DE102016209803B3 - Method for creating an earth observation image of a region by means of a synthetic aperture radar - Google Patents

Abstract

In the method, an earth observation image of a region is created by means of a synthetic aperture radar (SAR) and a phased array antenna composed of individual antennas having an azimuthal swivel angle range obtainable by a steering angle. Each individual antenna has an elementary element gain describing the individual antenna gain as a function of the solid angle, and the array antenna has an array factor describing the array antenna gain as a function of solid angle resulting from the arrangement and number of individual antennas, the array antenna gain having several main maxima which are substantially equal and which, however, are attenuated toward increasing control angles in view of the overall gain resulting from the element factor and the array antenna gain. The usable azimuthal swivel angle range is widened in the detection of a region, by setting the same control angle multiple times within the unique control angle range given by the geometric arrangement of the individual antennas, and filtering out the ambiguous main maximum pointing in the respective desired control direction, whereby geometric and / or radiometric resolution of the SAR is improved. The pulse repetition frequency with which the radar pulses are transmitted is selected so that the respective desired and the undesired main maxima are separated in order to reduce the undesired main maxima and thus filtering out these main maxima is possible.

Description

The invention relates to a method for generating an earth observation image of a region by means of a synthetic aperture radar (SAR) and a phased array antenna composed of individual antennas with an azimuthal swivel angle range obtainable by a control angle, each individual antenna having an elementary antenna gain as a function of the solid angle the array antenna has an array factor which describes the group antenna gain as a function of the solid angle resulting from the arrangement and number of individual antennas, the array antenna gain having a plurality of main maxima which are substantially equal but which, in view of the overall gain resulting from the Element factor and the group antenna gain results are weakened towards increasing control angles. More particularly, the invention relates to a method of improving resolution by extending the useful swing angle range of a phased array Synthetic Aperture Radar (SAR) and SAR image quality enhancement portions of the synthetic SNR poor aperture (hereinafter also LNPC) or omega-weighting).

Staring Spotlight [1] is the SAR (Synthetic Aperture Radar SAR) imaging mode with the longest illumination duration of a target, and thus the best resolution. The antenna is directed while the flyby of the radar steadily on the center of the scene to be imaged (see 1 ). The illumination duration and thus the azimuthal resolution is determined by the maximum azimuthal pivoting range of the antenna.

Very often, the antenna is realized as a phased array antenna, which allows rapid changes in the direction of the antenna sight without requirements on the position of the carrier or an influence on the same. The alignment of the viewing direction is carried out electronically by excitation coefficients for the respective individual antennas. For each control angle, a set of excitation coefficients is usually stored on board the carrier. As a result, the adjustable control angle are limited in their number or quantization. In general, the distance of the excitation coefficient addressable individual antennas is greater than half the wavelength, which leads to the occurrence of grating lobes (English: Grating praise).

One prior art SAR system with staring spotlight imaging mode and phased array antenna is TerraSAR-X (TSX) [1]. In this exemplary system, the on-board tabulated excitation coefficients cover a range of azimuthal control angles between -2.2 ° and + 2.2 °. This angle range of 4.4 ° corresponds to the angular distance Δθ _GL between the main maxima of the azimuth antenna pattern. Main maxima are the grating lobes and the main lobe. It is possible to extend the swing angle range by a different distribution of the control angle, but this leads to an increase in the quantization and thus to quality losses.

1 Figure 12 shows the oblique passing of a wearer on the spotlight scene to be photographed from an observation point opposite the trajectory of the wearer with the spotlight scene intervening on the ground. The carrier flies over the scene to be recorded laterally at the speed V _s . The azimuthal viewing direction of the antenna is guided in such a way that it is constantly directed to the scene center, ie "staring" (to stare). The in 1 shown swing angle range corresponds to the Δθ _{GL of} the TSX system. The antenna lobes shown in solid lines point into the control angle set by the excitation coefficients.

One technique that is supported in the Wrapped Starring Spotlight mode described below is that of the diverging PRF setting [1], [2]. In SAR systems, the signal, which is in principle not band-limited, is sampled in azimuth at the pulse repetition frequency (PRF). In this case, signals from directions which do not correspond to azimuth frequencies within the PRF band are also mirrored by refolding into the PRF band. Through selective selection of the PRF used, the positioning of an undesired controlled or refolded main maximum can now be set relative to a desired controlled or refolded main maximum. Diverging PRF setting now means to set the position of the desired and undesired main maxima as far apart as possible from each other. Then as much of the unwanted main maximum can be removed with a bandpass filtering. 2 shows in a thick, solid line a desired main maximum centered in the middle of the PRF band, left with two undesired main maxima in a thin dashed line in Converging PRF setting, ie with a centering of these also in the middle of the PRF band. In the right panel, the unwanted major maxima are in diverging PRF setting.

The conventional SAR processing in the azimuthal dimension performs, among other things, a correction of the azimuth antenna diagram in the azimuth spectrum and then a weighting of each look in azimuth to control or suppress the sidelobes of the impulse response. The sidelobe suppression can also be located in the azimuth spectrum. For the processing of data of the exemplary TSX system, a "General Cosine Window" [1] with an α-parameter of 0.6 is assumed below. This reduces the geometric resolution by a factor of 1.3 compared to processing without sidelobe suppression. The maximum geometrical azimuth resolution in Staring Spotlight imaging mode so far resulting from the azimuth slewing angle range of 4.4 ° for the exemplary TSX system is 0.165 m without spectral weighting for sidelobe suppression and 0.225 m with sidelobe suppression [1]. The maximum sidelobe ratio with respect to the Peak Sidelobe Ratio (PSLR) is -13.2 dB with no sidelobe suppression and -31.6 dB with suppression.

