JP2854578B2 - Image processing method and apparatus for synthetic aperture radar - Google Patents

Description

The present invention relates to a synthetic aperture radar (SAR) used for remote sensing of an arbitrary target surface, and is particularly applied to image processing of the radar data. And an image processing method and apparatus for a synthetic aperture radar.

(Prior Art) As is well known, in such a synthetic aperture radar, radar data collected through a synthetic aperture antenna mounted on a flying object such as an artificial satellite or an aircraft is converted into an azimuth direction which is a traveling direction of the synthetic aperture antenna. And the data in the range direction that is perpendicular to the traveling direction of the radio wave transmitted and received via the antenna, that is, the azimuth direction (assuming the mapping mode by side-looking). Each data in the direction and the range direction is image-processed as required to obtain a two-dimensional radar image of the target plane.

Here, the processing principle of the decomposition of the collected radar data will be described.

First, the processing related to the range direction will be described.

The decomposition of the signal in the range direction is performed using the time difference between the transmission pulse transmission and the reflection from the target object to the reception after transmission, as in the case of a normal radar as shown in FIG.

In FIG. 4, radio waves are transmitted from the radar (antenna) to the R
Td is the time it takes to be reflected and received from an object at a distance of It is. Here, c indicates the speed of light.

In addition, the ground resolution in the range direction is set as δR and When expressed using ΔTd (see FIG. 4 (b)),
The transmission pulse width is represented by ΔTd ≧ τ (3) where τ is τ. Therefore, the bandwidth of the system is Must. By the way, about 1m resolution is required
A bandwidth of 150 MHz, that is, a pulse width of about 6 ns is required.

In order to obtain such high resolution, it is necessary to widen the range of the system. However, if the pulse width is reduced, transmission power decreases and detection sensitivity (S / N) also deteriorates. To prevent this, increasing the peak power of the transmission pulse is technically and economically problematic. Therefore, a method is used in which a high-band signal obtained by modulating (spreading) a transmission pulse with a specific signal is transmitted, and after reception, demodulation (compression) is performed in a data processing stage to obtain high resolution.

In synthetic aperture radar, it is common to use chirp (Charp Rada
A linear FM (frequency modulation) called r) is used. As shown in FIG. 5 (a), the frequency is linearly modulated (frequency shift Δf) in the transmission pulse width, T, within the width T (characteristic of FIG. 5 (b)). A filter (Matched Fillter) with a linear frequency-delay time characteristic (the relationship between frequency and time is opposite to that on the transmitting side) as shown in FIG.
Thus, an output having an envelope waveform (a large output with a narrow pulse width) as shown in FIG. Due to this pulse compression action, the amplitude becomes It can be seen that the pulse width is 1 / (T · Δf) times, and the distance resolution and S / N are improved as Δf is increased.

Next, processing in the azimuth direction will be described.

Although the high resolution in the azimuth direction is a unique feature of the synthetic aperture radar, there are two approaches to this. One is that when the flying object advances from left to right in FIG.
First enters the antenna beam when the flying object comes to the position 1 and exits the antenna beam at 3 via 2. During this time, X is constantly irradiated with the radar antenna beam.
That is, the flying object sequentially feeds and transmits the received signal to each element of the array antenna which is formed by effectively combining the aperture surface of the antenna with the size of L or developed from 1 to 3 on the trajectory of the flying object. It can be thought that it was recorded.
In the case of a synthetic aperture radar, since a radio wave reciprocates, the phase difference between adjacent antenna elements is twice that of a normal array antenna having the same element spacing. Therefore, the array antenna beam width of length L at wavelength λ is And this is θa Becomes If the relationship of LθR (6) and the relationship that the beam width of the real aperture antenna having the diameter D is equal to θλ / D (7) with respect to the radio wave of the wavelength λ is used, the resolution in the azimuth direction (distance resolution) is obtained. ) Gives this as δs Becomes

This is, so to speak, a criterion for explanation that is taken as a measurement target.

