JP7614426B2 - Observation System - Google Patents

Description

This disclosure relates to observation systems, observation satellites, ground equipment, and programs.

In conventional technology, image resolution is limited by the pixel pitch of the detector of the optical observation device mounted on the observation satellite. Therefore, when improving the resolution of observing the Earth from a geostationary orbit about 36,000 km above the equator, the aperture diameter and focal length are restricted by the dimensions of the rocket fairing. For this reason, conventional technology has had the problem of making it difficult to achieve high resolution (for example, Patent Document 1).

JP 2008-107941 A

The purpose of this disclosure is to provide an observation system that increases the resolution of images without relying on the pixel pitch of the detector.

The observation system according to the present disclosure includes:
An observation satellite equipped with an optical observation device that captures high-resolution images and has an aperture diameter and focal length that are not restricted by the dimensions of a rocket fairing when observing the Earth from above a geostationary orbit, and a communication device;
and a ground facility for receiving and imaging image data captured by the optical observation device, the ground facility being adapted to image an object on the earth or in space,
The observation satellite,
A line of sight vector of the optical observation device is moved to capture an image of the same object to be imaged a plurality of times by the optical observation device, thereby generating image data of each of the plurality of images of the same object to be imaged, and transmitting the image data of each image to the ground by the communication device;
The ground equipment includes:
A communication control unit that receives imaging data of the same imaging subject each time;
an image control unit that performs super-resolution processing for processing each captured image data using Bayesian estimation;
Equipped with
The optical observation device includes a two-dimensional detector,
The two-dimensional detector includes:
During a time t required from capturing an image at a viewing angle η of the optical observation device corresponding to the number of pixels m in the line of sight vector movement direction to capturing an image at the next adjacent viewing angle η by moving the line of sight vector, (k-1) images are captured at a time interval of t/k for a prime number k that is a prime number of 2 or more other than a divisor of the number of pixels m and is a prime number other than 2, 3, and 5, and each piece of imaging data is obtained;
The communication device includes:
Each imaging data acquired by the two-dimensional detector is transmitted to the ground.

This disclosure provides an observation system that can increase the resolution of images without relying on the pixel pitch of the detector.

FIG. 1 is a diagram of the first embodiment, showing the configuration of an observation system 1000. FIG. 11 is a diagram of the first embodiment for explaining oversampling processing in the case of push-broom imaging. FIG. 11 is a diagram of the first embodiment for explaining oversampling processing in the case of push-broom imaging. FIG. 1 is a diagram for explaining Bayesian estimation in the first embodiment. FIG. 1 is a diagram for explaining Bayesian estimation in the first embodiment. FIG. 11 is a diagram showing the relationship between the line-of-sight vector movement distance x and GSD in the first embodiment. FIG. 11 is another diagram showing the relationship between the line-of-sight vector movement distance x and GSD in the first embodiment. FIG. 1 is a diagram of the first embodiment, showing a situation in which an observation satellite 100 flying in a geostationary orbit observes an imaging target 400 on the Earth's surface 501. FIG. 11 is a diagram of the first embodiment, showing observation of an image capture target 400 on the Earth's surface 501 from a geostationary orbit in a state where φ<θ/N deg. FIG. 11 is a diagram showing a comparative example in the first embodiment. FIG. 11 is another diagram showing a comparative example in the first embodiment. FIG. 11 is a diagram of the first embodiment, showing an observation method when assuming the observation area of FIG. 10 . FIG. 2 is a diagram showing the hardware configuration of an observation satellite 100 according to the first embodiment.

In the description of the embodiments and drawings, the same elements and corresponding elements are given the same reference numerals. Descriptions of elements given the same reference numerals are omitted or simplified as appropriate. In the following embodiments, "part" may be read as "circuit," "process," "procedure," "process," or "circuitry" as appropriate.

Embodiment 1.
An observation system 1000 according to the first embodiment will be described with reference to FIGS.
1 shows the configuration of an observation system 1000. The observation system 1000 includes an observation satellite 100 and a ground facility 200. The observation system 1000 is a system that captures images of the Earth 500 or features in space. The observation satellite 100 includes an optical observation device 110, a field of view changing device 120, and a communication device 130. The optical observation device 110 includes an optical system 111 and a detector 112.

<Observation System 1000>
The ground facility 200 receives the imaging data captured by the optical observation device 110 and converts it into an image. The observation satellite 100 moves the line of sight vector of the optical observation device 110 to capture images of the same imaging target multiple times with the optical observation device 110, generates imaging data for each of the multiple times of the same imaging target, and transmits the generated imaging data for each time to the ground via the communication device 130. The ground facility 200 includes a communication control unit 211 that receives the imaging data for each time of the same imaging target, and an image control unit 212 that performs super-resolution processing to process the imaging data for each time using Bayesian estimation as the image processing target.

