KR102123424B1 - Method and system for obtaining high-resolution observation images using micro vibration in geosynchronous satellite - Google Patents

Abstract

The present invention relates to a method and apparatus for obtaining high resolution images using microscopic vibrations in geostationary orbit satellites. An object of the present invention is to obtain both the high-resolution advantages of low-orbit satellite images and the constant observation advantages of geostationary satellite images, by using micro-vibration in geostationary orbit satellites to obtain multiple images, and using SR techniques It is to provide a method and apparatus for obtaining a high-resolution image using microscopic vibration in a geostationary orbit satellite, which ultimately enables a high-resolution image to be obtained by processing.

Description

{Method and system for obtaining high-resolution observation images using micro vibration in geosynchronous satellite}

The present invention relates to an optical satellite device and method for obtaining a high resolution image using microscopic vibration in geostationary orbit. More specifically, the present invention relates to a method and apparatus for operating a geostationary satellite that acquires a low-resolution image relative to a low-orbit satellite, using a micro vibration to obtain a high-resolution image even in a geostationary satellite. .

Currently, there are numerous satellites around the earth for observing various information. Using these satellites, you can perform tasks such as sending and receiving signals through a satellite communication network, acquire weather data through atmospheric observation, and remotely observe areas inaccessible to people to objects of interest. You can also perform tasks such as obtaining information about. In particular, it is obvious that the latter works rely heavily on visual information, that is, as a high resolution image is acquired from a satellite, the resultant quality of the above-described works is improved. Therefore, resolution is one of the most important performance evaluation indicators for satellite observation performance.

The resolution of an image that can be provided by a satellite is affected by the satellite camera's performance and the altitude of the satellite. Satellites can be classified in a variety of ways depending on size, purpose of use, orbit, altitude, etc. In particular, the satellite's orbit and altitude are closely related to each other, so in general, the satellite's orbit is low orbit (300~1,500km), medium orbit (1,500~10,000km), elliptical high orbit (10,000~40,000km), depending on the altitude. It is classified as a geostationary orbit (approximately 36,000 km). The low orbit satellite and the geostationary satellite will be described in more detail as follows.

As described above, the low-orbit satellite is a satellite having a satellite altitude of 300 to 1500 km, but currently, the altitude of the low-orbit satellite is widely used within about 600 km. These low-orbit satellites are mainly used for the construction of a satellite mobile communication network, and because they are arranged closer to the ground surface than other satellites such as mid-orbit, high-orbit, and geostationary, they can acquire high-resolution images. It can be used excellently. These low-orbit satellites are located in the atmosphere and acquire images while moving with the trajectory and speed determined according to the user's needs or purposes. For example, the speed of the low-orbit satellites is about 7 km/s, which is much faster than the Earth's rotational speed. Can be.

Geostationary satellites refer to satellites that are observed to be relatively stationary on Earth by moving at the same speed as the Earth's rotational speed. These geostationary satellites have a satellite altitude of 36,000 km. Geostationary satellites are also very useful as communication satellites, but they can also be used to acquire visual information, like low-orbit satellites, because they can observe any point on the earth at all times.

In the case of low-orbit satellites, the satellite altitude is about 600 km as described above, and in the case of a geostationary satellite, the satellite altitude is about 36,000 km as described above. Therefore, as a specific example, if a camera having the capability to acquire an image of 1m resolution when mounted on a low-orbit satellite is mounted on a geostationary-satellite satellite, an image of 60m resolution will be obtained. As another specific example of the same principle, when a camera having a 0.3m resolution performance in a low orbit is used in a still orbit, an 18m resolution image can be obtained.

As described above, low-orbit satellites have the advantage of being able to provide a much higher resolution image compared to geostationary-satellite satellites, but have the disadvantage that the time to see the region of interest is short because it moves at a speed faster than the rotation speed of the earth. On the other hand, a geostationary orbit satellite has the advantage of being able to observe the region of interest at all times compared to a low orbit satellite, but it has the disadvantage that there is a limit to increase the image resolution because the altitude is too high. Therefore, in order to increase the utilization of satellite images, it is necessary to simultaneously secure high resolution, which is an advantage of low-orbit satellites, and always observation, which is an advantage of geostationary satellites.