Another approach is based on radar (flying objects). In FIG. 6, if attention is paid to one target, X, the scattered radio wave due to X received when the flying object (platform) moves at the velocity, v, shows the Doppler effect due to the relative velocity. For example, in FIG. 7, the distance (R) between the radar and the trajectory of X, the speed of the platform (v),
(Which may be considered as the speed of the target) and the like, if there is no Doppler shift, X at a distance of x from point 0 where there is no Doppler shift
The Doppler frequency. That is, the history of the Doppler frequency when X passes through the radar beam can be calculated. This is a form in which the carrier frequency has undergone FM, as shown in FIG. 8 (a). Thus, by correlating between a series of radar received signals at a distance R and a reference signal of the calculated Doppler shift, it is entered in the received signal in a spread (with a continuously changing Doppler frequency) form. The signal of the target X is extracted, and the image is focused on the position of 0 and emerges. This will be described with reference to FIG. 8. If a signal subjected to FM as shown in FIG. 8A is applied to a matched filter having characteristics as shown in FIG. As a result, this is compressed to one point as shown in FIG. This is called azimuth compression, and actual processing of synthetic aperture radar data is also performed according to this concept.

Incidentally, the transmission wave F in Figure 7 _{(t) = A 0 exp (} iw 0 t) ... (9): Target X (R, x) the received signal of the scattered wave due to the (W ₀ transmit carrier angular frequency) Becomes

(C: value from the radar equation) Doppler shift due to motion The width of the Doppler shift that occurs from the time X enters the beam until it exits is Assuming that the system bandwidth is B, B ≧ 2fD (16) Since the time resolution is approximately B ⁻¹ , the azimuth resolution (distance resolution) δs is Which is the same as the result of Expression (8). That is, the resolution in the azimuth direction has such a property that the smaller the antenna diameter (D), the better. This means that the smaller the antenna, the wider the beam width, the longer the object to be measured will stay in it (or the longer the effective array antenna length L), and create (compress) the image.
This is due to the effect of an increase in the amount of data to be performed.

As described above, the signal can be decomposed in the azimuth direction by the same compression processing as in the range direction, and the received Doppler signal (Doppler frequency fd) in the azimuth direction is shown in FIG. It is clear from the principle of azimuth compression shown in FIG. 8 that the signal has the form shown in FIG. 8. However, in such coherent processing, a specific noise called speckle is usually generated in the reproduced image. Conventionally, a multi-look process, that is, a process of dividing a data group into a plurality of data groups on a frequency axis, obtaining image data for each of the divided data groups, and adding them to obtain one image. This data set for the azimuth direction
By averaging the noise by such addition, the occurrence of the speckle is suppressed.

For example, in the example shown in FIG. 9, the data group in the azimuth direction is divided into two data groups of Is and IIs (in this case, the number of looks is "2"), and these divisions are performed. After image processing is individually performed on each of the data groups thus obtained, these are added to obtain one image to be displayed on the azimuth direction axis. Note that the number of divisions (the number of wrecks) is fixedly set in advance at the stage of designing the radar, and may be set to a larger value such as “4”.

(Problems to be Solved by the Invention) When such multi-look processing is performed, the speckle noise is reduced in proportion to the number of the multi-look processing, so that the image quality of the displayed image is improved. ,
Increasing the number of looks in this way (in other words, employing multi-look processing) also shortens the synthetic aperture length per look, and conversely lowers the resolution.

Here, the conventional synthetic aperture radar performs such multi-look processing only on the data group in the azimuth direction as described above, and performs the image processing.Therefore, when displaying an image, the resolution in the range direction is improved. If the image quality of the two-dimensional surface as a whole is to be maintained to an extent that can be substantially satisfied while maintaining the same, the resolution is greatly different between the azimuth direction and the range direction, resulting in poor visibility. In order to give priority to obtaining images having the same resolution in the azimuth direction and the range direction, the resolution in the range direction has to be sacrificed without any improvement in image quality. Incidentally, the number of looks at the time of the multi-look processing has to be fixedly determined in advance at the time of designing the data in relation to the data group in the range direction.