***Configuration Description***
The hardware configuration of the ground equipment 200 will be described with reference to FIG. 1. The ground equipment 200 is a computer. The ground equipment 200 includes a processor 210 and other hardware such as a main memory device 220, an auxiliary memory device 230, an input IF 240, an output IF 250, and a communication IF 260. IF indicates an interface. The processor 210 is connected to other hardware via a signal line 270 and controls the other hardware. The ground equipment 200 includes a communication control unit 211 and an image control unit 212 as functional elements. The communication control unit 211 executes a reception process of the imaging data transmitted from the observation satellite 100. The image control unit 212 executes image processing using the imaging data. The functions of the communication control unit 211 and the image control unit 212 are realized by the cooperation of the processor 210 and a program 201. The software is stored in the auxiliary memory device 230. The ground side communication device 310 is connected to the communication IF 260. The communication control unit 211 communicates with the communication device 130 of the observation satellite 100 via a ground-side communication device 310 and a ground-side antenna 320 .

Program 201 is a program that causes a computer to execute each process, procedure, or step of communication control unit 211 and image control unit 212, with the "part" replaced with "process," "procedure," or "step." The observation method is a method that is performed by ground equipment 200, which is a computer, executing program 201. Program 201 may be provided by being stored in a computer-readable recording medium, or may be provided as a program product.

Next, observation by the observation satellite 100 will be described. Ground Sampling Distance (hereinafter, GSD) is a commonly known index of the resolution of the optical observation device 110 mounted on the observation satellite 100. In the example of Earth observation, the ground distance corresponding to the pixel pitch of the detector 112 in the ground image projected onto the detector 112 via the optical system 111 corresponds to the GSD. The viewing angle corresponding to the pixel pitch of the detector 112 is called the instantaneous viewing angle θ. For this reason, in the case of an optical observation device 110 equipped with a two-dimensional detection element 114E (described later) arranged two-dimensionally, the GSD of the image obtained in one imaging operation depends on the pixel pitch of the two-dimensional detection element 114E.

<Oversampling processing>
In response to this, a process called oversampling is known, which is known to produce an image with a higher spatial resolution than GSD, which depends on the pixel pitch of the detector 112, by synthesizing two images captured by shifting the image vertically and horizontally by, for example, 1/2 the pixel pitch.

<Pushbroom imaging>
2 and 3 are diagrams for explaining the oversampling process in the case of push broom imaging. An embodiment of the oversampling process in push broom imaging by the one-dimensional detector 113 will be described below. FIG. 2 shows one-dimensional detectors 113A and 113B as the one-dimensional detector 113. An example of the one-dimensional detector 113 is a CCD detector. The one-dimensional detector 113 has a plurality of one-dimensional detection elements 113E arranged in a row. Push broom imaging is an imaging method in which an optical observation device 110 equipped with the one-dimensional detection element 113E operates the one-dimensional detection element 113E according to the desired amount of movement of the line of sight vector while moving the line of sight vector in a direction 115 perpendicular to the detection element row of the one-dimensional detection element 113E, thereby generating a two-dimensional image. As shown in FIG. 2, the one-dimensional detector 113B is arranged at 0.5 pixels in the detection element row direction (CT direction) and at n+0.5 pixels (n is an integer) in the direction perpendicular to the pixel array (AT direction) based on the one-dimensional detector 113A.
3 shows imaging from one-dimensional detector 113A and one-dimensional detector 113B. When imaging is repeated for each pixel movement in direction 115 perpendicular to the detector element row, a two-dimensional image (black circle 141A) by one-dimensional detector 113A and a two-dimensional image (double circle 141B) by one-dimensional detector 113B are imaged. White circles 141C indicate points that are not imaged. The distance between pixel centers of one-dimensional detector 113A and one-dimensional detector 113B is 1/√2 times (1/2 ^1/2 ) the pixel pitch of each detector.

In oversampling processing, information on the grid indicated by the white circle 141C that is not captured is derived by interpolation processing from surrounding data on the ground. For example, the information on a total of four grids, the grid formed by the adjacent black circle 141A and the grid formed by the double circle 141B, is added and divided by four to generate information on the white circle 141C, generating an image as if the pixel pitch were 0.5.

In Figure 3, the pixel centers of two images, an image composed of black circles 141A and an image composed of double circles 141B, are shifted vertically and horizontally by 1/2 the pixel pitch, resulting in an arrangement at a 45-degree angle with a distance between pixel centers of 1/(√2), which can be understood as having the effect of increasing the resolution of the GSD by 1/(√2) times. However, this does not mean that the instantaneous field of view angle θ is multiplied by 1/(√2). Because information from adjacent pixels is mixed in, the image does not match that obtained by an optical observation device with a pixel pitch of 1/(√2).