On the other hand, as the development of IT technology and the use of mobile phones and surveillance cameras have been generalized in recent years, image processing technology is rapidly developing, and of course, this image processing technology is also used to analyze satellite images. As an example, Korean Patent Registration No. 1643886 ("Large-capacity high-resolution satellite image analysis device and method", 2016.07.25) discloses a technique for analyzing a large-capacity high-resolution image more efficiently.

One of image processing technologies, super resolution (SR), is a technique for obtaining high resolution images from low resolution images. 1 is a view for explaining the principle of the SR technique, the basic principle of the SR technique is to capture a stationary object while moving the camera or to shoot a moving object with a fixed camera to obtain multiple images of the same observation object, By processing this, a high-resolution image is obtained. More specifically, FIG. 1(a) shows a moving object with a fixed camera, FIG. 1(b) shows a still object while moving the camera, and FIG. 1(c) shows a video camera. Each image is displayed. One of the images containing the object of interest is taken as a reference image (in the example of FIG. 1, the image taken with FIG. 1(a) is a reference image), and the reference as shown in FIG. 1(d) Compared to images and other images, the subpixel shift is measured, and based on this, the missing parts of each other's images are compensated, and finally a high resolution image can be obtained. This SR technique is a technique that is currently applied to commercialized products, and thus a detailed description is omitted.

By applying this SR technique, as described above, it can be expected that a high-resolution always-observed image can be obtained by combining a high-resolution image obtained from a low-orbit satellite and a permanent observation image obtained from a geostationary-satellite satellite. However, in the case of a low-orbit satellite image, there is a limitation in applying the SR technique because the observation surface of the observation object varies depending on the viewpoint at which the satellite views the observation object. Especially, when the observation object is a stereoscopic object, the viewing surface of the object changes. The image is distorted. As described above, there is a problem in that accurate high-resolution images cannot be obtained simply by applying the existing SR technique.

1. Korea Patent Registration No. 1643886 ("Large-capacity high-resolution satellite image analysis device and method", 2016.07.25)

1. "Spatial Resolution Improvement Using Over Sampling and High Agile Maneuver in Remote Sensing Satellite" (KSAS, 2007)

Therefore, the present invention was devised to solve the problems of the prior art as described above, and the object of the present invention is to ensure that both the high-resolution advantages of low-orbit satellite images and the constant observation advantages of geostationary satellite images are stopped. A method and apparatus for acquiring high-resolution images using micro-vibration in geostationary-satellite satellites that acquire multiple multiple images by using micro-vibration in orbiting satellites and process them using SR techniques to ultimately obtain high-resolution images. In providing.

A high-resolution image acquisition method of the present invention for achieving the above object includes: a viewpoint positioning step in which a viewpoint of an image capturing unit provided in a geostationary satellite is directed to a predetermined region of interest; A vibration applying photographing step in which the region of interest is repeatedly photographed a plurality of times by the image capturing unit immediately after the vibration is applied to the geostationary satellite provided by the vibration applying unit provided in the geostationary orbit satellite; A high resolution calculation step in which a plurality of multiple overlapping images of the region of interest obtained by photographing immediately after vibration is applied are processed to produce a high resolution image; It may include.

In this case, the high resolution calculation step may be performed such that a high resolution image is calculated using a super resolution (SR) technique.

In addition, the vibration applying photographing step is made of an array sensor shape in which the image photographing unit is formed in a plane shape in which a plurality of pixels form a column and a row. It can be made to obtain a flat image at a time.

In addition, in the vibration applying photographing step, the vibration applying unit generates a microscopic movement of the camera in the region of interest, so that the camera gaze vector is moved to the sub-pixel level so that multiple photographing can be performed.