The present invention has been made in view of such circumstances, and maintains the same resolution in the azimuth direction and the range direction, and also maintains a level that is sufficiently satisfactory in image quality to improve the visibility of a displayed image. It is an object of the present invention to provide an image processing method and apparatus for a synthetic aperture radar which can be greatly enhanced.

[Configuration of the invention]

(Means for Solving the Problems) In order to achieve the above object, the invention according to claim 1 is mounted on a flying object such as an artificial satellite or an aircraft, and the resolution in the azimuth direction which is the traveling direction of the flying object and the azimuth In an image processing method for a synthetic aperture radar in which data collection conditions including multi-look processing are set such that resolutions in a range direction orthogonal to the direction become equal to the resolution in the azimuth direction and the range direction collected by the synthetic aperture radar, The first two-dimensional image having the same resolution in the azimuth direction and the resolution in the range direction with respect to the target object is formed by directly converting the original data under the condition that the resolution of the original data is equal to the image data. Performs the multi-look processing of the same number of looks in both the azimuth direction and the range direction to obtain the target in the azimuth direction. And forming a second two-dimensional image having the same resolution in the range direction as the first two-dimensional image and displaying the first two-dimensional image and the second two-dimensional image on display means so that they can be referred to.

The invention according to claim 2 is mounted on a flying object such as an artificial satellite or an aircraft so that the resolution in the azimuth direction which is the traveling direction of the flying object is equal to the resolution in the range direction orthogonal to the azimuth direction. In an image processing method for a synthetic aperture radar in which data collection conditions including multi-look processing are set, the original data under the condition that the resolution in the azimuth direction and the resolution in the range direction collected by the synthetic aperture radar are equal to direct image data. First image processing means for forming a first two-dimensional image having the same resolution in the azimuth direction and the resolution in the range direction with respect to the target object; Performs multi-look processing with the same number of looks in both directions, so that the azimuth resolution and range resolution of the target are equal. Second image processing means for forming a second two-dimensional image, a first two-dimensional image formed by the first image processing means, and a second two-dimensional image formed by the second image forming means. Display control means for displaying an image and a referenceable image.

The invention according to claim 3 is the invention according to claim 2,
The display control means is configured to control the first two-dimensional image and the second
Is switched and displayed on one display device.

The invention according to claim 4 is the invention according to claim 2,
The display control means is configured to control the first two-dimensional image and the second
Is simultaneously displayed on at least two displays.

(Function) According to the present invention, the original data under the condition that the resolution in the azimuth direction and the resolution in the range direction collected by the synthetic aperture radar are equal to each other, are directly converted into image data, thereby achieving A first two-dimensional image having the same resolution and the same resolution in the range direction is formed, and multi-look processing of the same number of looks is performed on the original data in both the azimuth direction and the range direction to obtain a target image. A second two-dimensional image having the same azimuth resolution and range resolution with respect to the object is formed.

Here, since the first two-dimensional image is obtained by directly converting the original data into image data, the image has a high resolution but a narrow field of view, and the second two-dimensional image has an azimuth Since the multi-look processing of the same number of looks is executed in both the direction and the range direction, the resolution is reduced, but the image has excellent recognizability and good image quality.

Then, the first two-dimensional image and the second two-dimensional image are displayed on the display means so that they can be referred to.

According to such a configuration, it is possible to refer to the first two-dimensional image which has a high resolution but has a narrow field of view and the second two-dimensional image which has a low resolving power but has excellent recognizability and good image quality. Thus, it is possible to easily obtain useful information on the target.

