JPS59228180A - Radar image simulation system - Google Patents

Abstract

PURPOSE:To calculate the position of a photographing point on earth corresponding to each pixel of a radar image by using data for the position and attitude of a satellite and the ground surface to calculate the position of the photographing point on earth corresponding to each point on an observation image. CONSTITUTION:Line numbers of an observation are inputted into a satellite position/attitude calculating section 6 to determine the position and attitude of the satellite. A rotary matrix M between the earth reference coordinate system and the satellite reference coordinate system is found with a coordinate conversion matrix calculating section 3 and the ground surface DTM is converted to the satellite reference coordinate system with a coordinate converting section 7. A cross line between a beam plane and the ground surface DTM' converted is determined with an intersection calculating section 8. On the other hand, the beam distance to the photographing point from a satellite is determined with a beam distance calculating section 9. The photographing point is decided with a point of intersection calculating section 10 and inputted into a coordinate converting section 11 to be converted to earth reference coordinate system.

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a method for generating a simulated image, and in particular to a radar image simulation method for determining the position of a corresponding ground photographing point for each point in a satellite or aircraft-mounted rake image. Regarding the method.

[Background of the invention]

Observation of satellites, aircraft, etc. I [! ] In order to analyze and use the I-image, it is first necessary to completely correct the geometric distortion of the image so that it overlaps with the map. The distortion of the observed image is
Satellite, aircraft position at each point in the image.

It occurs due to changes in posture or the effects of terrain undulations.

In particular, when a satellite observes the entire surface of the earth diagonally below, due to the effects of topographical undulations, even if point 18 is actually photographed by VC as shown in Figure 4, if an optical sensor is used, it will be as if point 22 was photographed. In the case of radar, the image is distorted as if it were a point 23 equidistant from the satellite (parallax). FIG. 5 shows the flow of these three-dimensional distortion correction processes. Among these, observation image 26 is based on the ground elevation data 25(L)TM) and the flight conditions of the satellite.
By simulating the three-dimensional insect of
It is possible to know the correspondence between each point of the image 27 and each point of the corrected image 27 corrected by the stereoscopic distortion correction unit 28.

When simulating images observed by conventional optical sensors mounted on satellites, spacecraft, etc., an apparatus having the configuration shown in FIG. 1 is used. Each pixel of the observed image (from the position (t, p) of the basic unit of the image, based on the time when that pixel was photographed, the satellite position r (represents the entire vector) in the earth reference coordinate system 16 in Fig. 3) , the attitude θ (Euler angle) is determined from the satellite position and the cost meter nose 1.The line-of-sight vector calculation unit 2 calculates the line-of-sight vector S representing the direction of the sensor's line of sight in the satellite reference coordinate system 15 of FIG. is found. Next, 1
The rotation matrix M between the earth reference coordinate system 16 and the satellite reference coordinate system is converted into the earth reference coordinate system by the coordinate transformation unit 4, which is converted into the earth reference coordinate system by the coordinate conversion matrix calculation unit 3 according to r and θ. Convert the line of sight vector to . Then, the intersection calculation unit 5 calculates the straight line determined by the satellite position r and line-of-sight vector 1 and the ground surface DTM.
The coordinates rek of the photographing point on the earth's surface corresponding to the pixel position Ct, p) are calculated by calculating the intersection with the pixel position Ct, p). DTM is often given approximately as a set of polygons in an earth reference coordinate system.

However, when simulating a radar image, the pixel position (t
, p) is not determined by the line-of-sight vector 2 as in the case of optical sensors, but by the length of the line-of-sight. Therefore, the above-mentioned method has a problem in that it is not possible to simulate a radar image.

[Purpose of the invention]

An object of the present invention is to improve the drawbacks of the above-mentioned conventional methods and to provide an entire radar image simulation method that can determine the photographing point on the earth's surface with high precision for each pixel of an observed image in radar image simulation. vchru.

[Summary of the invention]

In order to achieve all of the above objects, the present invention expresses data on the earth's surface in a satellite reference coordinate system with the satellite position as the origin, and furthermore, expresses data on the earth's surface in a satellite reference coordinate system with the satellite position as the origin, and furthermore The length of the line of sight determined by the pixel position of the observed image is the total radius, and the intersection point of the arc on the plane of the radar pulse centered on the origin of the satellite reference coordinate system and the above-mentioned intersection line is calculated. The feature is that by calculating all the coordinates of the photographing point, the intersection point between the arc in the three-dimensional space and the plane that approaches the ground surface is calculated on the two-dimensional plane.