In addition, in the vibration applying photographing step, the vibration applying unit may be configured to apply vibration at least once in each of the horizontal and vertical directions of the region of interest.

In addition, the high-resolution image acquisition method, after the high-resolution calculation step, a region of interest change step of changing the region of interest according to a predetermined shooting scenario; Further comprising, after the step of changing the region of interest, the viewpoint positioning step, the vibration applying photographing step, and the high-resolution calculation step may be sequentially performed to generate a high-resolution image of the changed region of interest.

In addition, the high-resolution image acquisition device of the present invention, a high-resolution image acquisition device using the high-resolution image acquisition method as described above, geostationary satellite; An image photographing unit provided on the geostationary satellite to photograph a region of interest; A vibration applying unit that applies vibration to the geostationary orbit satellite; An image analysis unit for collecting and processing a plurality of multiple overlapping images photographed by the image photographing unit; It may include.

In this case, the image analysis unit may calculate a high-resolution image using a super resolution (SR) technique.

In addition, the image capturing unit may be formed in the form of an array sensor in which a plurality of pixels are formed in a flat form in which columns and rows are formed.

Also, the vibration applying unit may apply vibration by applying torque to a geostationary satellite.

According to the present invention, there is a great effect that a high resolution image can be obtained, which secures both the high resolution advantage of the low-orbit satellite image and the constant observation advantage of the geostationary satellite image. In more detail, according to the present invention, it is possible to greatly improve resolution by acquiring multiple overlapping images using microscopic vibrations in a geostationary orbit satellite and processing these images using SR techniques. At this time, since an image obtained from a geostationary satellite basically has always observed, the high-resolution image obtained as described above naturally has always observed, that is, a high-resolution image.

As described above, according to the present invention, there is an effect of acquiring a high-resolution image with a geostationary satellite, and accordingly, an effect that the utilization of the satellite image can be greatly expanded compared to the conventional one can be obtained.

1 is a principle of the SR technique.
2 is an example of multiple imaging in a low-orbit satellite.
3 is an example of an image obtained from a low-orbit satellite.
4 is an example of an image obtained from a geostationary satellite.
5 is an example of camera line-of-sight movement when a line sensor is used in a geostationary satellite.
6 is a state diagram of the geostationary satellite toward the gaze of interest.
7 is a state diagram of the micro-vibration applied to the geostationary orbit satellite.
8 is an example of a photographing area when vibrating in the horizontal and vertical directions, respectively.
Figure 9 is an example of a captured image obtained when the vibration in each of the horizontal and vertical directions.
10 is an example of overlapping overlapping areas of the photographed images of FIG. 9.

Hereinafter, a method and apparatus for obtaining a high resolution image using microscopic vibration in a geostationary satellite according to the present invention having the above-described configuration will be described in detail with reference to the accompanying drawings.

The high-resolution image acquisition method and apparatus of the present invention is basically for improving the resolution of an image obtained from a geostationary satellite, and uses a super resolution (SR) technique. Therefore, in order to understand the present invention, it is desirable to understand in advance the principle of image acquisition in low orbit satellites and geostationary orbit satellites and problems when applying the SR technique thereto. Accordingly, first, the principle of image acquisition in low-orbit satellites and geostationary-satellite satellites and the simple application of difficulties in SR techniques accordingly, will be briefly described, and then the high-resolution image acquisition method and apparatus of the present invention will be described in detail.

Image acquisition principle and SR technique applied to low orbit satellites and geostationary orbit satellites

First, the principle of image acquisition in a low orbit satellite will be described.