There are two principal ways to graphically represent the available azimuthal bandwidth: in maximum geometric resolution, or by dividing the bandwidth into multiple views ("looks") and incoherently adding them, giving a representation with reduced speckle and degraded geometric resolution. Speckle is a fluctuation in brightness that is inherent in all SAR images and that makes it difficult to detect targets. Looks are often overlapped in SAR images, and a common measure used to quantify the reduction of Speckle is the Equivalent Number of Looks (ENL).

Another important quality parameter of SAR images is the amount of noise in the signal, which can be described by the signal-to-noise ratio (SNR). The dominant part of the noise is usually the thermal noise at the radar receiver. A high amount of noise in the SAR signal appears in the SAR image like a gray haze over the image. High noise also reduces the detectability of targets.

Both speckle and thermal noise fluctuate the measured radar backscatter cross section. For area targets, one can summarize ENL and SNR into a quality parameter and describe the detectability in the SAR image by the "radiometric resolution RR" as

wobei N_p die Anzahl von Bildpixeln mit Grauwerten von 0 bis 255 ist und g_p der Grauwert eines Pixels. Je kleiner C_rms ist, umso ähnlicher ist das Bild einem einzigen, idealen, das ganze Bild ausfüllende Flächenziel ohne Speckle und ohne Rauschen. Je größer C_rms für eine Szene mit gemischten Zielen, also Flächenzielen, urbane Gebieten und Punktzielen ist, umso weniger Rauschanteil ist vorhanden. Der maximale Kontrast ergibt sich, wenn die Hälfte der Pixel maximale und die andere Hälfte minimale Graustufe aufzeigen. Der maximale Kontrast entspricht der Hälfte der höchsten Graustufe, also 127,5 in Gleichung (1). Die Darstellung der zurückgestreuten Radarleistung erfolgt wegen des günstigeren Dynamikbereiches oft in der Wurzel der Intensität, also in Amplitude. Im Folgenden wird der Grauwert Null der Amplitude Null zugeordnet und der Grauwert 255 der Wurzel des 95. Perzentils der Intensität.The contrast is a helpful measure for comparing differently displayed SAR images. It provides conclusions about a change in the noise component and describes the detectability of scenes with heterogeneous backscatter cross-section (English: "Backscatter"), such as occurs in urban areas. There is no uniform definition of contrast for SAR. The following is the definition of the "Root Mean Square Contrast" _rms , which is defined similarly for optical images:

where N _{p is} the number of gray-scale image pixels from 0 to 255 and g _{p is} the gray value of a pixel. The smaller C _rms is, the more similar the image is to a single, ideal, full-frame area target without speckle and noise. The larger C _{rms is} for a scene with mixed targets, ie area targets, urban areas, and point targets, the less noise there is. The maximum contrast is obtained when half of the pixels show maximum gray scale and the other half. The maximum contrast is half of the highest gray level, which is 127.5 in Equation (1). The representation of the backscattered radar power is often due to the favorable dynamic range in the root of the intensity, ie in amplitude. In the following, the gray value zero is assigned to the amplitude zero and the gray value 255 to the root of the 95th percentile of the intensity.

The representation of the amplitude in 3 from the value 0 to the root of the 95th percentile of all intensity values corresponds linearly to the gray values from 0 to 255. In Table 1, the value of the root is given as 11.76. The measured contrast of the whole in 3 The area shown is 56.5 and for the area D of the urban area 75.2. An area target should ideally show the lowest possible contrast, ie low speckle and thermal noise. For the three area targets A, B and C, the contrast is in the table 1, as well as the measured ENL values and the mean SNR. From the last two results according to equation (1) the radiometric resolution to about 2.3 for the three surface targets.

Tabelle 1: Bildqualitätsparameter für den Ausschnitt aus einem Staring Spotlight Bild des TerraSAR-X Systems.

Table 1: Image quality parameter for the section from a staring spotlight image of the TerraSAR-X system.

• a) In Staring Spotlight SAR, either the azimuthal control angle range or the quantization of the control angles is limited by the available on-board memory. Due to the limited control angle range while maintaining the quantization, the maximum possible geometric resolution is limited. Maintaining geometric resolution limits radiometric resolution.
• b) In SAR acquisitions, when the angle range is extended to include angles in the synthetic aperture where the antenna gain is poor, the thermal noise in conventional SAR processing increases greatly and the SNR in the SAR image is poor , Therefore, these angular ranges are not processed and their information potential remains unused.

The object of the invention is to improve the resolution and image quality in SAR systems.

• – der nutzbare azimutale Schwenkwinkelbereich bei der Erfassung einer Region erweitert wird durch mehrmaliges Einstellen des gleichen Steuerwinkels innerhalb des eindeutigen, durch die geometrische Anordnung der Einzelantennen gegebenen Steuerwinkelbereichs, und durch Herausfiltern desjenigen mehrdeutigen Hauptmaximums, das in die jeweils gewünschte Steuerrichtung zeigt, wodurch die geometrische und/oder radiometrische Auflösung des SAR verbessert wird, wobei die Pulswiederholfrequenz, mit der die Radarpulse gesendet werden, so gewählt wird, dass die jeweils gewünschten und die ungewünschten Hauptmaxima zur Reduzierung der ungewünschten Hauptmaxima separtiert sind und somit ein Herausfiltern dieser Hauptmaxima möglich ist, und
• – mindestens eine Aufnahme der Region zur Erstellung eines Erdbeobachtungsbilds verarbeitet wird.

