth angle of the 360th scan line is 359 degrees, and 360 scan lines are constructed; the data on each scan line, and then based on the scan line data of each layer of the original radar three-dimensional reflectivity data, the 360 constructed 360 scan lines are obtained by interpolation calculation method. Scan line data, thereby completing the normalization processing of radar 3D scan data.

Compared with the prior art, the present invention has the following beneficial effects:

In view of the low efficiency of the conventional method of obtaining the three-dimensional isosurface of the radar reflectivity data in the current technology, it is difficult to apply it to the daily three-dimensional weather real-time monitoring business. In order to solve this problem, the present invention normalizes the scanning data of each layer of the radar, and then directly processes the data points detected by the radar to construct a hexahedral grid, and then directly obtains the radar reflectivity data by processing the hexahedron directly. 3D isosurface.

The whole process of the invention reduces the amount of calculation data and avoids redundant data, greatly improves the acquisition efficiency of the three-dimensional isosurface of the radar reflectivity data, and can be easily applied to the real-time three-dimensional weather monitoring business.

The invention can quickly obtain the three-dimensional isosurface of the radar reflectivity data, so that the radar product data can be well applied to the three-dimensional scene, and can better provide effective decision support for disaster reduction and disaster prevention of meteorological disasters, and has good social and economic benefits. At the same time, its ability to quickly obtain the shape of rainfall cloud clusters can shorten the response time of relevant departments to disasters.

Description of drawings:

1 is a schematic top view of normalized radar reflectivity data in Embodiment 1;

2 is a schematic side view of the normalized radar reflectivity data in Embodiment 1;

3 is a schematic diagram of the construction of a hexahedron of radar reflectivity data in Embodiment 1;

FIG. 4 is a schematic diagram of the construction of a hexahedron of radar reflectivity data (degenerate) in a special case in Embodiment 1;

5 is a schematic diagram of the numbering of vertices and sides of a hexahedron in Embodiment 1;

6 is a schematic diagram of 15 basic situations in the process of obtaining an isosurface in Embodiment 1;

Fig. 7 is the 40dbz reflectance value isosurface data rendering diagram in test example one;

Fig. 8 is the 50dbz reflectivity value isosurface data rendering diagram in test example two;

Detailed ways:

Example 1:

The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

The method for quickly obtaining the three-dimensional isosurface of radar reflectivity data provided by the present invention includes the following steps:

1. Normalization of radar 3D scanning data;

2. Hexahedral grid construction of radar data points;

3. Calculate the three-dimensional isosurface of radar reflectivity data; among them,

1. When normalizing the radar 3D scan data, traverse each layer of the radar volume scan data, perform the following normalization operation on the data of each layer, and normalize the data of each scan layer to have 360 scan lines The standard layer of the data, the radar scan data with the upper and lower corresponding data points of each scanning layer is obtained:

a) Take any scan layer, according to the existing scan line data in the selected layer, construct a scan line whose azimuth angle is 0 degrees, and the elevation angle is the elevation angle of this scan layer, and then obtain the scan line on the newly constructed scan line through interpolation calculation method Line data, the spatial interval of each data point is 1 km, and the number of data on the line data is the original data number (the number of libraries) of the scanning line of the layer;

b) According to the method of step a and the parameters of the elevation angle, data interval and the number of data, construct line data with azimuth angles of 1 degree, 2 degrees, 3 degrees, and up to 359 degrees in this layer respectively; The normalization operation is completed;

c) According to the above steps, perform the above normalization operation on all the echo data scan layers in the radar volume scan data in turn. After normalization, the corresponding azimuth scan line data of each scan layer of the radar corresponds to the upper and lower positions, as shown in Figure 1 and As shown in Figure 2;

2. When constructing a hexahedral grid of radar data points, perform the following operations on each layer of radar data:

In the radar scanning layer (assuming that it is currently the Lth layer), traverse each data line and perform the following operations: Assuming that the current data line is the jth root, take two adjacent data points on the data line in turn (set to The point coordinates of the i and i+1 data points), and at the same time, two adjacent data points (i and i+1) at the same position are taken on the adjacent j+1 line in this data layer (L layer). data points), and connect the four data points as vertices to form a quadrilateral; at the same time, for the L+1 data layer, also take the four data points at the corresponding positions (jth and j+1 scan lines). The point coordinates of the i-th and i+1st data points) are also connected as vertices to form a quadrilateral according to the above method; on this basis, the upper and lower data points of the upper and lower layers are A hexahedron, as shown in Figure 3, where L represents the number of scan layers, j represents the azimuth angle corresponding to the scan line, j is an integer of 0≤j≤359, i represents any data point on the scan line, and the formed hexahedron is Irregular hexahedron, an irregular hexahedron refers to a hexahedron with non-parallel opposite sides on a certain quadrilateral face (that is, in the 6 faces of a hexahedron, as long as a set of opposite sides is not parallel, the hexahedron is called an irregular hexahedron);

b) If the number of data points on the scan line in the L+1 layer is less than the number of data points on the scan line in the L layer, that is, when the data obtained from the jth and j+1th scan lines in the Lth layer The four data points of , when the jth and j+1th scan lines in the L+1th layer lack the upper and lower corresponding data points, the L layer will generate redundant quadrilaterals that cannot be matched. At this moment, Then reuse the last point corresponding to the two scan lines in the L+1 layer (two points are used as four data points so that they can correspond to the four data points of the L layer), and follow the construction of the quadrilateral Method, use the degenerate quadrilateral connected into a line (the connection line is still a quadrilateral, but its four vertices have been overlapped), and the vertices of the redundant quadrilaterals in the L layer are all connected with the degenerate quadrilateral of the L+1 layer. The vertices are connected correspondingly to form a degenerate hexahedron. In this case, since the data points in the L+1 layer are reused, the hexahedron constructed here is essentially a degenerate hexahedron, similar in shape to a triangular prism, but it can still be used. It is considered to be composed of eight vertices, but there are two sets of points that overlap, as shown in Figure 4.

c) According to the above method, all data points in the layer are constructed as hexahedrons;

d) Repeat the above steps until the scanning data points of each layer of the radar are constructed into hexahedrons. All radar reflectivity data points form a hexahedral grid, waiting for the data extraction operation of the 3D isosurface;

3. The calculation process of the three-dimensional isosurface of the radar reflectivity data is as follows:

3.1) Set the reflectance value d of the isosurface according to user requirements, traverse all the hexahedrons generated in step 2 one by one, perform MC (Marching cube) algorithm processing on them, and detect whether the current hexahedron contains reflectance values. is the triangular patch of d, if not, process the next hexahedron; if there is, get the set of triangular patches with reflectance value d contained in the current hexahedron, and save it;

The specific process of the algorithm processing of MC (Marching cube) includes the following steps:

First, set the reflectance value of the isosurface as d; and number the 8 vertices and 12 sides of the hexahedron as shown in Figure 5.

Second, the 8 vertices of the hexahedron are judged one by one, and the vertex numbers whose reflectance values are greater than or equal to d are recorded.

Third, according to the result of the vertex number that meets the requirements recorded in the second step, it is classified into the 15 possible situations in Figure 6 (this example corresponds to the situation shown in Figure 3 in Figure 6), and according to the first In the case of Fig. 3, the coordinates of each point whose reflectance value is d in the corresponding side is obtained according to the following formula, and then these obtained point coordinates are connected according to the corresponding method in Fig. 3 to form a number of triangular patches ( In this case, there are two triangular patches).

，

，对应的雷达反射率值为

，

，等值面值为d，则反射率为d的点坐标：（公式中K为中间比值变量，其表示，当前等值面数值d与坐标A点的数值V₁之差与两坐标点A、B的数值之差的比值。该比值可以用于求算坐标点A、B之间数值为d的点坐标）Let the coordinates of the two endpoints of any eligible edge be

,

, the corresponding radar reflectivity value is

,

, the isosurface value is d, and the reflectivity is the point coordinates of d: (K in the formula is the intermediate ratio variable, which means that the difference between the current isosurface value d and the value V ₁ of the coordinate point A and the two coordinate points A, The ratio of the difference between the values of B. This ratio can be used to calculate the coordinates of the point with the value d between the coordinate points A and B)

3.2) According to the process in the above step 3.1, after each hexahedron is processed, the obtained triangular patches are stored together; then these triangular patches are the isosurface triangular mesh data to be constructed.