(Example) As shown in FIG. 5 (b), since the transmission pulse of the synthetic aperture radar is chirped, the signal (data group) decomposed in the range direction of the received radio wave is also obtained.
The frequency changes linearly with respect to the time axis t as shown in FIG. That is, this is common to the Doppler frequency characteristic of the signal (data group) decomposed in the azimuth direction shown in FIG. 2 between I _R and II _R as shown in FIG. 10
This means that the data can be divided into two data groups and subjected to the same multi-look processing as described above.

FIG. 1 shows an embodiment of an image processing apparatus for a synthetic aperture radar according to the present invention constructed based on such logic.

In this example, synthetic aperture radar itself, for example, the (2) and (8) (or (17)) in basis, so that its range resolution [delta] _R and azimuth resolution δs is equal, for example, the antenna diameter It is assumed that the design has been made for the transmission waveform (pulse width) and the like.
Therefore, the synthetic aperture radar (SAR) original data supplied to the image processing circuit 10a shown in FIG. 1 by using a magnetic tape or an appropriate communication means as a medium is itself a data group in the range direction and a data group in the azimuth direction. When these data groups are decomposed into data groups, the resolutions in these two directions are equal.

In the image processing circuit 10a shown in FIG. 1, the range FFT unit 11 compresses the target (target plane) information of the SAR original data, which is spread in two dimensions (range direction and azimuth direction), in the range direction. Part. As is well known, this compression is realized by cross-correlating the data group in the range direction with a predetermined reference function. Here, FFT (Fast Fourier Transform) is used as a method for obtaining such cross-correlation.

Also, in the image processing circuit 10a, range multi-look compressing unit 12, the range compressed data group, for example, as shown in FIG. 10, the two groups of data in the example of I _R and II _R he says This is a portion that is divided and inverse FFT is performed on each of the divided data groups. As a result, the data group in the range direction is divided into the first data group I _R
When it comes to be image processed independently by the second data group II _R.

The corner turn unit 13 in the image processing circuit 10a
Is a part for rearranging the range multi-look compressed data group stored in the range direction in the azimuth direction in order to efficiently perform the subsequent processing in the azimuth direction. That is, this processing corresponds to transposition of rows and columns of two-dimensional data.

In the same image processing circuit 10a, the azimuth FFT unit 14
Is a part for further compressing the rearranged range multi-look compressed data group in the azimuth direction. This compression is also realized by cross-correlating these data groups with a predetermined reference function.

Further, in the image processing circuit 10a, the azimuth multi-look compression unit 15 converts the above-mentioned azimuthally compressed data group into, for example, two data groups of Is and IIs as shown in FIG. And performs inverse FFT on each of the divided data groups. As a result,
From the azimuth multi-look compression unit 15, and the divided data units I _R and II _R for the previous range direction, four data groups respectively corresponding to the divided data units I s and II s for this azimuth direction Will be output.

Then, the look addition processing unit 16 in the image processing circuit 10a is a part that takes the absolute values of the four types of data groups subjected to the inverse FFT and adds them incoherently to each other. Through such addition processing, noise included in each data group is also averaged.

As described above, the image processing circuit 10a supplies the S
A circuit that decomposes the AR original data into a data group in the range direction and a data group in the azimuth direction, and performs the above-described multi-look processing on both of the decomposed data groups. Display control circuit
Supplied to 20a.

The display control circuit 20a is a circuit that performs well-known control so that the supplied data group is two-dimensionally displayed on an appropriate display 30a in accordance with the range direction and the azimuth direction.

As described above, according to the embodiment shown in FIG.
Same resolution in range and azimuth directions
Since the SAR original data is used to perform multi-look processing using the same number of looks (the number of looks is “2”) for each of the separated data groups in the range direction and the azimuth direction, the data is displayed on the display 30a. The two-dimensional image also slightly decreases in terms of resolution (in the case of the above example, the resolution is doubled in each of the range direction and the azimuth direction).
However, the image quality is greatly improved, and an image with excellent visibility that maintains the same resolution in the range direction and the azimuth direction is obtained.