To solve this problem, the invention proposes a method for generating an earth observation image of a region by means of a synthetic aperture radar (SAR) and a phased array antenna composed of individual antennas with an achievable by a control angle azimuthal swivel angle range, each individual antenna in a single antenna gain in Dependent on the spatial angle descriptive element factor and the array antenna has an array factor that describes the resulting from the arrangement and number of individual antennas group antenna gain as a function of solid angle, the array antenna gain has several main maxima, which are substantially equal and which, however, with regard to Overall gain resulting from the element factor and the array antenna gain are attenuated toward increasing steering angles, with the method

• - The usable azimuthal swivel angle range in the detection of a region is extended by repeatedly setting the same control angle within the unique, given by the geometric arrangement of the individual control angle control range, and by filtering out that ambiguous main maximum pointing in the respective desired direction of control, whereby the geometric and / or radiometric resolution of the SAR is improved, the pulse repetition frequency with which the Radar pulses are sent, is selected so that the respective desired and the unwanted main maxima to reduce the unwanted main maxima are separtiert and thus filtering out these main maxima is possible, and
• - At least one image of the region is processed to create an Earth observation image.

According to one embodiment of the invention, in addition to the expansion of the usable azimuthal swivel angle range, the processing of the earth observation image takes place on the basis of multi-look processing, ie. H. is created by processing a total image of a region, which is composed of several partial shots (so-called looks) of the region by illuminating across different viewing angle ranges, the partial images are filtered out of the total recording, where

• Portions of the synthetic aperture which, at certain angles, have a signal-to-noise ratio which is less than a predeterminable limit and is therefore "poor", so that these portions of the synthetic aperture have not been used in previous methods through a signal-to-noise-dependent summation of the individual looks, with the summation either
• Derived directly from the azimuth antenna diagram, is derived by a normalized correction of the antenna pattern in the individual looks (LNPC), first of all correcting the antenna pattern and then normalizing each look in such a way that at a certain azimuth angle or, if the operation in the spectrum, at a certain azimuth frequency specific to each look, the product of inverse antenna pattern and normalization gives a value of one in each look, or
• Depending on the location, a signal-to-noise-dependent weighting of the looks (Ω-weighting), which is different at each pixel position, is performed, wherein the weighting function is created according to desired optimization criteria.

• – eine möglichst gute radiometrische Auflösung und/oder
• – eine möglichst gutes Signal-zu-Rausch-Verhältnis im Multi-Look-Bild und/oder
• – eine möglichst hohe Anzahl Equivalent Number of Looks handelt/handeln kann.

Further possible variants of the invention relate to a location-dependent weighting of the looks, by changing the optimization criterion for the creation of the weighting function

• - The best possible radiometric resolution and / or
• A best possible signal-to-noise ratio in the multi-look image and / or
• - can trade / trade as high as possible Equivalent Number of Looks.

A further aspect of the invention can be seen in that the recording in the spotlight mode and thus by extending the illumination of the region to be recorded as a result of the expansion of the azimuthal swivel angle range, or in TOPSAR mode by extending the azimuthal swivel angle range, or bi-directional SAR mode or in forward / backward looking SAR mode.

The various aspects of the invention are explained below with reference to the drawing. In detail, they show:

1 a graphical representation of the well-known Staring Spotlight mode with the maximum azimuthal control angles of the TerraSAR-X system (TSX),

2 a graphic representation of the Converging PRF setting (left) and Diverging PRF setting (right),

3 an exemplary excerpt from a staring spotlight image of the TerraSAR-X system (TSX), where three area targets (A, B and C) and one urban area (D) are marked with buildings,

4 a schematic representation of the invention Staring Spotlight mode with extended azimuthal control angle range by exploiting the main maxima using the example of the TerraSAR-X system (TSX),

5 a selection of the desired signal components or the suppression of the unwanted signal components, by time-dependent band-pass filtering and diverging PRF setting, the given angles being exemplary for the TerraSAR-X system (TSX),

6 the basic block diagram of the inventive LNPC multilook processing on the range compressed and azimuth phase corrected full azimuth spectrum,

7 the basic block diagram of the Ω-weighting multilook processing according to the invention on the basis of the range compressed and azimuth phase-corrected complete azimuth spectrum,

8th a graphical representation of the control angle of the left TerraSAR-X Wrapped-Staring-Spotlight-Aufnahme according to 9 , where in thin line the commanded, repetitive, so "wrapped" control angle, and in thick line the "unwrapped" control angle, that is made available by filtering and diverging PRF setting and ultimately processed control angle is shown,

9 Exemplary Wrapped Staring Spotlight images of the TerraSAR-X system, where on the left a scene with urban areas and area targets is shown and a section for an evaluation described in the description of the figure is marked and on the right a scene with a marked point target is shown in the figure description is analyzed in more detail,

10 on the left the azimuth resolution, measured for the in 9 selected point target, where in dashed lines the measurements for Element Pattern Weighting (EPW), as a solid line the theoretical / simulated values and dotted in comparison the theoretical / simulated resolution values for a complete correction of the antenna diagram are shown and where on the right the measured and simulated values are shown for the highest sidelobe ratio of the impulse response,

11 Azimuth antenna diagrams and spectra, left for LNPC and right for Ω weighting, taking into account the different scaling of the amplitudes shown, and in the diagrams (a) the profile of the azimuth spectrum in solid and the calculated antenna diagram shown in dashed line is shown in the diagrams (b) the azimuth antenna pattern correction specific for each of the nine looks shown alternately in solid and dashed lines, in the diagrams (c) the profile of the azimuth spectrum after the antenna pattern correction shown and in the diagrams (d) the profile of the azimuth spectrum after correction of the antenna pattern and weighting for sidelobe suppression are shown,

12 To the left, a cutout optical Google Earth image 9 Left and right, the amplitude image of the SAR image in classic processing of 9L _FPC , where the nine looks are completely corrected in terms of the azimuth antenna diagram and where on the left and on the right, different area targets and mixed context destination areas are marked by white frames.