The principle of high resolution in the first embodiment described below is not limited to a pixel pitch shift of 1/2, nor is it limited to two images. Furthermore, the principle of high resolution in the first embodiment does not require different conditions for the vertical and horizontal directions.

<Bayesian Estimation>
It is known that Bayesian estimation of multiple images of the same target can have the effect of increasing the resolution, and this is called "super-resolution processing." This super-resolution processing enables high-resolution imaging by using multiple images taken when the line-of-sight vector of the observation satellite 100 has moved θ/N deg (N is a natural number between 2 and 100) which is smaller than the instantaneous field-of-view angle θ deg.

<Procedure for Bayesian estimation>
The procedure of Bayesian estimation is shown below. The series of procedures of Bayesian estimation described below are performed by the image control unit 212.
4 and 5 are diagrams for explaining Bayesian estimation. In FIG. 4, two images are shown: image 1 shown by a solid line and image 2 shown by a dashed line. The image control unit 212 selects one image and interpolates the values of the grid points (hereinafter, called mixels) of the desired surface model based on the values of each pixel (a pixel of an observed image is called a pixel). The dashed lines in FIG. 4 indicate the surface model, and the white circles indicate mixels. The image control unit 212 performs registration between the first image and the other images with sub-pixel accuracy, so that the distance between each mixel and each pixel can be obtained. The image control unit 212 obtains the pixel value mi of mixel i by the following formula, and sets it as the initial composite image.

(a) Likelihood The image control unit 212 calculates a likelihood function that gives the probability of the pixel value of each pixel when the pixel value of each mixel and the observation conditions (PSF, registration information) are given. Next, assuming that the observation error of each pixel is independent of the adjacent values, the following equation is obtained as the likelihood of the entire image.

尤度と先見確率との積が最大となる条件は次式で与えられる。

(b) Look-ahead Probability The degree to which the values of each mixel are related is called the look-ahead probability. Taking into account the continuity of the mixel values, the following formula is obtained. Figure 5 conceptually shows the generation of mixel values using the following formula.

The condition for maximizing the product of the likelihood and the prior probability is given by the following equation.

When the mixel spacing is large, the first term in the numerator on the right hand side of the above △mi becomes dominant, resulting in a solution similar to that of the maximum likelihood method. The second and third terms in the numerator are terms related to continuity and act as noise suppressors.

This is the procedure for Bayesian estimation.

<Super-resolution processing>
As described above, the procedure of achieving high resolution by processing multiple images of the same object captured multiple times by the optical observation device 110 in high resolution using Bayesian estimation processing is called super-resolution processing. By performing super-resolution processing, it is possible to extract information with a higher spatial frequency than the GSD of an image obtained by one capture by the optical observation device 110. However, it goes without saying that there is a limit to how much resolution can be achieved in the optical observation device 110 due to the diffraction limit of the optical system 111.

<Low Earth Orbit>
In the above explanation of the oversampling process performed by the observation satellite 100 and the super-resolution process performed by the ground facility 200, for simplicity, a GSD was used in which the line-of-sight vector is parallel to the normal vector of the "earth's surface 501 considered to be a plane," as shown in FIG. 6, which will be described later. Since the observation satellite 100 orbiting in a low orbit has a high ground speed, if push broom imaging is performed in the direction of the satellite's movement, the line-of-sight vector moves in the direction of the satellite's movement with respect to the earth's surface, even if the angle of the line-of-sight vector is constant. This makes it possible to perform oversampling and super-resolution processes.

<Observation from geostationary orbit>
On the other hand, as shown in FIG. 8, when observing the Earth 500 from an observation satellite 100 flying near a geostationary orbit (altitude approximately 36,000 km), the geostationary orbit satellite flies in synchronization with the rotation of the Earth 500. Therefore, since the ground speed of a satellite flying in a geostationary orbit is almost zero, the line of sight vector is moved by the attitude angle of the observation satellite 100 main body or the field of view changing device 120 equipped in the optical observation device 110. In this case, it is reasonable to use the instantaneous field of view angle θ as a reference rather than the GSD. For example, when observing Japan from a geostationary orbit, the GSD per pixel is different from the GSD for nadir viewing due to the spherical effect of the Earth, the oblique effect of the line of sight vector when observing the mid-latitudes from above the equator, and the fact that the distance from the optical device to the earth's surface to be observed is not uniform.
Therefore, as shown in FIG. 9 described later, the instantaneous viewing angle θdeg of the detection element of the optical observation device 110 is used as a reference, and the observation satellite 100 determines whether the movement angle φ of the line of sight vector is smaller than the instantaneous viewing angle θdeg.
θ/N deg (N is a natural number between 2 and 100)
The ground equipment 200 receives the imaging data, and the image control unit 212 converts the imaging data into an image, and then performs super-resolution processing using Bayesian estimation.