As can be seen from its name, low-orbit satellites have lower satellite altitudes than other types of satellites, so it is possible to obtain sufficiently high-resolution images simply by increasing the camera's performance. However, since the low-orbit satellite moves at a speed faster than the rotation speed of the Earth, it is necessary to change the attitude of the satellite in order to acquire a plurality of images at the same point. As described in detail with reference to FIG. 2 when a point where a plurality of images are to be obtained is referred to as a region of interest P as follows. As described above, the low-orbit satellite uses the satellite maneuvering capability to position the low-orbit satellite before reaching the location of interest (P) (position (a) in FIG. 2), and the low-orbit satellite reaches the location of interest (P). The location of interest (position (b) in FIG. 2) and the low-orbit satellite may pass the location of the region of interest (P), and then photograph the region of interest at the location (position (c) in FIG. 2). At this time, as shown in FIG. 2, the attitude of the low-orbit satellite in the position (a) is tilted toward the direction of the satellite, (b) the location of the low-orbit satellite accurately faces the ground, and (c) the position of the low-orbit satellite in the position. Since the posture is inclined toward the opposite direction of satellite progression, the camera viewpoint can be directed to the region of interest P at each position.

In most low-orbit satellites, a time delay integation (TDI) line sensor is mainly used for high-precision observation. The line sensor is a sensor in which a plurality of pixels are formed in a straight line forming columns or rows, and for example, may have a form in which 1000 pixels are arranged in one line. The image captured by the line sensor becomes a 1000x1 pixel linear image, and the region of interest (P) is not a linear shape but a flat shape, and a plurality of these linear images are collected to collect the region of interest (P). You can get a planar image of the entire area. FIG. 3 conceptually illustrates an example in which images obtained from low-orbit satellites are collected to form one large image, that is, an image of the entire area of interest P. At this time, if the time for acquiring each linear image, that is, the sampling time and the satellite movement speed is exactly the same, as shown in Fig. 3(a), the images of the entire area of interest P are connected by attaching the respective linear images. You will be able to get However, if the sampling time is slow, an area that has not yet been photographed will occur as shown in FIG. 3(b), and if the sampling time is fast, an overlapped area will be generated as shown in FIG. 3(c). In particular, when multiple overlapping images are generated as shown in FIG. 3(c), it can be expected that the resolution can be increased by applying the SR technique described above. This attempt has been well disclosed in "Spatial Resolution Improvement Using Over Sampling and High Agile Maneuver in Remote Sensing Satellite" (KSAS, 2007).

However, in the case of low-orbit satellites, as shown in FIG. 2, there is a problem in that the viewpoint of the satellite looking at the observation object is different for each imaging time of each image. That is, the observation surface of the observation object photographed in each image is different. In particular, when the observation object is a stereoscopic object, if the SR technique is simply applied, the image is distorted. That is, it is difficult to obtain an accurate high-resolution accurate image by simply applying the SR technique when photographing a region of interest using a TDI line sensor in a low-orbit satellite.

Meanwhile, the principle of image acquisition in geostationary satellites is as follows.

In general, geostationary satellites are used to observe a relatively large area compared to low orbit satellites. Specifically, first, the camera view of the geostationary orbit satellite is directed to a certain region of interest, and a photograph is exposed for a certain period of time to obtain a flat image of the entire area of the region of interest. When the image acquisition is completed, the camera view is changed to another region of interest, and then the above-described process is repeated. Since most geostationary images have a main purpose of observing a large area, they are often made to change a region of interest according to a predetermined shooting scenario. FIG. 4 conceptually illustrates an example of acquiring an image from a geostationary orbit satellite while changing a region of interest according to a predetermined shooting scenario. On the other hand, in the case of the geostationary satellite, the distance between the satellite and the observation point is longer than that of the low orbit satellite, so the exposure time must be long to obtain a sufficient signal. It is obvious that the shooting scenario for changing the region of interest should be made so that one region of interest can be photographed after a sufficient time has elapsed after taking this point into consideration.