13 , a plot of the optimized Ω weights for M = 32 classes and a SNR _{ML, min} of 0 dB, where the X-axis shows the look number and the Y-axis the SNR of the central look of the classes,

14 ENL, SNR and radiometric resolution for the optimized classes of the embodiment, in thick dashed line for a SNR _{ML, min} of 0 dB and in a thin solid line for 3dB and

15 marked excerpts of the in 9 SAR acquisition shown on the left, with 3L _{FPC standing} for three looks with full antenna pattern correction - this is very close to the operationally implemented staring spotlight mode of the TerraSAR-X system -, 9L _{LNPC stands} for nine looks with LNPC multilooking and 9L _{Ω, 0dB} , 9L _{Ω, 3dB} for Ω weighting multilooking in the variants 0 dB and 3 dB minimum desired SNR in the multi-look image.

In an existing phased array SAR system, the capture mode of the present invention (hereafter referred to as wrapped staring spotlight mode) substantially extends the slew angle range over the conventional staring spotlight capture mode, without additional on-board space requirements and without a deterioration of the quantization of the control angle. Prerequisite is that the angular distance Δθ _GL between the main maxima is covered in sufficient quantization of the control angle. The expansion then takes place by repeatedly setting the control angles within Δθ _GL , but not adding the signal components of the control direction of the synthetic aperture actually set by the excitation coefficients, but those signal components coming from the desired direction of another principal maximum of the antenna pattern.

4 again shows the acquisition geometry of the exemplary TSX system, but for a wrapped staring spotlight image. The middle three antenna positions are already in 1 shown for the staring spotlight mode. The two outer antenna positions in 4 now show the extension of the swivel angle range. In order to record signals from the directions ± 4,4 °, at the two Positions again set the excitation coefficients for 0 ° control angle. By diverging PRF setting and corresponding filtering in the processing then the desired control angle can be used. For point targets, the extended control angle range directly improves the geometric resolution.

The selection of the desired signal components or the suppression of unwanted is done by a time-dependent bandpass filtering. This process is in 5 illustrated. During the acquisition, the desired control angle is set directly at any time, or by taking advantage of an ambiguous main maximum. This is in 5 shown by the thick line drawn in dashed lines, which decreases over the azimuth time t _a , ie the recording time, steadily from + 4,4 ° steering angle to -4,4 ° steering angle. The vertical extent of the band drawn around this line in a thin solid line around this control angle is the so-called scene angle, which indicates the extent of the processed scene in the azimuth direction. It is set either by an azimuth time-dependent bandpass filter in the processing, or equivalent after processing by selecting the scene extent in the finished processed image. Specifically, the azimuth time-dependent filter can be realized, for example, by sub-aperture processing with different center frequencies in the sub-apertures [3]. The variation of the desired control angle over the azimuth acquisition time determines the maximum geometric resolution and is referred to as the unwrapped resolution angle. The main maxima shown in dotted lines in the upper right and lower left parts of the illustration are shifted from the desired one by the angle Δθ _GL . The illustrated gain ratios are qualitatively similar to those of the exemplary TerraSAR-X system. The bottom row of plots represents the relative positioning of the desired and strongest unwanted major maxima in the PRF band for diverging PRF setting. For the ± 2.2 ° control angles, the desired and unwanted major maxima are equal but due to diverging PRF -Setting separated from each other. The unwanted main maxima can be suppressed by the scene angle corresponding to a band pass. For the very large control in the exemplary TerraSAR-X system up to ± 4.4 °, the gain values from the desired control angle directions are substantially lower than those from undesired directions. Even with diverging PRF setting, the remaining gain from these directions is larger than that from the desired directions.

The SAR images resulting from the formation of the synthetic aperture involving poor SNR control angle regions can be compared to conventional SAR processing by a new single weighted SNR based on the two quality parameters "contrast" and "radiometric resolution" Look-process can be significantly improved. Two main variants of the novel multi-look processing are cited below as an exemplary embodiment, namely Look Normalized Pattern Correction (LNPC) with a constant weighting of the individual look images, which is constant over the pixels and thus location-independent. Weighting (English: Ω-weighting) with a pixel-dependent and thus location-weighted weighting of the look images. LNPC is easier to implement. Ω weighting is more flexible and can be better tuned to the detection of specific goals.

LNPC variant

LNPC weights the individual azimuth look images differently in the addition to the multi-look image. The different weighting results from a normalized correction of the azimuth antenna diagram. For example, in each look, the normalization may be related to the mean azimuth frequency of the look, i. H. at the mean azimuth frequencies of the looks, the correction would be 1. Each pixel in the resulting multi-look image has exactly equal weighted portions of each look. LNPC is thus location-invariant.

6 shows the principal block diagram of LNPC Multi-Looking. The starting point is the complete azimuth spectrum, which is compressed in the range direction and completely corrected in azimuth, including the correction of the azimuth modulation.

Important for the LNPC multi-looking is the spectral correction of the antenna diagram by the amplitude _weighting function A _{LNPC, k} (f _a ) in equation (3), which depends on the azimuth frequency f _a and the number of the look. The amplitude of the two-way azimuth antenna diagram of the total spectrum is designated AAP _2W (f _a ). The central Doppler centroid (Doppler centroid) of the look k is denoted by f _{c , k} . The rectangle function rect selects for each look k from the total spectrum the look bandwidth B _L centered around f _{c, k} . The antenna diagram is corrected for all frequencies f _a but standardized by the multiplication with AAP _2W (f _{c, k} ) to the value 1 at the central Doppler frequency.

