JPS61193268A - Satellite image correction system - Google Patents

Abstract

PURPOSE:To simplify and enhance accuracy of correction in processing of earth station terminals by effecting distortion correction coefficient in high accuracy at central earth stations and by utilizing such coefft. in distortion correction, transmitting same to the various terminal earth stations. CONSTITUTION:Observation image by the earth observation satellite 1 are transmitted to the earth as digital data in radio waves and the radio waves are received by the central receiving station 2 and one or more terminal receiving station 3. Based on the received images, the processor 6 shall create distortion correction coefft. and the distortion correction coefft. shall be transmitted to the various terminal receiving stations 3 via the transmitter 7 and antenna 8. At the terminal receiving stations, the coefft. are received via the antenna 9 and receiver 10 which are stored in the distortion correction coefft buffer 11. At the point when the coefft. are obtained, distortion correction processing of the observed images stored in the buffer 15 is already possible.

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a satellite image processing method, and in particular, a method for easily and easily eliminating shape distortion contained in images received from a satellite by a receiving station mounted on a ship or on the ground. The present invention also relates to a method suitable for highly accurate correction.

[Background of the invention]

There are attempts to estimate sea surface temperature distribution, ocean current conditions, etc. from satellite observation images such as the meteorological satellite N0AA and use them for fishing purposes. In order to use satellite observation images for this purpose,
Correct shape distortions in observed images caused by fluctuations in the shape and rotation of the earth, fluctuations in the orbit and attitude of satellites, fluctuations in sensor scanning speed, etc., so that each point in the image corresponds to a point on the earth's surface with high precision. There is a need. Two methods have been tried in the past to perform distortion correction processing for this purpose.

The first is "Development of Fishery Remote Sensing Image Processing System" by Honma et al. in the April 1958 issue of Hitachi Hyoron.
This method is exemplified in the paper titled. The central satellite ground station performs processing on the image received from the satellite to remove shape and shading distortions included in the image to obtain a corrected image. An estimated value of the seawater temperature at the corresponding sea surface point is determined using a regression equation prepared in advance from the density of each point in the corrected image obtained from the infrared sensor data. From now on, more isotherm maps of the sea surface, or ocean state maps showing patterns of ocean currents, will be created. These drawings, along with fishing location information obtained from reports from the fishing boats, are sent to the fishing boats by facsimile. The fishing boat receives this map, and the crew uses this map to locate fishing spots. To correct the distortion of the received image in this system, we create a geometric model that associates each pixel of the observed image with a ground point based on the satellite's orbit and attitude data and the earth's shape model, and use this to create a geometric model that associates each pixel of the observed image with a ground point. A corrected image is obtained by creating a mapping between points on the uncorrected observed image, that is, a distortion correction coefficient, and using this to determine the image intensity on the corresponding observed image for each pixel of the corrected image. At this time, errors included in the satellite's orbit and attitude data cause a decrease in distortion correction accuracy. On the other hand, this system can receive images of a wide range of areas including land, so it is possible to receive images of a wide range of areas including land.
The orbit and attitude of the satellite are estimated with high accuracy using the position information in the image of the trol point (referred to as trol point), and processing is usually performed to improve the accuracy of distortion correction.

The second one is Aerospace Research Notes, Sky and Sea, No. 6-jj.

Kitano in 1983, “APT that can be used for fisheries.”
This method is exemplified in the paper titled "Image Receiving Device". This is a system installed on ships, etc.

The image data sent from the satellite is directly received by the antenna,
The image obtained after distortion correction processing and temperature conversion processing is displayed on a display device.

The content of the processing is the same as that of the first conventional technique, but in order to simplify the processing, it is common to simplify the geometric model for distortion correction. Furthermore, high-precision correction using GCP is not normally performed.

Now, the purposes of using satellite images on fishing boats, etc. are (1) to obtain information quickly, (2) to have high accuracy of information, and (
3) It is desirable to have a large amount of information.If we look at the conventional technology from these points, there are the following problems. In the first conventional technology, requirement (2) above can be achieved through precise correction processing using GCP, but it takes a long time to process data from a wide range of areas at once through centralized processing at the central ground station. ) is not satisfied, and the processing results are sea surface temperature maps and sea state maps.

It is sent to the fishing boat by facsimile as a drawing with lines, letters, and symbols such as a fishing map. This is to reduce the load on wireless transmission, but part of the information in the corrected image is lost, and (3) is not satisfied.In the second prior art, images from satellites are directly received, so ( 1) and (3) are satisfied. However, do not perform high-precision correction using GCP&1'
In F', (2) is not satisfied.GCP processing is difficult because there is often no GCP when a fishing boat receives an image of a small area of interest centered on its own position. Therefore, receiving images over a wide range of areas including land has the problem of increasing equipment requirements, which is undesirable for systems mounted on fishing boats. That was difficult.

[Object of the invention]

□An object of the present invention is to solve the above-mentioned problems of the prior art by easily and highly accurately detecting shape distortions contained in images received from a satellite at a receiving station mounted on a ship or installed on the ground. The purpose is to provide a means for correction.
□ [Summary of the Invention] In order to achieve the above object, the present invention provides a plurality of terminal receiving stations mounted on a ship or installed on the ground.

One central receiving station is installed, and the central receiving station performs precise processing to create a distortion correction coefficient and transmits it to the single receiving station, and each terminal receiving station uses the distortion correction coefficient to perform distortion correction processing on the received image. It is distinctive in what it does.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to FIGS. 1 to 4. FIG. 1 is a diagram showing the overall structure of this invention.
Observation images by earth observation satellites 1 such as N0AA are sent to the ground as digital data by radio waves. The central receiving station 2 and one or more terminal receiving stations 3 receive this radio wave. The central receiving station 2 sends the received radio waves to the antenna 4. It is received via the receiver 5. Processing device from this received image! 6 creates distortion correction coefficients through processing to be described later. This distortion correction coefficient is the transmitter 7. It is sent to each terminal receiving station 3 via antenna '8.