[Embodiments of the invention]

FIG. 2 shows a block configuration of an embodiment of the present invention.

First, the satellite emits an image line by line in a plane perpendicular to the entire traveling direction of the radar pulse, and samples the reflected waves from the ground surface over time to obtain an image of 1247 minutes.

Therefore, the photographing time of the pixel (t, p) in the observed image is determined only by the line number t and is not dependent on p. Therefore, the satellite position r and attitude θ at that time are also determined by the satellite position and attitude calculation unit 6 from only t.

The rotation matrix M between the earth reference coordinate system 16 and the satellite reference coordinate system 15 is calculated by the coordinate transformation matrix calculation unit 3 using r and θ, as in FIG.

It is assumed that the ground surface data DTM is approximated by a set of polygons n in the earth reference coordinate system. By using this fr and M, the coordinate conversion unit 7 converts it to the satellite reference coordinate system and DT
Get M'. Furthermore, as shown in FIG.
), and in the intersection calculation unit 8 with the beam plane,
Since the DTM for determining the intersection line 13 (DTM)" between this plane and the DTM' is approximated by a set of polygons, the DTM" is a broken line.

On the other hand, from the pixel (tsp), the beam distance d from the satellite 14 to the photographing point 18 is calculated in the beam distance calculating section 9 as shown in FIG.

Therefore, the photographing point is located on the plane 12 on an arc 17 centered on the position of the satellite 14 and having a radius d. The shooting point 18 is determined by using this arc 17 and the previously obtained broken line 13 (
This is done by finding the intersection with DTM"). Since these are all on the plane 12 (x = 0),
The processing only needs to be performed with respect to y and two coordinates, and is simplified to a two-dimensional problem.

Regarding the method of intersection of a circular arc and a broken line in the intersection calculation unit 10, the following processing is performed for each line segment that is a component of a broken line. Referring to Figure 3, if we calculate the entire distance between the end points 20°21 of the line segment and the satellite 14, and if the radius d of the arc 17 is between them, we can see that the arc intersects with the line segment. Karu. Next, a similar distance is determined for the center 19 of the line segment, and this is compared in magnitude with d to determine which of the midpoints the intersection 18 is located at. It can be seen in FIG. 3 that the intersection point 18 exists between the midpoint 19 and the endpoint 21. And then line segment 19-2
Similar processing is performed for 1. By repeating this process, the line segment is divided into two parts and converged to one point,
As a result, intersection point 19 is determined. Let that coordinate be white.

If the coordinates of Q are t(y, z), they are (0, y, z) in the satellite reference coordinate system 15, and are converted to the earth reference coordinate system 16 in the total coordinate conversion unit 11 using M and rf. coordinates rc
f get.

By performing this process for all pixel numbers p of the tth line of the image, the process for one line is completed. For the pixel on the (t+11th line), start by recalculating the satellite position r and the attitude.

Through the processing described above, all pixels (
t, p), the corresponding ground surface coordinates re are calculated.

〔Effect of the invention〕

As explained above, according to the present invention, there is an effect that the position of the ground photographing point corresponding to each pixel of the radar image can be calculated with full accuracy. In addition, since processing is performed on a line-by-line basis, calculations are more efficient as a side effect.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of a device for simulating an observation image using a conventional short-term optical sensor, Fig. 2 is a block diagram of a radar image simulating device according to the present invention, and Fig. 3 is a diagram showing the relationship between the beam surface and the ground surface curvature. Calculating the intersection between the intersection line and the beam? The explanatory diagrams, FIG. 4, are diagrams showing the principle of balax in images observed by an optical sensor or radar, and FIG. 5 is a schematic diagram of the three-dimensional distortion correction process of satellite images. Fig. 1 %Z Fig.￥J 3 Fig.

Claims (1)

[Claims] 1. In radar images of satellites, aircraft, etc., the position of the photographing point on the ground corresponding to each point on the observed image can be determined by using data on the position, attitude, and ground surface of the satellite. A radar image simulation method that is characterized by 2. The location of the photographing point on the ground is determined by a step of expressing the ground surface data in a coordinate system with the satellite as a reference, a step of finding the line of intersection with the surface containing the radar beam, and a step of calculating the line of intersection between this line and the pixel on the image. 2. The radar image simulation method according to claim 1, further comprising the step of calculating an intersection point with a circular arc whose radius is determined by a position in a two-dimensional coordinate system.