In the above formula for θ/N, when N is 2 and imaging is performed by moving θ/N degrees in both the latitudinal and longitudinal directions, this corresponds to the example in Figure 3, which explains the principle of oversampling imaging. When imaging is performed by moving θ/N degrees only in the latitudinal direction, high resolution can be expected only in the latitudinal direction. Note that the direction of movement of the line of sight vector is arbitrary. Also, as long as there is a margin up to the analytical limit of the optical system 111, the larger N is, the greater the effect can be expected.

(Gaze vector movement distance < GSD/N)
The ground facility 200 receives, as imaging data, imaging data captured in a state where the moving distance x of the line of sight vector is GSD/N deg (N is a natural number equal to or greater than 2) smaller than the Ground Sampling Distance value GSD, which indicates the ground distance per detection element 112E of the detector 112 provided in the optical observation device 110.
That is, N is an integer equal to or greater than 2,
x < GSD/N
It is.
The relationship between the line-of-sight vector movement distance x and the GSD will be specifically described with reference to Figures 6 and 7. Figure 6 shows the GSD. In Figures 6 and 7, the observation satellite 100 flies to the right from time t1 to tn. Figure 7 shows the relationship between the line-of-sight vector movement distance x and the GSD. As shown in Figure 7, the range of the GSD includes a plurality of "line-of-sight vector movement distances x". This allows multiple imaging data of the same imaging target to be acquired, enabling super-resolution processing by the ground equipment 200 using multiple images of the same imaging target.

(φ<θ/N)
The ground equipment 200 receives from the observation satellite 100 imaging data captured in a state where the movement angle φ of the line of sight vector of detection element 112E of the detector 112 equipped in the optical observation device 110 is θ/N deg (N is a natural number greater than or equal to 2) smaller than the instantaneous field of view angle θdeg of detection element 112E.
The details are explained below.
FIG. 8 shows a situation in which an observation satellite 100 flying in a geostationary orbit observes an image target 400 on the Earth's surface 501 .
FIG. 9 shows the observation of an image target 400 on the earth's surface 501 from a geostationary orbit.
φ＜θ/N deg (N is a natural number equal to or greater than 2)
FIG. 13 is a diagram showing observation in the state of
8 and 9 are the contents described in the above section "Observation from Geostationary Orbit". The angle made by the two dotted lines in Fig. 9 is φ. Therefore, even when the observation satellite 100 flies in a geostationary orbit, multiple pieces of imaging data of the same imaging target 400 can be obtained within the range of the instantaneous field of view angle θ. This allows multiple pieces of imaging data of the same imaging target to be obtained, making it possible for the ground equipment 200 to perform super-resolution processing using multiple images of the same imaging target.

(One-dimensional push broom)
The optical observation device 110 may include a one-dimensional detector 113 that generates imaging data by push-broom imaging. In this case, as described in Fig. 2 and Fig. 3, by using oversampling processing in push-broom imaging, multiple imaging data of the same imaging target can be obtained, so that super-resolution processing by the ground facility 200 using multiple images of the same imaging target becomes possible.

(Rotating the line of sight vector at a constant angular velocity)
The optical observation device 110 may employ a method in which the line-of-sight vector is rotated at a constant angular velocity, and imaging data is generated by capturing images while moving the line-of-sight vector.
The specific details are explained below.
10 and 11 are comparative examples of the observation system 1000 of the first embodiment. In the following, it will be described that the observation system 1000 can solve the problems of the comparative examples of FIG. 10 and FIG. 11. FIG. 10 and FIG. 11 show a method of performing wide-area imaging by repeating an operation called step-and-stair by an optical observation device mounted on an observation satellite flying near a geostationary orbit. The entire observation area 411 in FIG. 10 is a collection of squares with thick frames. Of the entire observation area 411, the easternmost row is called the easternmost observation area 412. Of the entire observation area 411, the row adjacent to the west side of the easternmost observation area 412 is called the western adjacent area 413. Of the entire observation area 411, the westernmost row is called the westernmost observation area 414.

Figure 11 shows the easternmost observation area 412 and the western adjacent area 413. In step-and-stair, the observation satellite repeatedly changes the line-of-sight vector of the optical observation device until the field of view of the optical observation device moves from the northernmost end of the easternmost observation area 412 to the southernmost end. In other words, the observation satellite flying near the geostationary orbit moves the line-of-sight vector of the optical observation device in the latitudinal direction (southward). The observation satellite then moves the field of view of the optical observation device from the southernmost end of the easternmost observation area 412 to the southernmost end of the western adjacent area 413 adjacent to the west. The observation satellite then repeatedly changes the line-of-sight vector of the optical observation device until the field of view of the optical observation device moves from the southernmost end of the easternmost observation area 412 to the northernmost end of the western adjacent area 413. This is step-and-stair.