Here, the distortion supplement coefficient and its creation method will be explained. In FIG. 2, image 20 represents both uncorrected observed images, and image 21
Cue the respective coordinate systems indicating the desired corrected image.
p) pcxe y), now let the coordinates of any point 22 in the corrected image be (xey), then the mapping 24 showing the coordinates (Qt p) of the corresponding point 23 in both uncorrected images is a distortion model. call. Generally speaking, the distortion model can be expressed as follows. Here f...! , indicates a function. This function form calculates the satellite's orbit, attitude data, and
It is determined by performing three-dimensional geometric calculations using a sensor model, a shape model of the earth, etc.

For details, please refer to the above-mentioned two documents, as well as the April 1958 issue of ``Development of Fishery Remote Sensing System''.Furthermore, the orbits of regular satellites.

Attitude data contains errors, so in order to reduce the errors in the distortion model, we use GCP whose position on the ground surface in the observed image is accurately known to estimate the orbit and attitude more accurately. However, a method is used to create a distortion model. Further, as the function of equation (1), the corrected image is divided into a plurality of partial regions (blocks) and then polynomial approximation is performed for each block. For example, the following dark linear function is used.

At this time, the coefficients aOT...tb3 of this polynomial are called distortion correction coefficients.

On the other hand, the terminal receiving station 3 receives the image data sent from the satellite 1 through the antenna 12. The image is received via the receiver 13 and stored in the received image buffer 15. In the case of the N0AA satellite, the dimensions of images that can be received are approximately 4000 km on the earth's surface in the orbital direction (north-south) and approximately 2 km in the orthogonal direction to the orbit (east-west).
All of this may be stored in the buffer 15, which covers a wide range of 1,000 km, but since the range required for each ship is usually small, the capacity of the buffer 15 can be reduced in a linear manner. Satellite orbit information 18 is generally obtained by means such as telex. Receiving area calculation bag [19
determines the area of the observation image to be received and stored in the buffer 15 from this orbit information 18 and the region of interest designation data 14 given from the outside, and transmits it to the receiver 13. The principle of processing of the device 19 will be explained with reference to FIG. Orbit information 1
8, the trajectory 30 of the point just below the orbit of the receiving satellite and the boundaries 31 and 32 of the observation range of the scanning sensor are determined. It is also assumed that the coordinates (X o e 7 o ) of the center point 33 corresponding to the region 34 in the figure and the dimension J x +Ay of the region are given as the region of interest specification data 14 . At this time, from the trajectory 30 of the trajectory, time t corresponding to the scanning line 35 at which reception should start and time t1 corresponding to the scanning line 36 at which reception should end are determined. In addition, the time Jtt at which data storage in the buffer should start and the time Δt at which storage should end on each scanning line are determined. jy ja found in this way
y Atxv Atz is sent to the receiver 13 and used for selective reception by the receiver.

Now, the distortion correction coefficients created by the central receiving station 2 using the method described above are sent to each terminal receiving station 3. At the terminal receiving station, antenna 9. This is received via the receiver 1o and stored in the distortion correction coefficient buffer 11. Once this coefficient is obtained, distortion correction processing can be performed on the observed image already stored in the buffer 15.

FIG. 4 is a block diagram of the processing device 16 that performs distortion correction processing. The device 16 consists of a sampling device 40° corrected image variation 412 and an enhancement processing device 42. The resampling device 40 performs the following processing on each point of the corrected image defined by the region of interest specification data 14, that is, uses the coefficients stored in the buffer 11 to extract the corresponding uncorrected image according to equations (1) and (2). Find the coordinates (Q, p) of the image. At this time, (Q, p) generally does not take an integer value corresponding to each point of the observed image, and the observed image intensity corresponding to this coordinate must be determined by interpolation. Now, when the coordinates of each point in the observed image around (a, p) are (it j), the intensity at (Q, P) is determined by the following interpolation formula.

Here, W is a known weighting function. For (i, j), it is sufficient to take 4 to 16 points around (Ωe p).The corrected image obtained as a result of performing this interpolation process, that is, resampling process on each point of the corrected image, is stored in the buffer 4.
The 6-enhancement processing device 42 outputted to 1 performs shading conversion, pseudo color enhancement, etc. when displaying the corrected image on the display device 17.

As described above, according to this embodiment, since the distortion correction coefficients created with high accuracy at the center receiving station are used, there is an effect that the correction accuracy at the terminal receiving station at any location can be improved.

〔Effect of the invention〕

According to the present invention, a distortion correction coefficient is applied with high precision at the central ground station and can be sent to each terminal ground station to be used for distortion correction, which has the effect of simplifying processing at the terminal ground station and increasing correction accuracy. There is.

[Brief explanation of drawings]

Fig. 1 is an overall configuration diagram of the satellite image correction method according to the present invention, Fig. 2 is a diagram for explaining the distortion correction coefficient, Fig. 3 is a diagram for explaining the principle of reception area calculation, and Fig. 4 FIG. 1 is a configuration diagram of a processing device in a terminal ground station according to the present invention. 1...Earth observation satellite, 2...Central receiving station, 3...
Terminal receiving station, 20...Uncorrected image, 21...Corrected image No. 6 is shown in Figure Z.

Claims (1)

[Claims] In an earth observation system consisting of an earth observation satellite, one central receiving station, and one or more terminal receiving stations, the distortion correction coefficients created by the central receiving station are transmitted to the terminal receiving station, and the coefficients received by the terminal receiving station are A satellite image correction method characterized by performing distortion correction processing on a received image using.