At the same place in the expiration of 6 where the multiplication by A _{LNPC, k} (f _a ) occurs, the desired _{balance for sidelobe suppression} is also performed by multiplication by M _{L , k} (f _a ). This weighting function extends over the bandwidth B _L and is shifted for each look to be centered around f _{c, k} . After the inverse Fast Fourier Transformation (IFFT) in 6 For each image pixel p, there are respectively k look pixels with the intensities I _{p, k} . These are then summed according to the rule in equation (4) to the multi-look pixel intensity I _{p, ML} .

C _cal is an arbitrary calibration constant to which all necessary, but not required for the description of the method constants are summarized. Because of the different weighting of the look pixels in the summation to the multi-look pixel, a normalization of the sum by the denominator in equation (4) is required. Here, W _{I, S} [k] is a factor for look k defined in equation (5), which describes the ratio of the intensity of a look pixel before and after the spectral LNPC weighting and the weighting for sidelobe suppression. The integral over the look bandwidth B _L is centered around f _{c, k} , and includes the weighting of the intensity through the azimuth antenna pattern, which is effective over the bandwidth of the look k on it.

Ω Weighting variant

The name Ω-Weighting should make it clear that this weighting is the last one that will be applied to the individual looks, and only in the look screen area. As in 7 shown, the correction of the antenna pattern and the weighting for sidelobe suppression before done, still in the azimuth spectral range. The correction of the azimuth antenna diagram is done completely by multiplying the amplitude function A _{FPC, k} (f _a ) to the amplitude spectrum _m , as shown in equation (6). The index FPC stands for complete correction of the antenna pattern (English: Full Pattern Correction).

In Ω weighting, the multi-look summation of the intensities of the individual look pixels I _{p, k, FPC} is done differently for each pixel p over the looks k as described in equation (7). The determination of the individual weights Ω _{p, k} is dependent on the SNR of the individual look pixels and will be described below.

W _{I, S, FPC} [k] is the factor for look k defined in equation (8), which describes the ratio of the intensity of a look pixel before and after the complete spectral correction of the antenna pattern and the weighting for sidelobe suppression.

Determination of the weights Ω _{p, k}

• i) dass gegenüber einer konventionellen SAR-Verarbeitung eine verbesserte radiometrische Auflösung zeigt und
• ii) bei dem sich der Bildkontrast durch die Erhöhung des thermischen Rauschens nicht verschlechtert bzw. sogar verbessert.

The goal in optimizing weights is to generate a multi-look image that includes poor SNR looks,

• i) that shows improved radiometric resolution compared to conventional SAR processing, and
• ii) in which the image contrast does not deteriorate or even improve by the increase of the thermal noise.

A simple solution would be to derive the weight Ω _{p, k} directly from the measured SNR of the pixel p in the look k, ie directly from S ~ NR _{p, FPC} [k]. The tilde displays a reading. The downside of this is that look pixels with large measured intensity are always heavily weighted and pixels low intensity are always weighted weakly, which increases the speckle-induced intensity fluctuations, ie increases the speckle. This solution is therefore eliminated.

To limit the influence of speckle on the weights, Ω weighting uses the information of all the looks of an image pixel. As in the basic block diagram of 7 1, the method according to the invention uses the antenna pattern AAP _2W (f _a ) and the measured SNR values of all pixels and looks S ~ NR _{p, FPC} [k] to define classes of pixel SNR curves over the looks. Then the weights Ω _m [k] are optimized for each class m so that the radiometric resolution becomes as small as possible for a given minimum required SNR in the multi-look image SNR _{ML, min} . If SNR _{ML, min is} not reached for a class, all weights of that class are set to zero. Finally, for each multi-look image pixel to be formed with the SNR curve of the measured look pixels S ~ NR _{p, FPC} [k], the associated class m is determined and the weight Ω _m [k] of this class in the multi-look -Summation of equation (7), ie, Ω _{p, k} [k] is set to Ω _m [k].

For Ω weighting, the thermal noise is reduced for areas with low backscatter, ie. H. exactly in the areas where the thermal noise disturbs the most. Higher backscatter cross-section tolerates more noise because it improves the number of looks and ultimately the radiometric resolution.

Without loss of generality, one possibility for defining the classes, for optimizing the weights of the classes, and for assigning the SNR profile of a pixel over the looks to a class in the exemplary embodiment is shown below.

In summary, the aspects essential to the invention can be described as follows.

In an existing SAR system with phased array antenna and a distance of excitation coefficient responsive single antennas greater than half the wavelength and on-board stored excitation coefficients for azimuth antenna control, the wrapped-spotlight spotlighting mode will extend the pan angle range over the conventional staring spotlight recording mode essential. This is done by an additional utilization of main maxima, which form next to the main maximum in the actually set main maximum direction, without additional on-board storage space requirement and without a deterioration of the quantization of the control angle. Diverging PRF setting during data acquisition supports the selection of the desired signal components or the suppression of unwanted signal components, which is achieved by time-dependent band-pass filtering. The added value for dot targets is the improvement in geometric resolution associated with the expansion of the pan angle range. The added value for area targets and mixed image content is improved radiometric resolution and improved contrast in multi-look images.

The core of the Wrapped Staring Spotlight method can also be used for other SAR modes, such as bi-directional SAR or forward / backward-looking SAR, ie existing ones can be used SAR systems azimuth viewing directions are evaluated, which are not in the originally designed range of control angles.

The LNPC-Look Normalized Pattern Correction option improves the resulting multi-look SAR images with respect to the "contrast" and "quality" quality parameters in the formation of the synthetic aperture, including control angle ranges with poor SNR. radiometric resolution "over conventional state-of-the-art SAR processing through a novel one-shot SNR-based weighted multi-look processing. With LNPC, a weighting of the individual look images is constant over the pixel positions and thus independent of location.

LNPC can also be used for other SAR modes with azimuthal location-varying SNR, such as ScanSAR, TOPSAR, Staring Spotlight, and Sliding Spotlight.