In wide-area imaging by repeating step-and-steer, for example, for an observation target more than 36,000 km away from a geostationary orbit, the repeated movement and settling of the line-of-sight vector is cumbersome, and it is difficult to achieve pointing accuracy and pointing stability. In addition, when imaging while moving the line-of-sight vector, there is a problem that the image data captured once is blurred because information from adjacent pixels that have moved during the exposure time is mixed in.
In contrast, if the optical observation device 110 rotates the line-of-sight vector at a constant angular velocity and captures images while moving the line-of-sight vector to generate imaging data, it has the advantage of being able to obtain images with excellent resolution without the need for complicated step-and-stair operations.

((k-1) images are taken at time intervals of t/k)
The observation system 1000 can also employ the following method.
The optical observation device 110 includes a two-dimensional detector 114. The two-dimensional detector 114 acquires each piece of imaging data obtained by capturing (k-1) images at a time interval of t/k for an integer k of 2 or more other than a divisor of the number of pixels m during a time t required from capturing an image at a viewing angle η of the optical observation device 110 corresponding to the number of pixels m in the line-of-sight vector movement direction to capturing an image at the next adjacent viewing angle η due to movement of the line-of-sight vector. The communication device 130 transmits each piece of imaging data acquired by the two-dimensional detector 114 to the ground.
Oversampling and super-resolution processing do not need to be limited to the same pixel, and as long as the pixels include the same monitoring target in the instantaneous field of view, the image control unit 212 of the ground facility 200 can extract corresponding pixels and perform super-resolution processing. Therefore, if the two-dimensional detector 114 selects and processes pixels including the same observation target within the field of view, and if multiple images that are appropriately distributed within the field of view are acquired without being limited to the instantaneous field of view of a specific pixel, oversampling or super-resolution processing is possible.
By continuously repeating the operation of capturing images multiple times between adjacent view angles η and the next view angle η while maintaining a constant moving speed of the line of sight vector, it becomes possible to obtain multiple images on which oversampling processing or super-resolution processing can be performed over a wide area. In this case, if the integer k is set to a divisor of the number of pixels of the two-dimensional detector 114, such as 2 or 3, there is a possibility that the pixel center positions will coincide by chance, so the image control unit 212 deliberately selects an integer k other than a divisor of the number of pixels m.
If (k-1) images are taken at a time interval of t/k for k other than a divisor of the number of pixels m, k pixels capturing an image of a specific point of the observation target can be obtained, and since the center positions of the k pixels do not coincide but are appropriately distributed, an image data set suitable for super-resolution processing using Bayesian estimation can be obtained. It is desirable to select k as a prime number other than 2, 3, or 5, such as 7, 11, 13, or 17.
Conventionally, observation satellites have had to perform a step-and-steer operation in which the line-of-sight vector is once fixed at a field of view η, an image is captured, and then the line-of-sight vector is moved and the line-of-sight vector is again fixed at an adjacent field of view η to capture an image. However, with this method, the line-of-sight vector can be moved at a constant speed while ensuring a sufficient imaging interval, making it possible to obtain an image set suitable for super-resolution processing.

(The line of sight vector movement direction is at an angle of arctan (1/m) rad or more)
Furthermore, the observation system 1000 can also employ the following method: The strict line-of-sight vector movement direction with respect to the number of pixels m in the line-of-sight vector movement direction of the two-dimensional detection element 114E is separated by an angle of arctan (1/m) rad or more from the direction of the detection element row closer to the line-of-sight movement direction.
This separation will now be described.
The observation system using the method of "(k-1) images taken at a time interval of t/k" enables super-resolution processing using multiple images of the same imaged object in the direction of movement of the line of sight vector. However, when the direction of movement of the line of sight vector and the direction of the row of detection elements are exactly the same, there is a problem that super-resolution processing cannot be performed in the direction perpendicular to the direction of movement of the line of sight vector. Therefore, by making the strict direction of movement of the line of sight vector separate from the direction of the row of detection elements by an angle of arctan (1/m) rad or more, super-resolution processing becomes possible in the direction perpendicular to the direction of movement of the line of sight vector.

(Mixture of one-dimensional detector 113 and two-dimensional detector 114)
The observation system 1000 may also adopt the following method. Specifically, the method is as follows. The optical observation device 110 includes a one-dimensional detector 113 having a plurality of one-dimensional detection elements 113E arranged one-dimensionally, and a two-dimensional detector 114 having a plurality of two-dimensional detection elements 114E arranged two-dimensionally. The communication device 130 transmits to the ground a plurality of one-dimensional imaging data, which is imaging data obtained by push broom imaging by each one-dimensional detection element 113E in a direction substantially perpendicular to the element row of the one-dimensional detection element 113E, and a two-dimensional imaging data, which is imaging data by the two-dimensional detection element 114E. The ground equipment 200 receives the plurality of one-dimensional imaging data and the two-dimensional imaging data by the communication control unit 211. The image control unit 212 images each one-dimensional imaging data of the plurality of one-dimensional imaging data, and then estimates an error in the line-of-sight vector movement speed due to the push broom imaging using the two-dimensional imaging data. The image control unit 212 then performs geometric correction on each image that has been imaged from each one-dimensional imaging data, and executes super-resolution processing using each geometrically corrected image.

