JPS5815178A - Picture processor of synthetic aperture radar - Google Patents

Abstract

PURPOSE:To easily attain compensation of distortion and reduction of a noise, by using both a data obtained by Fourier conversion of a conjugate function of a transmitting signal at the time of range compression, and a data obtained by Fourier conversion of a time domain reference function at the time of azimuth compression. CONSTITUTION:In a range compression part A, a data obtained by high speed Fourier conversion in an FET circuit 9, as to a conjugate function of a transmitting signal obtained by a time domain reference function generating circuit 8 is multiplied by a data obtained by high speed Fourier conversion in the range direction in an FET circuit 10, as to a synthetic aperture radar (SAR) data, by a multiplier 11. The data obtained in this way is subjected to Fourier reverse conversion in an IFFT circuit 12, and after that, is arranged in the azimuth direction by a corner turning circuit 13. In an azimuth compression part B, the processing is executed by a time domain reference function generating circuit 15, FET circuits 14, 16, a multiplying circuit 17 and an IFFT circuit 18 in the same way.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides digital processing of synthetic aperture radar acquisition data.
l device.

Synthetic Aperture Radar
e Radar (hereinafter referred to as 8AR) is a type of side-looking radar, and is an all-weather high-resolution sensor that is mounted on a mobile platform such as an aircraft or an artificial satellite and takes pictures of the earth's surface using radio waves.

Normally, the distance resolution of a radar improves as the effective pulse width of the transmitted pulse becomes narrower, and the azimuth resolution improves as the beam width of the antenna becomes narrower. Therefore, in order to obtain high resolution,
It requires a narrow pulse, high power transmitter and a large diameter antenna.

In contrast, in SAR, the 8 synthetic aperture technique uses synthetic aperture technology and pulse compression technology, which emits transmitted pulses at regular intervals from a relatively small-diameter antenna as the platform moves, and reduces the amount of reflected waves from the target. By recording amplitude and phase information and processing these information,
Obtains high azimuth resolution equivalent to a large diameter antenna (azimuth compression).

In addition, in pulse compression technology, instead of reducing the peak power of a pulse 4 ['ft], the pulse width is increased to increase the average power, and this long pulse is subjected to appropriate modulation (linear frequency modulation). etc.) and sends a pulse with
By processing the received signal from the target), the total power in the received signal is compressed to one point in time, and high range resolution is obtained (distance compression).

Conventionally, optical processing using a lens system as shown in the KMI diagram for processing received signals has been commonly used. That is, this method applies the principle of porography to optically process the 8AR data recorded on the film 1 to obtain an image. As shown in the figure, a conical lens 2 whose focal length becomes shorter toward the bottom is placed in front of the 8AR data film, and phase correction is performed so that the wavefront becomes a ray parallel to the X-axis in any part of the film. , [Azimuth (X-axis direction) FE reduction], the light exiting the conical lens is.

It is a cylindrical diverging wave in the y-axis direction. Therefore, the cylindrical lens 3 in the y-axis direction is used to correct the phase so that the rays become parallel to the y-axis direction [distance (y-axis direction) compression]
, the ray obtained by K in this way is.

A reproduced image was obtained on the image film 5 by condensing light beams parallel to both the X-axis and y-axis directions using a circular lens 4.

However, in such an optical processing method.

The drawback is that it is difficult to configure the lens system with high precision and to process the data (noise reduction, distortion correction, etc.).

The present invention solves the above-mentioned drawbacks by performing all digital processing on data acquired by a single-synthetic open radar, and enables highly flexible processing such as distortion correction and noise reduction during data processing, resulting in excellent accuracy. The present invention provides a synthetic aperture radar image processing device.

In other words, one F! In processing the synthetic open radar data, A uses data obtained by Fourier transforming the conjugate function of the transmitted signal during distance compression, and data obtained by Fourier transforming the time domain reference function during azimuth compression. A reproduced image is obtained through digital processing, making it possible to easily perform distortion correction and noise reduction operations during the processing, and achieving improved processing accuracy.