When the TDI line sensor described above is provided in the geostationary orbit satellite, in order to obtain a flat image of the entire area of a certain region of interest, a plurality of linear images are obtained by using the satellite maneuvering capability in the low orbit satellite, and the flat image is obtained by combining them. Similar to that obtained, in a geostationary orbit satellite, a plurality of linear images may be obtained while changing the line-of-sight vector, and a flat image may be obtained by combining them. 5 shows an example of a camera gaze vector motion for applying the SR technique in a geostationary orbit satellite. In order to apply the SR technique to a geostationary orbit satellite, multiple overlapping images are required, that is, images obtained by photographing the same region of interest (P) in different ways. As in the previous description, if a horizontal image is obtained while moving in the vertical direction and a flat image is obtained, then the gaze vector is rotated 90 degrees, and then the gaze vector is moved horizontally from left to right to move the region of interest ( P) to obtain vertical straight images. Then, by combining them and obtaining another plane image for the same region of interest (P), two overlapping images can be obtained, and it is expected that the resolution can be increased by using the SR technique using these two overlapping images. Can be.

However, in the process of obtaining two overlapping images, a process of rotating the gaze vector by 90 degrees is essential. In order to rotate the gaze vector, the satellite must be rotated, and a geostationary orbit satellite has a considerable volume and weight, which causes a considerable amount of time and power consumption to rotate the satellite.

High-resolution image acquisition method and apparatus of the present invention

As described above, the ultimate object of the present invention is to obtain multiple high-definition images smoothly from a geostationary satellite, and to obtain high-resolution images by applying SR techniques to the multi-overlapping images thus obtained. At this time, as described above, when attempting to obtain a plurality of multiple overlapping images to apply the SR technique, in the case of low-orbit satellites, the observation time of each multiple overlapping images is different, and thus the image is distorted. Since multiple overlapping images are obtained by moving faster than the rotation speed of the Earth, it is impossible to apply the method of low-orbit satellites to geostationary orbit satellites. Also, in the case of a geostationary satellite using a TDI line sensor, the process of rotating the satellite by 90 degrees is essential to obtain multiple overlapping images, and thus, the time and power consumption is a significant problem. In the present invention, in order to solve all of these problems, a geostationary satellite using an array sensor is applied to apply microscopic vibration when photographing a region of interest.

First, the configuration of the high-resolution image acquisition device of the present invention will be described. The high-resolution image acquisition device of the present invention includes a geostationary satellite, an image capturing unit, a vibration applying unit, and an image analyzing unit. The parts will be described in more detail as follows.

The image capturing unit is provided in the geostationary satellite and serves to photograph a region of interest. In this case, it is preferable that the image photographing unit is formed in the form of an array sensor. The array sensor is a sensor formed in a planar form in which a plurality of pixels form columns and rows, for example, a form in which 1000x1000 pixels are arranged in columns and rows. Can have The image captured by the array sensor becomes a 1000x1000-pixel planar image. That is, by making the image capturing unit in the form of an array sensor, it is possible to acquire a flat image of the entire area of the region of interest at a time in a single shot.

The vibration applying unit serves to apply vibration to the geostationary satellite. The specific configuration of the vibration applying unit applies micro-vibration to the geostationary satellite so that the gaze vector of the camera moves at the sub-pixel level during shooting. As one example of such a micro-vibration generating device, the vibration applying unit may be configured to apply vibration by applying torque to the geostationary orbit satellite through a rotational speed change. For example, in the case of a reaction wheel, a change in angular velocity of the wheel is generated to generate torque.

The image analysis unit serves to collect and process a plurality of multiple overlapping images photographed by the image photographing unit. At this time, (as will be described in more detail later), since the plurality of multiple overlapping images obtained in the present invention have little difference in observation time between each other and the same object (region of interest) is not the same, the SR technique is applied. It has very optimal conditions. Therefore, in the present invention, the image analysis unit can produce a high-resolution image very smoothly using a super resolution (SR) technique, regardless of various problems in the conventional low-orbit or geostationary satellite described above.

Now, the high-resolution image acquisition method of the present invention will be described in detail. The high-resolution image acquisition method of the present invention may include a viewpoint positioning step, a vibration applying photographing step, and a high-resolution calculation step, and may further include a region of interest change step. Each step will be described in more detail as follows.