In the push broom imaging of the one-dimensional detector 113, the positional accuracy of the acquired image in the moving direction of the line of sight vector depends on the accuracy of the line of sight vector moving speed. Therefore, when the push broom imaging data is imaged by the image control unit 212, there is a problem that image distortion occurs if there is an error or detection error in the line of sight vector moving speed.
On the other hand, the imaging data captured by the two-dimensional detection element 114E can accurately represent the line of sight vector movement direction and the relative positional relationship in the orthogonal direction at a specific moment.
The image control unit 212 compares the image generated by the push broom imaging with the image generated from the imaging data of the two-dimensional detection element. If there is a difference as a result of the comparison, it becomes apparent that there is an error in the line of sight vector movement speed of the one-dimensional detection element 113E, and the image control unit 212 can compensate for the speed error using correct position information from the two-dimensional detection element 114E.
Furthermore, even if there is an angle error in the line of sight vector movement direction of the one-dimensional detector 113, the error can be estimated from the difference in the images. This has the effect of correcting geometric errors contained in the image captured by the push broom. The geometric error may also be called a position error. In general, the one-dimensional detection element 113E is capable of wide-area observation in the observation width direction, so that there is an effect of being able to obtain an accurate high-resolution image of a wide area by using the narrow-area observation image by the two-dimensional detection element 114E.

(Flying in geostationary orbit with a two-dimensional detector 114 installed)
The observation system 1000 may adopt the following system: The observation satellite 100 flies along a geostationary orbit. The optical observation device 110 includes a two-dimensional detector 114, and observes the mid-latitudes from above the equator using the two-dimensional detector 114.

If an optical observation device 110 equipped with a two-dimensional detector 114 with equal vertical and horizontal pixel pitches were to capture an image of Japan, which is located in the mid-latitudes, from above the equator, there would be a problem in that the Earth's surface equivalent distance per pixel would be longer in the latitudinal direction than in the longitudinal direction due to the strabismus effect. Therefore, by performing super-resolution processing in the latitudinal direction, it would be possible to obtain images with the same resolution in both the longitudinal and latitudinal directions.

(geostationary orbit flight, push broom)
The observation system 1000 may also employ the following system: The observation satellite 100 may fly near a geostationary orbit.
Based on Fig. 12, a wide-area imaging method when flying near a geostationary orbit will be described. Fig. 12 assumes the observation area of Fig. 10. The one-dimensional detector 113 of the optical observation device 110 has 16 visible detection elements arranged in a horizontal row. The observation satellite 100 continues to change the line-of-sight vector of the optical observation device 110 until the field of view of the optical observation device 110 moves from the northernmost end of the easternmost observation area 412 to the southernmost end. In other words, the observation satellite 100 moves the line-of-sight vector of the optical observation device 110 in the latitudinal direction (southward). During this time, the optical observation device 110 continues imaging. After that, the observation satellite 100 moves the field of view of the optical observation device 110 from the southernmost end of the easternmost observation area 412 to the southernmost end of the observation area adjacent to the west (west adjacent area 413). After that, the observation satellite 100 continues to change the line of sight vector of the optical observation device 110 until the viewing range of the optical observation device 110 moves from the southernmost end to the northernmost end of the observation area adjacent to the west side. Then, this operation is repeated, and the viewing range of the one-dimensional detector 113 moves to the westernmost end of the entire observation area 411.

In an observation system 1000 equipped with a high-resolution optical observation device 110 that constantly observes the Earth 500 from near a geostationary orbit, it may take a whole day to comprehensively observe the vicinity of Japan. As the saying goes, "the sun rises in the east and sets in the west," the position where sunlight is most suitable for sunlight conditions shifts from east to west over time.
For this reason, it is reasonable to capture images from east to west, as shown in Fig. 12. It is also reasonable to first capture images in the latitudinal direction using push broom capture, then move the line of sight vector to an area with an adjacent observation swath on the west side, and then capture images using push broom capture in the latitudinal direction again. When capturing images using push broom capture in the latitudinal direction, it is possible to make the ground distance in the latitudinal direction equal to that in the longitudinal direction by repeating the capture under conditions where the angle of movement of the line of sight vector is smaller than the instantaneous field of view angle θ, and it is also possible to obtain the effect of increasing the resolution by super-resolution processing.