The Ω weighting improves the resulting multi-look SAR images with respect to the quality parameters "contrast" and "radiometric resolution" in comparison to a conventional one in the formation of the synthetic aperture including control angle ranges with poor SNR State of the art SAR processing by a novel SNR of the individual looks dependent weighted multi-look processing. With the Ω weighting, the weighting of the look images is dependent on the location of the pixels and thus on the location.

The Ω weighting can also be applied to other SAR modes with azimuthal location-variant SNR, such as ScanSAR, TOPSAR, Staring Spotlight, and Sliding Spotlight.

As described above, with the TerraSAR-X system, two acquisitions were made in Wrapped Starring Spotlight Capture mode. The commanded over the acquisition time angle for the left in 9 Shown are in 8th shown in a thin line. The repeated setting of the control angle within Δθ _GL = 4.4 ° is clearly visible. The achieved by exploiting the grating lobes wide control angle range of 8.8 ° is plotted in thick black line, which overplotted the thin line in the central region of about 3.5 s to 10 s.

9 shows the two Wrapped-Staring-Spotlight images in an overview with several looks. The azimuth processing was performed with the full acquired azimuth control angle range of 8.8 °. Parts of the two images are reviewed below to demonstrate the benefits of the Wrapped Staring Spotlight acquisition process, as well as LNPC multi-look processing and Ω weighting multi-look processing.

The extent of wrapped staring spotlight images in 9 corresponds in azimuth direction to a scene angle in the size of the azimuth 3dB opening angle of the antenna, ie the projection of the opening angle to the earth's surface at the distance of the scene center. The marked section in the left picture in 9 is in 3 shown in a multi-look presentation. 3 is approximately the same as the prior art staring spotlight imaging mode of the exemplary TSX system. The marked section in the right shot of 9 contains a three-way reflector (English: Corner Reflector), which appears in the picture as a point target. Both in 9 marked sections are analyzed below and thus the proof of the advantages of the invention is led.

point targets

The impulse response to the point target in the right section of 9 has been analyzed for growing processed bandwidth. The bandwidth of the TerraSAR-X system in the staring spotlight mode is 38 kHz. All other bandwidths have been generated by using the Wrapped Staring Spotlight mode. The measured geometrical azimuth resolution improves with increasing processed bandwidth.

The processing was performed without correction of the antenna pattern in azimuth. This preserves the weight of the element antenna pattern in the raw data record. This weighting is referred to below as EPW (Element Pattern Weighting). With increasing bandwidth, this weighting changes from a similarity to the rectangular weighting to a weighting with strong suppression of the sidelobes. This is reflected in 10 in the right diagram, which shows the strongest peak-to-side lobe ratio (PSLR) of the point target response over the Processed bandwidth shows. The PSLR values of an ideal simulation are shown in dashed lines. The values measured in the TerraSAR X Wrapped Staring Spotlight image for the point target are shown in solid lines. The PSLR drops from the value for a rectangular weighting of -13.2 dB with increasing bandwidth, since the effective weighting develops towards a stronger sidelobe suppression. The measured values agree very well with the simulation.

The geometric resolution left in 10 is measured over the bandwidth of the existing Staring Spotlight mode of the TerraSAR-X system for EPW weighting. It agrees very well with the theoretical / simulated values. From 55 kHz results in a small deviation; for the maximum processed bandwidth of 76 kHz, it is 5%, presumably due to the high altitude information requirements of the location of the point target. However, this is a requirement of the method used for focusing in the azimuth direction. However, the increasing improvement of the measured geometric resolution shows the improvement of the geometric resolution compared to the operational existing staring spotlight mode of the TerraSAR-X system, which provides an azimuth resolution of 0.19 m at 38 kHz maximum bandwidth and EPW. The additionally opened by Wrapped Staring Spotlight control angle ranges can thus be successfully integrated for point targets in the synthetic aperture and extend it thereby or allow an improved geometric resolution in azimuth.

For comparison, the simulated geometric resolution for a rectangle weighting (rect), ie for a complete correction of the azimuth antenna diagram, is given.

Area targets and mixed image content

In order to demonstrate an improvement of the image quality for area targets and for image parts with mixed target areas, over the whole range of wrapped staring spotlight recording in 9 formed a number of nine looks with a bandwidth of 15.2 kHz on the left. Looks overlap by 50% and each look is weighted with the sidelobe suppression weight described above. 11 shows profiles of the azimuth spectrum for LNPC (left) and Ω weighting (right). The upper diagrams (a) are identical and show the profile of the spectrum and the calculated antenna diagram in dashed line. Both the spectrum and the antenna diagram show variations due to the quantization of the antenna control and the traveling pulses (ie partly different transmission and reception antenna diagrams because of the long signal propagation time> 1 / PRF). In the outer, that is the magnitude larger azimuth frequencies that in comparison to the antenna pattern (English: Pattern) higher profile of the spectrum shows a greatly increased noise in the real recorded data. The calculated antenna diagram contains no noise.

The diagrams (b) of the 11 show the applied corrections of the antenna diagram. For LNPC, the amplitude of the correction is equal to one at the central look frequency. With Ω-weighting, a complete correction of the antenna pattern (English: Full Pattern Correction FPC) takes place. The profiles of the corrected azimuth spectra show strong upward at the ends of the outer looks, which means "dragging up" the noise. This effect is much more pronounced in Ω weighting. Note the different scaling of the amplitudes as well as the limitation of the displayed amplitude. The lower plots (d) show the spectra after additional weighting for sidelobe suppression.