In the viewpoint positioning step, the viewpoint of the image capturing unit provided in the geostationary satellite is directed to a predetermined region of interest (P). 6 shows a state diagram of a geostationary orbit satellite facing the gaze toward the region of interest (P). As described above, when the image photographing unit is formed in the form of an array sensor in which a plurality of arrays of pixels are formed in a planar form of columns and rows, even when only a single shot is taken in the state of FIG. 6, a planar image of the entire area of the region of interest P It can be obtained at once.

In the vibration applying photographing step, the region of interest P is repeatedly photographed multiple times by the image photographing unit immediately after vibration is applied to the geostationary satellite by the vibration applying unit provided in the geostationary satellite. 7 shows a state diagram in which micro vibration is applied to a geostationary orbit satellite. In FIG. 7, the size of the vibration is exaggerated to make it easier to visually confirm that the vibration is being applied, but in reality, the vibration applied by the vibration applying unit is a very small micro-vibration, and the images taken by the imaging unit are almost identical to each other. It becomes multiple overlapping images (that is, only a slight difference), and thus is an optimal condition for applying the SR technique as described above. On the other hand, if the vibration applied by the vibration applying unit is excessively large, errors such as distortion may occur in the image processing process, so the size of the vibration applied by the vibration applying unit is desirably small. More specifically, in the vibration applying photographing step, the vibration applying unit generates a microscopic movement of the camera in the region of interest, so that the camera gaze vector moves to a sub-pixel level so that multiple photographing can be performed. In addition, in order to obtain multiple types of multiple overlapping images, the vibration applying photographing step may include at least one or more times in each of the horizontal and vertical directions of the region of interest P, as shown in FIG. 7. It is preferably made to apply vibration.

The operation of applying the vibration in the vibration applying unit will be described in more detail as follows. The vibration applying unit that generates micro-vibration must be mounted in a direction that is not the same to generate camera movement. Therefore, as an example, when the camera gaze vector direction is a Z-axis, two or more vibration applying parts may be mounted to generate torque in an X-Y plane perpendicular to the Z-axis. When there are multiple vibration applying parts, it is possible to prepare for a failure and variously adjust the camera movement. On the other hand, according to the laws of physics, it is well known that acceleration or angular acceleration occurs when a force or torque is applied to an object, and if the force or torque is not applied, the object moves at a constant or constant velocity. Using this principle, in order to move the camera in a geostationary orbit satellite, the vibration applying unit generates a micro vibration pulse in one direction for a very short moment to change the satellite attitude by the generated torque. When the vibration applying unit does not work, the satellite is subjected to a constant-speed motion, and imaging is performed several times in a constant-speed section for the region of interest. The simplest form of the vibration applied by the vibration applying unit may be a form in which micro-vibrations are sequentially generated in the left, right, up and down directions, imaging is performed after the micro-vibrations are generated, and then the micro-vibrations are generated.

In the high-resolution calculation step, a plurality of multiple overlapping images of the region of interest (P) obtained by photographing immediately after vibration is applied are collected and processed to produce a high-resolution image. As described above, since multiple overlapping images obtained while applying micro-vibrations have optimal conditions for applying the SR technique, it is preferable that the high-resolution calculation step is performed such that the high-resolution images are calculated using the Super Resolution (SR) technique. Do.

The multiple overlapping images obtained in the high resolution calculation step and the SR technique processing using the same will be described in more detail as follows.

8 shows an example of a photographing area when vibrating in the horizontal and vertical directions, respectively. When the vibration applying unit vibrates the geostationary orbit satellite in the horizontal direction (X-axis direction), as shown in FIG. , X5. X3 denotes a state in which the photographing area coincides with the region of interest P, X1 and X2 denote a state in which the photographing region is deflected to the left by the vibration to the left, and X4 and X5 are interested in the photographing region by vibration Each of the states deflected to the right of the area P is shown. Similarly, when the vibration applying unit vibrates the geostationary satellite in the vertical direction (Y-axis direction), in the image capturing unit, Y1,… as shown in FIG. 8(b). , You can get images like Y5. Y3 denotes a state in which the photographing area coincides with the region of interest P, Y1 and Y2 denote a state in which the photographing region is deflected downward from the region of interest P by vibration, and Y4 and Y5 are interested in the photographing region by vibration Each of the states deflected above the area P is shown.