In the above example, the number of looks in the multi-look processing, that is, the number of divisions of the data group in the range multi-look compression unit 12 and the azimuth multi-look compression unit 15 is “2”. The settings are optional,
In particular, in an apparatus such as this embodiment, which performs multi-look processing with the same number of looks for both data groups in the range direction and the azimuth direction having the same resolution in the original data, the number of looks may be arbitrarily set. It is also possible to adopt a configuration that can perform the following.

Further, in the embodiment shown in FIG. 1, only the image subjected to the multi-look processing as described above is displayed on an appropriate display, but here the SAR original data itself is displayed in the range direction as described above. And the azimuth direction, the resolution is the same. Therefore, if the multi-look processing is not performed on this SAR original data and image processing is performed directly on it, it is the highest resolution image of the system. A full resolution image can also be displayed on a display as an image having the same resolution in the range direction and the azimuth direction. An embodiment in such a case is shown in FIGS. 2 and 3, respectively.

First, the embodiment apparatus shown in FIG. 2 is capable of selectively displaying one image 30a subjected to the multi-look processing and a full resolution image not subjected to the multi-look processing on one display 30a. It is.

That is, in the apparatus shown in FIG.
Reference numeral 10b denotes a data group that has been range-compressed by the range FFT unit 11 and the range compression unit 17 (both are well known) in addition to the above-described elements constituting the image processing circuit 10a shown in FIG. A corner turn unit 13 'for extracting and rearranging the data in the azimuth direction, an azimuth FFT unit 14' for further compressing the rearranged data group in the azimuth direction and outputting the compressed data group directly to the display control circuit 20b, and an azimuth compression unit The display control circuit 20b includes a multi-look processing data (an output of the look addition processing section 16) supplied from the image processing circuit 10b. Data) and the bypass processing data (output data of the azimuth compression unit 18) are selectively switched and displayed on the display 30a based on a switching signal appropriately applied. .

This makes it possible to switch and display between the multi-look processing image and the full resolution image according to an arbitrary request of the user.

In the embodiment shown in FIG. 1, if the number of divisions (the number of looks) set in each of the range multi-look compression unit 12 and the azimuth multi-look compression unit 15 is "1", the above-described full resolution It is possible to obtain a resolution image, and when switching and displaying two or more types of images, the number of divisions may be changed and the embodiment apparatus shown in FIG. 1 may be used repeatedly. . That is, this makes it possible to use the image properly according to the purpose.

Further, the embodiment apparatus shown in FIG. 3 can simultaneously display an image subjected to the multi-look processing and a full resolution image not subjected to the multi-look processing on the two displays 30a and 30b. Things.

That is, in the apparatus shown in FIG.
10b is a bypass circuit (range compression unit 17, corner turn unit 13 'azimuth FFT unit 1) similar to that shown in FIG.
4 ′ and an azimuth compression unit 18). The display control circuit 20c includes multi-look processing data (output data of the look addition processing unit 16) and bypass processing data (azimuth compression unit) supplied from the image processing circuit 10b. 18 output data)
Are processed separately and simultaneously, and are simultaneously displayed on the first and second two displays 30a and 30b.

This allows simultaneous display of the multi-look processed image and the full resolution image described above.

Further, if the embodiment shown in FIGS. 1 to 3 is further developed, a plurality of image processing circuits having a different number of looks (a circuit having the number of looks "1" for reproducing the full resolution image or a bypass circuit). It is of course possible to prepare a plurality of displays corresponding to this number and display a plurality of types of images reproduced by the plurality of image processing circuits on a plurality of displays simultaneously. is there.