Next, an embodiment of the present invention will be described with reference to the drawings (FIG. 2). This figure shows the basic flow of digital image processing of synthetic aperture radar acquired data according to the present invention. The parameter calculation unit 7 calculates 8Al (from the data header file 6).

It reads the orbit and attitude data of the moving platform, calculates parameters such as velocity vectors on the earth's surface necessary for subsequent processing, and sends them to the distance compression section and azimuth compression section B. The distance compression section compresses the linear frequency modulated wave of the transmitted pulse into a sharp pulse, and compresses the target information spread as a two-dimensional hologram in the distance direction. This operation takes one line by correlating the received signal with the conjugate function of the transmitted signal, but it requires a large amount of data.

FFT t- is used to calculate in the frequency domain to speed up the calculation. The calculation is first performed by fast Fourier transforming the conjugate function (reference function) of the transmission signal [τ: width of the transmission signal, k: frequency sweep rate] obtained by the 1-time domain reference function generation circuit 8 using the FFT circuit 9. The data thus obtained is multiplied by 1iitll by the data obtained by performing fast Fourier transform in the distance direction using the 8AR data QFFT circuit 10. 12 performs inverse Fourier transform. The result is distance-compressed data arranged in the distance direction.

In order to azimuthally compress this data (described later), the corner turning circuit 13 performs an operation to rearrange the data in the azimuth direction.
Let's do it.

On the other hand, in the two-direction compression section B, the data spread in the azimuth direction (direction perpendicular to the distance direction).

Compress in azimuth direction. This operation is similar to distance compression,
This is done by correlating the 8AR data compressed in the distance direction with a reference function, but since the amount of data is large, FFT
= i is used to calculate in the frequency domain to speed up the calculation. tzu1
Time domain reference function generation circuit [F2: slope of Doppler frequency] Data fast Fourier transformed by the t-FFT circuit 16,
Distance-compressed 8AR data '1 Data obtained after fast Fourier transform in the azimuth direction by the FFT circuit 14 is multiplied by the multiplier 17? , the IPFT circuit 18 performs inverse Fourier transform.

Since the output data of this IPFT circuit 18 is complex data, its absolute value is outputted as image data 19. In order to completely reduce speckle and noise, the azimuth spectrum is divided, each is independently compressed and imaged, and then incoherently superimposed (multi-look processing) is performed. Software can perform distortion correction, immersion conversion, texture analysis, smoothing and sharpening of a single image, detection of image lines and boundaries, etc. on the image data.

The present invention, as described above, is accomplished by configuring a digital system as shown in FIG.

Accurate SAR image processing can be performed.

And with software, you can easily reduce noise, correct distortion,
This has the effect of allowing you to perform operations such as:

[Brief explanation of drawings]

FIG. 1 is a diagram showing an example of the configuration of an optical processing apparatus using a lens system that has been widely used in the past, and FIG. 2 is a block diagram showing an embodiment of the present invention. 1...SAR data film, 2...Conical lens, 3...Cylindrical lens,
4. Old... Circular lens, 5... Image film% 6... SAW data file, 7...
...Parameter calculation unit, 8゜15...Time domain reference function generation circuit, 9,10゜14.16...
・Fast Fourier transform circuit, 11.17... Multiplier, 12.18... Fast Fourier inverse transform circuit,
13... Corner turning circuit, 19...
Festival 1 time

Claims (1)

[Claims] In a synthetic aperture radar image processing device that improves range resolution and azimuth resolution by compressing synthetic aperture radar acquired data in the range direction and azimuth direction, the acquired data is compressed by Fourier processing in the range direction and azimuth direction. A synthetic aperture radar image processing device characterized in that data obtained by the conversion is multiplied by data obtained by Fourier transform by a conjugate function of a transmission signal, thereby compressing the data in the distance direction and the azimuth direction, respectively.