Test example one:

According to the above inventive method, an example is processed for a weather radar in a certain region of China. And select the reflectivity value of 40dbz as the isosurface value to calculate the three-dimensional isosurface of the radar reflectivity data, and obtain the effect diagram as shown in Figure 7:

Test example two:

According to the above inventive method, an example is processed for a weather radar in Nanjing, China. And select the reflectivity value of 50dbz as the isosurface value to calculate the three-dimensional isosurface of the radar reflectivity data, and obtain the effect diagram as shown in Figure 8.

Claims (2)

1. a method for quickly obtaining the three-dimensional isosurface of radar reflectivity data, is characterized in that, comprises the steps: Normalization of radar 3D scanning data; hexahedral grid construction of radar data points; 3D isosurface calculation of radar reflectivity data; among which, 1) Normalization of radar 3D scan data: Normalize each scan layer of radar 3D scan data to a standard layer with 360 scan line data, and obtain radar scans with upper and lower corresponding data points in each scan layer data; 2) Hexahedral grid construction of radar data points: 2.1) For the radar scan data obtained by the above normalization, take the adjacent jth and j+1th scan lines on any scanning layer L and take the point coordinates of the adjacent ith and i+1th data points respectively. , a total of four points are connected as vertices to form a quadrilateral; Similarly, take the point coordinates of the ith and i+1th data points of the adjacent jth and j+1th scan lines on the L+1 scan layer, and a total of four points are connected as vertices to form a quadrilateral; The point coordinates taken by the Lth and L+1th scan layers are connected according to the upper and lower relationship, and a hexahedron is constructed; where L represents the number of scan layers, j represents the azimuth angle corresponding to the scan line, and j takes 0≤j≤ An integer of 359, i represents any data point on the scan line; 2.2) When the number of data points on the L+1th layer in step 2.1) is less than the number of data points on the Lth layer, that is, when the jth and j+1th scan lines in the Lth layer are obtained. When the jth and j+1th scan lines in the L+1th layer lack the corresponding data points in the upper and lower positions, the last data points of the two scanlines corresponding to the L+1th layer are repeatedly taken. Two data points, and the obtained data points are correspondingly connected to construct a hexahedron. In this case, due to the repeated use of data points in the L+1 layer, the constructed hexahedron is essentially a degenerate hexahedron; 2.3) Repeat steps 2.1) and 2.2), and sequentially construct the above-mentioned hexahedron for all data points on each adjacent scanning layer, and obtain a hexahedral grid constructed from all radar reflectivity data; 3) Calculation of three-dimensional isosurface of radar reflectivity data: The hexahedral mesh obtained in step 2) is calculated by the MC algorithm to obtain the three-dimensional isosurface data of the radar reflectivity data; the specific steps are as follows: 3.1) For the constructed hexahedron mesh, take any hexahedron, and perform MC algorithm processing according to a preset reflectivity value d to detect whether each edge of the current hexahedron contains a point with a reflectivity value d, if so, then Connect the points with the reflectance value d on each side into several triangular patches, and save these generated triangular patches; if not, process the next hexahedron; 3.2) Repeat step 3.1) to complete the calculation of the MC algorithm for each hexahedron in the hexahedral mesh, and all the saved triangular patches constitute the three-dimensional isosurface data of the radar reflectivity data. 2. the method for rapidly obtaining the three-dimensional isosurface of radar reflectivity data according to claim 1, is characterized in that: the concrete process of described normalization is as follows: in each scanning layer, take the radar antenna position as the center, The first scan line is constructed with an azimuth angle of 0 degrees, and each integer azimuth angle constructs a scan line, and so on, the azimuth angle of the 360th scan line is located at 359 degrees, and 360 scan lines are constructed; Then, according to the scan line data of each layer of the original radar three-dimensional reflectivity data, the 360 scan line data constructed by the interpolation calculation method are obtained, thereby completing the normalization of the radar three-dimensional scan data.

Images