(elliptical orbit, push broom)
The observation system 1000 may employ the following method: The observation satellite 100 flies in an elliptical orbit and performs push broom imaging in the satellite's flight direction. The observation satellite 100 maintains the imaging period near the perigee and also captures images near the apogee.
When an observation satellite flies in a low-earth elliptical orbit and observes the Earth, if the instantaneous field-of-view angle θ is constant, the image will be high-resolution with a small GSD at the perigee, and low-resolution with a large GSD at the apogee. Therefore, when taking push-broom images, the orthodox method is to make the imaging period longer near the apogee than the imaging period near the perigee.
On the other hand, if push broom imaging is performed at an imaging period equivalent to that of the perigee, multiple images can be generated that are overlapped with respect to the instantaneous field of view angle θ in the satellite's flight direction, making oversampling processing and super-resolution processing possible. By performing super-resolution processing, it is possible to obtain images with a higher resolution compared to the ground distance near the apogee. In addition, since the ground speed near the apogee is slower than near the perigee, it is possible to observe a specific observation target for a long period of time.

(WGS84)
The observation system 1000 may also employ the following method: The ground facility 200 transmits an imaging command including the position coordinates in the Earth-fixed coordinate system of the imaging target to the observation satellite. Specifically, the communication control unit 211 generates an imaging command including the position coordinates and transmits it to the observation satellite 100 via the ground communication device 310.
WGS84 is known as an earth-fixed coordinate system, and is used in the GPS and other navigation and positioning satellite systems.
In an observation satellite observing the earth, if the position coordinates of the imaging target are specified, the flying position and velocity vector of the observation satellite relative to the earth-fixed coordinate system are known. Therefore, the line-of-sight vector of the optical imaging device is set in front of the imaging target, and an appropriate line-of-sight vector movement is applied, enabling accurate super-resolution processing. For an orbiting satellite flying at a low altitude, it is reasonable to apply the line-of-sight vector movement by utilizing the satellite's ground speed, and for a geostationary satellite, it is reasonable to apply the line-of-sight vector movement by changing the satellite's attitude angle. Furthermore, by appropriately combining the satellite's ground speed and attitude angle fluctuation, a desired line-of-sight vector movement speed can be achieved with respect to the imaging target, so that any number of imaging data suitable for super-resolution processing can be obtained.

(Space object)
The observation system 1000 can also employ the following method. The same imaging target is a space object. In recent years, with the increase in the number of space objects, monitoring activities for space objects have been strengthened for the purpose of avoiding the risk of debris collisions with artificial satellites that play an active role as social infrastructure. As a means of monitoring activities, there is an increasing need to monitor space objects such as debris with high resolution using observation satellites.

The orbital information of a space object is known in advance by a space situation monitoring system called SSA (Space Situation Awareness), and the flight position of a space object at a specific time in the future can be known as position coordinates in an Earth-fixed coordinate system, and the velocity vector is also known.

The flight position and velocity vector of the observation satellite at a specific time are also known. Therefore, the communication control unit 211 can send an imaging command to the observation satellite 100, including the position coordinates in the Earth-fixed coordinate system of the specific space object to be imaged. This determines the direction of the line-of-sight vector that the optical observation device 110 equipped on the observation satellite 100 will use to capture the space object, and the relative movement direction and relative velocity can also be calculated. Therefore, by appropriately combining the relative velocity and the attitude angle change of the observation satellite, it is possible to achieve the desired line-of-sight vector movement velocity for the imaged object, and to obtain any number of imaging data suitable for super-resolution processing.

There is an increasing need to monitor space objects flying near geostationary orbit from observation satellites flying near geostationary orbit, and there is also an increasing need to monitor space objects from low-altitude orbiting satellites. When imaging space objects such as debris from observation satellites, the effect is that a fine image can be obtained by imaging the same object multiple times and performing super-resolution processing. If the space object is an artificial object, it is possible to identify whether the object is an artificial satellite or rocket remnants, and by imaging the shape at high resolution, it is also possible to identify the type of artificial satellite, such as a communication satellite or an observation satellite. In addition, pushbroom imaging allows wide-area continuous imaging, which has the effect of reliably imaging a specific object in the vastness of space.

(Additional Information on the Hardware Configuration of the Observation Satellite 100)
Fig. 13 shows the hardware configuration of the observation satellite 100. The hardware configuration of the observation satellite 100 will be described with reference to Fig. 13.

The configuration of the observation satellite 100 will be described with reference to FIG.
The observation satellite 100 comprises an optical observation device 110 , a field of view changing device 120 , a communication device 130 , an attitude control device 140 , a propulsion device 150 , a satellite control device 160 , and a power supply device 170 .

The optical observation device 110 is as described above.

The field of view change device 120 changes the observation field of view of the optical observation device 110.

The communication device 130 is a communication device for communicating with the ground equipment 200. For example, the communication device 130 receives various commands from the ground equipment 200. The communication device 130 also transmits observation data obtained by the optical observation device 101 to the ground equipment 200.