Both LNPC and Ω weighting multilooking produce the same form of impulse responses to a point target in all looks. Ie. the signal spectra of a point target are equally weighted. In 11 the overall spectra are shown, ie signal and noise. The differences in 11 (d) arise due to the different "raised" noise. For point targets in each look, this leads to equal performance values of the point target response: geometric resolution and sidelobe relationships.

12 shows an optical Google Earth image of the clipping in the left 9 Left. Right in 12 is the result of classical azimuth processing of the nine looks with complete correction of the azimuth antenna pattern and sidelobe suppression, hereafter referred to as 9L _FPC . Different area targets and target areas with mixed context are marked by white borders.

Out 12 It immediately becomes clear that the SNR in the multilook picture with classical processing 9L _{FPC is} completely insufficient for image evaluation due to the poor SNR of the outer looks. 15 shows the results for LNPC Multilooking (9L _LNPC ) and Ω weighting multilooking in the variants 0 dB minimum desired SNR in the multi-look image (9L _{Ω, 0dB} ) and 3 dB (9L _{Ω, 3dB} ). For comparison, the variant with three looks and full azimuth-antenna-pattern-correction (3L _FPC ) is shown, which is close to operational implemented staring spotlight mode of the TerraSAR-X system. The representation of the amplitude is in all the amplitude images shown in FIG 15 as described above. The azimuth resolution in all looks and in the multi-look image is 0.55 m with a scan of 0.26 m. In the distance dimension (here oblique view), the geometric resolution is 0.58 m for a sample of 0.22 m.

• i) Das SNR wird in jedem Look für jeden Pixel gemessen. Um den Einfluss des Speckle zu verringern, wird dabei eine gleitende Mittelwertbildung unter Einbeziehung der acht benachbarten Pixel angewendet.
• ii) Für den zentralen Look wird ein Histogramm erstellt und M Klassen werden so definiert, dass jede Klasse m in etwa gleich viele SNR-Werte enthält.
• iii) Die SNR-Werte der anderen Looks werden ausgehend vom SNR-Wert des zentralen Looks mit Hilfe des Azimut-Antennendiagramms ergänzt.
• iv) Die Optimierung wird für jede Klasse durchgeführt, so wie oben beschrieben. Es wird also für ein vorgegebenes gewünschtes minimales SNR eines Pixels des Multi-Look-Bildes SNR_ML,min eine Optimierung der Ω-Gewichte Ω_m[k] der einzelnen Looks k so durchgeführt, dass eine öglichst gute radiometrische Auflösung erzielt wird.
• v) An jeder Pixelposition wird dem Verlauf der SNR-Werte (ohne gleitende Mittelwertbildung) über den einzelnen Looks eine Klasse zugeordnet. Dies erfolgt über den minimalen Gesamtabstand zwischen dem Verlauf der gemessenen SNR-Werte und der SNR-Werte der Klassen.
• vi) Die für die Klasse optimierten Ω-Gewichte werden in der Multi-Look-Summation für die jeweilige Pixelposition laut Gleichung (7) verwendet.

The adjustment of the individual weights of the Ω weighting was carried out in this embodiment as follows.

• i) The SNR is measured in each look for each pixel. In order to reduce the influence of speckle, a moving averaging with the eight neighboring pixels is used.
• ii) A histogram is created for the central look and M classes are defined so that each class m contains approximately equal numbers of SNR values.
• iii) The SNR values of the other looks are supplemented from the SNR value of the central look using the azimuth antenna diagram.
• iv) Optimization is performed for each class as described above. Thus, for a given desired minimum SNR of a pixel of the multi-look image SNR _{ML, min} an optimization of the Ω weights Ω _m [k] of the individual looks k is performed in such a way that the best possible radiometric resolution is achieved.
• v) At each pixel location, a class is assigned to the history of SNR values (without moving averaging) across each look. This is done via the minimum total distance between the course of the measured SNR values and the SNR values of the classes.
• vi) The Ω weights optimized for the class are used in the multi-look summation for the respective pixel position according to equation (7).

13 shows the optimized weights for M = 32 classes and a minimum desired SNR of 0 dB in each pixel of the multi-look image. The y-axis does not directly show the number of the class, but the SNR of the central look of the class. 14 shows the result of optimizing the Ω weights for the classes, namely the Equivalent Number of Looks (ENL), the SNR and the Radiometric Resolution (RR).

Comparison of the image results for the entire drawing file, surface targets and mixed image content

Tabelle 2: Messwerte für die in Fig. 15 dargestellten Multi-Look-Bilder, die mit verschiedenen Multi-Look-Methoden generiert wurden. Die Werte für 3L_FPC wurden teilweise schon in Tabelle 1 angegeben. Table 2 summarizes the overview for the respective multilook picture in 15 Multilook method used together indicates the amplitude limits used to represent the gray levels and gives measurements for different areas of the image.

Table 2: Measured values for the multi-look images shown in FIG. 15, which were generated with different multi-look methods. The values for 3L _FPC have been partially given in Table 1.

Whole fields

The image 3L _FPC contains little noise because only the inner 3 looks with good SNR were used, but a high speckle, which is reflected in the ENL value of about 2 for the surface targets. Although a fixed low weight is used for the outer looks for 9L _LNPC , the ENL is improved to a value of 4. However, the noise is increased due to the outer looks, which is visible in the image by the gray haze, which can be seen in the comparison with 3L _FPC . Another indicator of the increased thermal noise is the reduction of the contrast from 56.5 to 48.3. The improvement in terms of speckle is reflected in the improved ENL values. Image contrast is improved to 57.1 for Ω weighting at a desired minimum multilook SNR of SNR _{ML, min} = 3 dB.

area targets

The improvement of the ENL is for the three area targets A, B, and C in 15 remeasured. They have a moderate and high radar backscatter cross section. The improvement in the ENL is greatest for 9L _LNPC as the increase in noise also contributes to the improvement in ENL statistics. The improvement for 9L _{Ω, 0dB} and 9L _{Ω, 3dB} , however, is in the same order of magnitude. For the high radar backscatter cross section of area target C, the improvement in the Ω weighting images is higher than in the 9L _LNPC because more looks were summed with higher weight as a consequence of the better SNR. The higher weighting is in 13 visible, noticeable. The measured SNR is best for 3L _FPC and lower for the other three images. However, the radiometric resolution, which results from SNR and ENL, is substantially improved for the methods according to the invention "LNPC and Ω-weighting". The measured image contrast in Table 2 substantially decreases in the area targets for both inventive methods due to the improved radiometric resolution.