FIG. 9 is an example of a captured image obtained in each of the horizontal and vertical vibrations, and shows the X1 to X5 and Y1 to Y5 separately overlapped in FIG. 8. Specifically, FIG. 9(a) shows photographed images obtained during horizontal vibration, and FIG. 9(b) shows photographed images acquired during vertical vibration. 9(a) When looking at the X1 photographed image, the geostationary satellite is the image photographed in the state most deflected to the left by vibration. In the X1 photographed image, the region on the left is more photographed than the region of interest (this region is " At the same time, the part of the right side of the region of interest cannot be photographed (referred to as the "shot lost"). FIG. 9(a) The X2 photographed image is an image photographed in a state in which the geostationary satellite is less deflected to the left than in the X1 photographed image. In the X2 photographed image, the size of the surplus and lost images is smaller than that of the X1 photographed image. It does exist. 9(a) The X3 photographed image is an image photographed in a state where the photographing area coincides with the region of interest P. In the X3 photographed image, surplus photographing and photographing loss are not generated. 9(a) In the X4 and X5 photographed images, the directions differ only to the right, and similarly, there are surplus photographs and lost photographs in the X2 and X1 photographs, respectively. 9(b), as shown in FIG. 9(b), Y3, Y2, and Y4 are not generated only in the Y3 photographed image (the image photographed while the photographing area coincides with the region of interest P). , In Y5 photographed images, it is well shown that there are surplus photographs and missing photographs on the upper or lower side.

10 shows an example of overlapping overlapping areas of the photographed images of FIG. 9. In FIG. 10, the part indicated by the bold border line is the region of interest P, and the part indicated by the light square area is the photographed images of FIGS. 8 and 9. Each photographed image was marked so as to be dark as the overlapping part overlapped. Specifically, for example, the region indicated by S1 in FIG. 10 is X1,. , X5 and Y1,… , Y5, a total of 10 photographed images overlap, and the area indicated by S2 is X1,… , X5 and Y2,… , Y5, a total of 9 shooting images will all overlap. Overall, as can be seen in FIG. 10, it can be seen that fewer images are superimposed toward the edge of the region of interest P, and more images are superimposed toward the center of the region of interest P. have.

As can be seen from the description of the principle of the SR technique with reference to FIG. 1, the SR technique can increase the resolution as the number of overlapping images for one observation object increases. As is well shown in FIG. 10, when 5 shots are obtained by vibration in the horizontal direction and 5 shots by the vibration in the vertical direction, 10 multiple overlapping images can be obtained in the central portion of the region of interest P, that is, the S1 region. , It is confirmed that the resolution of the S1 region can be significantly improved by processing the SR technique.

In the example of FIGS. 8 to 10, the magnitude of vibration is exaggerated for convenience in order to visually recognize the surplus and minutes lost due to vibration, and in addition, per horizontal or vertical vibration for easy understanding One maximum amplitude position (X1/Y1)-One intermediate amplitude position (X2/Y2)-Match position (X3/Y3)-The other intermediate amplitude position (X4/Y4)-The other maximum amplitude position (X5/Y5) It was explained that it was only located. However, when the technology of the present invention is actually applied to a geostationary orbit satellite, the amplitude of the micro-vibration will be formed much smaller than the horizontal or vertical length of the region of interest (P). It can be easily inferred that the area of interest P will be formed to be almost the same as the entire area. In addition, sampling can be performed more than 5 times in one oscillation cycle, and in this case, it can be expected that the number of multiple overlapping images obtained will be much larger. That is, high-resolution images can be obtained by using the SR technique in most areas of interest (P).