〔The invention's effect〕

As described above, according to the present invention, the azimuth relative to the target object is directly converted into image data under the condition that the resolution in the azimuth direction and the resolution in the range direction collected by the synthetic aperture radar are equal. Forming a first two-dimensional image having the same resolution in the azimuth direction and the resolution in the range direction, and performing the multi-look processing of the same number of looks in both the azimuth direction and the range direction on the original data. And forming a second two-dimensional image having the same resolution in the azimuth direction and the range direction with respect to the target object, and forming the first two-dimensional image and the second two-dimensional image.
Since the two-dimensional image is configured to be displayed on the display means so that it can be referred to, it is possible to refer to the first two-dimensional image having a high resolution but having a narrow field of view and the second two-dimensional image having excellent recognizability and high image quality. , And thereby, useful information regarding the target object can be easily obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of an image processing apparatus according to the present invention. FIGS. 2 and 3 are block diagrams showing another embodiment of the image processing apparatus according to the present invention.
FIG. 4 is a schematic diagram showing the concept of the range direction processing of the synthetic aperture radar, FIG. 5 is a diagram showing the concept of chirp compression, FIG. 6 is a schematic diagram showing the concept of the azimuth direction processing of the synthetic aperture radar, and FIG. FIG. 8 is a schematic diagram showing the relationship between the target position and the Doppler frequency in the aperture radar, FIG. 8 is a diagram showing the concept of azimuth compression, FIG. 9 is a diagram showing the Doppler frequency transition by a point target in the azimuth direction, FIG. FIG. 3 is a diagram showing a frequency transition of a reception pulse in a range direction. 10a, 10b ... Image processing circuit, 11 ... Range FFT unit, 12 ...
Range multi-look compression section, 13, 13 ': corner turn section, 14, 14': azimuth FFT section, 15: azimuth multi-look compression section, 16: look addition processing section, 17: range compression section, 18 azimuth compression section, 20a, 20b, 20c display control circuit, 30a, 30b display.

 ──────────────────────────────────────────────────続 き Continuation of front page (56) References JP-A-61-100679 (JP, A) JP-A-61-100680 (JP, A) JP-A-61-124881 (JP, A) JP-A-61-100 213784 (JP, A)

Claims (4)

(57) [Claims] Claims: 1. A flying object such as an artificial satellite or an aircraft,
An image processing method for a synthetic aperture radar in which data collection conditions including multi-look processing are set such that resolution in an azimuth direction that is the traveling direction of the flying object and resolution in a range direction orthogonal to the azimuth direction are equal, The azimuth resolution and the range resolution with respect to the target are obtained by directly converting the original data under the condition that the resolution in the azimuth direction and the resolution in the range direction collected by the synthetic aperture radar are equal to each other. Equal first
And a multi-look process of the same number of looks in both the azimuth direction and the range direction is performed on the original data, and the resolution in the azimuth direction with respect to the target is improved. A synthetic aperture radar, wherein a second two-dimensional image having the same resolution in the range direction is formed, and the first two-dimensional image and the second two-dimensional image are displayed on display means so as to be referred to. Image processing method. 2. A vehicle mounted on a flying object such as an artificial satellite or an aircraft,
In a synthetic aperture radar image processing apparatus in which data collection conditions including multi-look processing are set so that the resolution in the azimuth direction that is the traveling direction of the flying object and the resolution in the range direction orthogonal to the azimuth direction are equal, The azimuth resolution and the range resolution with respect to the target are obtained by directly converting the original data under the condition that the resolution in the azimuth direction and the resolution in the range direction collected by the synthetic aperture radar are equal to each other. Equal first
A first image processing means for forming a two-dimensional image of the following; performing multi-look processing of the same number of looks in both the azimuth direction and the range direction on the original data; Forming a second two-dimensional image in which the resolution in the azimuth direction and the resolution in the range direction are equal.
Image processing means; display control means for displaying the first two-dimensional image formed by the first image processing means and the second two-dimensional image formed by the second image forming means so as to be referred to; An image processing apparatus for a synthetic aperture radar, comprising: 3. The synthetic aperture according to claim 2, wherein said display control means switches and displays said first two-dimensional image and said second two-dimensional image on a single display. Radar image processing device. 4. The combination according to claim 2, wherein said display control means simultaneously displays said first two-dimensional image and said second two-dimensional image on at least two displays. Aperture radar image processing device.