The attitude control device 140 is a device for controlling attitude elements such as the attitude of the observation satellite 100 and the angular velocity of the observation satellite 100. The attitude control device 140 changes each attitude element to a desired direction. Alternatively, the attitude control device 140 maintains each attitude element in a desired direction. The attitude control device 140 includes an attitude sensor, an actuator, and a controller. The attitude sensor is a gyroscope, an earth sensor, a sun sensor, a star tracker, a thruster, a magnetic sensor, etc. The actuator is an attitude control thruster, a momentum wheel, a reaction wheel, a control moment gyro, etc. The controller controls the actuator according to the measurement data of the attitude sensor or various commands from the ground equipment 200. The attitude control device 140 can be used to change the azimuth angle of the line of sight vector of the optical observation device 101.

The propulsion device 150 is a device that provides propulsive force to the observation satellite 100 and changes the speed of the observation satellite 100. Specifically, the propulsion device 150 is an electric propulsion device. For example, the propulsion device 150 is an ion engine or a Hall thruster.

The satellite control device 160 is a computer that controls each device of the observation satellite 100 and includes a processing circuit. For example, the satellite control device 160 controls each device according to various commands transmitted from the ground equipment 200.

The power supply unit 170 includes a solar cell, a battery, a power control device, etc., and supplies power to each device on the observation satellite 100.

The processing circuit provided in the satellite control device 160 will be described. The processing circuit may be dedicated hardware, or a processor that executes a program stored in a memory. In the processing circuit, some functions may be realized by dedicated hardware, and the remaining functions may be realized by software or firmware. In other words, the processing circuit may be realized by hardware, software, firmware, or a combination of these. Specifically, the dedicated hardware is a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination of these. ASIC stands for Application Specific Integrated Circuit.
FPGA is an abbreviation for Field Programmable Gate Array.

The above describes embodiment 1. Of the multiple technical aspects of embodiment 1, one may be implemented partially, or two or more of the multiple technical aspects of embodiment 1 may be implemented in combination.

100 Observation satellite, 110 Optical observation device, 111 Optical system, 112 Detector, 112E Detector element, 113, 113A, 113B One-dimensional detector, 113E One-dimensional detector element, 114 Two-dimensional detector, 114E Two-dimensional detector element, 115 Direction, 120 Field of view change device, 130 Communication device, 140 Attitude control device, 150 Propulsion device, 160 Satellite control device, 170 Power supply device, 141A Black circle, 141B Double circle, 141C White circle, 200 Ground equipment, 210 Processor, 211 Communication control unit, 212 Image control unit, 220 Main memory device, 230 Auxiliary memory device, 240 Input IF, 250 Input IF, 260 Input IF, 270 Signal line, 310 Ground communication device, 320 Ground antenna, 400 Imaged object, 410 Observation area, 411 Entire observation area, 412 Easternmost observation area, 413 Western adjacent area, 414 Westernmost observation area, 500 Earth, 501 Earth's surface, 1000 Observation system.

Claims (3)

An observation satellite equipped with an optical observation device that captures high-resolution images and has an aperture diameter and focal length that are not restricted by the dimensions of a rocket fairing when observing the Earth from above a geostationary orbit, and a communication device;
and a ground facility for receiving and imaging image data captured by the optical observation device, the ground facility being adapted to image an object on the earth or in space,
The observation satellite,
A line of sight vector of the optical observation device is moved to capture an image of the same object to be imaged a plurality of times by the optical observation device, thereby generating image data of each of the plurality of images of the same object to be imaged, and transmitting the image data of each image to the ground by the communication device;
The ground equipment includes:
A communication control unit that receives imaging data of the same imaging subject each time;
an image control unit that performs super-resolution processing for processing each captured image data using Bayesian estimation;
Equipped with
The optical observation device includes a two-dimensional detector,
The two-dimensional detector includes:
During a time t required from capturing an image at a viewing angle η of the optical observation device corresponding to the number of pixels m in the line of sight vector movement direction to capturing an image at the next adjacent viewing angle η by moving the line of sight vector, (k-1) images are captured at a time interval of t/k for a prime number k that is a prime number of 2 or more other than a divisor of the number of pixels m and is a prime number other than 2, 3, and 5, and each piece of imaging data is obtained;
The communication device includes:
An observation system that transmits each of the imaging data acquired by the two-dimensional detector to the ground. The optical observation device is
2. The observation system according to claim 1, further comprising a two-dimensional detector that rotates a line-of-sight vector at a constant angular velocity and captures images while moving the line-of-sight vector to generate the imaging data. The ground equipment includes, as the imaging data,
Receive imaging data captured in a state where the movement angle φ of the line of sight vector of the detection element of the detector included in the optical observation device is θ/N deg (N is a natural number equal to or greater than 2) smaller than the instantaneous field of view angle θ deg of the detection element;
3. The observation system according to claim 2.

Images