As described above, according to the present invention, by applying micro-vibration to a geostationary orbit satellite to obtain multiple overlapping images of a region of interest and processing them using the SR technique, a much higher resolution image can be obtained than in the prior art. At this time, the image obtained from the geostationary orbit satellite basically has always observed, so the image thus obtained becomes a high-resolution image. That is, according to the present invention, it is possible to obtain an image having both the high-resolution advantages of the aforementioned low-orbit satellite and the constant observation advantages of the geostationary satellite.

Meanwhile, as described above, the high-resolution image acquisition method of the present invention may further include a step of changing a region of interest. The region of interest change step is performed after the high resolution calculation step, and in the region of interest change step, the region of interest is changed according to a preset shooting scenario. Of course, after the step of changing the region of interest as described above, the viewpoint positioning step, the vibration applying photographing step, and the high-resolution calculation step may be sequentially performed, so that a high-resolution image may be calculated for the changed region of interest.

The present invention is not limited to the above-described embodiments, and of course, the scope of application is diverse, and anyone who has ordinary knowledge in the field to which the present invention pertains without departing from the gist of the present invention as claimed in the claims. Of course, various modifications are possible.

P: Region of interest
X1~X5: Recorded image obtained when vibrating horizontally
Y1~Y5: Recorded image obtained during vertical vibration
S1, S2: overlapping area

Claims (10)

A viewpoint positioning step in which the viewpoint of the image capturing unit provided in the geostationary satellite is directed to a predetermined region of interest;
A vibration applying photographing step in which the region of interest is repeatedly photographed a plurality of times by the image capturing unit immediately after vibration is applied to the geostationary satellite provided by the vibration applying unit provided in the geostationary orbit satellite;
A high-resolution calculation step in which a plurality of multiple overlapping images of the region of interest obtained by photographing immediately after vibration is applied are processed to produce a high-resolution image;
Including,
The vibration applying photographing step,
The vibration applying unit applies the vibration using the reaction wheel angular velocity change to the geostationary orbit satellite to apply vibrations at least once in each of the horizontal and vertical directions of the region of interest to actively move the microscopic movement of the camera in the region of interest. A high-resolution image acquisition method characterized in that the camera's gaze vector is moved to a sub-pixel level so that multiple shots are performed.
According to claim 1, The high-resolution calculation step,
High-resolution image acquisition method characterized in that it is made to produce a high-resolution image using the SR (Super Resolution) technique.
The method of claim 1, wherein the vibration applying photographing step,
The imaging unit is formed in the form of an array sensor in which a plurality of pixels are formed in a flat form in which columns and rows are formed.
A high-resolution image acquisition method, characterized in that the flat image of the entire area of interest is acquired at a time when a single shot is taken using the image capturing unit.
delete delete The method of claim 1, wherein the high-resolution image acquisition method,
After the high-resolution calculation step,
A region of interest changing step in which the region of interest is changed according to a preset shooting scenario;
Further comprising,
After the step of changing the region of interest, the viewpoint positioning step, the vibration applying photographing step, and the high-resolution calculation step are sequentially performed, so that a high-resolution image for the changed region of interest is calculated.
In the high-resolution image acquisition device using the high-resolution image acquisition method according to claim 1,
Geostationary satellite;
An image photographing unit provided on the geostationary satellite to photograph a region of interest;
A vibration applying unit that actively applies vibration by applying torque using the reaction wheel angular velocity change to the geostationary orbit satellite;
An image analysis unit for collecting and processing a plurality of multiple overlapping images photographed by the image photographing unit;
High-resolution image acquisition device comprising a.
The method of claim 7, wherein the image analysis unit,
High-resolution image acquisition device, characterized in that for calculating the high-resolution image using the SR (Super Resolution) technique.
The method of claim 7, wherein the imaging unit,
A high-resolution image acquisition device, characterized in that the plurality of pixels is formed in the form of an array sensor that is formed in a planar form of columns and rows.
delete

Images