Mixed image content

The improvement in speckle and noise is also evident in the scenery of area targets and driveways of the target area E. In 3L _FPC the detectability of the paths is limited by the high speckle and in 9L _LNPC by the high noise. The best detectability for this scene is achieved with 9L _{Ω, 3 dB,} ie with Ω-weighting and a high SNR required in multi-look image. The diminished speckle is obvious and the measured contrast is similar to 3L _FPC .

Artificial objects such as buildings generate strong backscatter, z. B. edges or surfaces perpendicular to the direction of the radar, and extremely low backscatter in shaded areas. This leads to a high image contrast. High noise with low backscatter reduces the image contrast and thus the possibilities for detection and classification for urban areas. Table 2 gives the measured contrast for the target area D with buildings and the target area F in 15 at. The increase in noise from 3L _FPC to 9L _LNPC reduces the contrast because the areas with low radar backscatter cross section appear brighter due to the noise floor. Ω-weighting significantly improves image contrast through the high weighting of the outer looks for strong targets and the simultaneous low weighting of the outer looks for the weak radar backscatter cross section of shaded areas. Target area F includes a roof structure. The pictures with Ω-weighting show reduced speckle for the area target around the roof structure. The improvement in SNR from 9L _LNPC over 9L _{Ω, 0dB} to 9L _{Ω, 3dB} is well observable. For 9L _{Ω, 3dB} , several pixels are set to zero because the expected minimum multi-look SNR is less than the desired 3 dB.

Existing SAR systems can be improved by the Wrapped Staring Spotlight mode in the swivel angle range and thus in the geometric resolution in azimuth. Examples of systems currently in operation are the TerraSAR-X system, the Canadian Radarsat system, the Japanese ALOS 2 / PalSAR 2 system or the Italian Cosmo-Skymed system. The Spanish PAZ system, which is about to launch, is also a potential system that can be improved with wrapped staring spotlight. All future SAR systems with phased array antennas will benefit from the Wrapped Staring Spotlight mode. Airborne SAR systems with azimuthally controllable, phased array antenna can be improved by the Wrapped Staring Spotlight mode subsequently with respect to geometric and / or radiometric resolution. The core of the Wrapped Staring Spotlight method can also be used for other SAR modes, such as Bi-directional SAR or Forward / Backward-Looking SAR, i. H. In existing SAR systems, azimuth viewing directions that are not in the originally designed range of the control angles can be evaluated. The TOPSAR mode can also be extended by extending the swivel angle range. An example of such a system is Sentinel-1.

Image quality can be significantly improved by LNPC or Ω weighting in SAR acquisitions with parts of the synthetic aperture of poor SNR. Both methods can be incorporated into SAR processors from the beginning or later, and allow for the enhancement of multi-look SAR images in terms of contrast and radiometric resolution quality parameters. Examples are once again the Wrapped Staring Spotlight mode SAR systems mentioned above.

bibliography

• [1] J. Mittermayer, S. Wollstadt, P. Prats-Iraola, R. Scheiber: "The TerraSAR-X Staring Spotlight Fashion Concept", IEEE Trans. To Geosc. Remote Sens., Vol. 52, no. 6, pp. 3695-3706, June 2014.
• [2] J. Mittermayer, S. Wollstadt, P. Prats, P. Lopez-Dekker, G. Krieger, A. Moreira: "Bidirectional SAR Imaging Mode", IEEE Trans. To Geosc. Remote Sens., Vol. 51, no. 1, pp. 601-614, January, 2013.
• [3] J. Mittermayer, A. Moreira and O. Loffeld, "Spotlight SAR Data Processing Using the Frequency Scaling Algorithm," IEEE Trans. To Geosc. Remote Sens., Vol. 37, no. 5, pp. 2198-2214, September, 1999.

Claims (4)

A method for creating an earth observation image of a region by means of a synthetic aperture radar (SAR) and a phased array antenna composed of individual antennas having an azimuthal swivel angle range obtainable by a steering angle, each individual antenna an elementary element gain describing the individual antenna gain as a function of the solid angle, and the array antenna having an array factor describing the group antenna gain as a function of solid angle resulting from the arrangement and number of individual antennas, the array antenna gain having a plurality of main maxima which are substantially equal and which, however with respect to the overall gain resulting from the element factor and the array antenna gain, are attenuated toward increasing control angles, wherein in the method - the useful azimuthal swing angle range is increased upon detection of a region by setting the same control angle multiple times within the unique, by the geometric arrangement of the individual antennas given control angle range and by filtering out that ambiguous main maximum, which points in the respectively desired direction of control, whereby the geometric and / or radiometric resolution of the SAR is improved, wherein the pulse repetition frequency at which the radar pulses are sent, is selected so that the respectively desired and the unwanted main maxima are separated to reduce the unwanted main maxima and thus filtering out these main maxima is possible and - at least one photograph of the region is processed to produce an earth observation image. A method according to claim 1, characterized in that the earth observation image is produced on the basis of a multi-look processing, ie on the processing of a total image of a region that is composed of a plurality of partial views of the region created by illuminating over different viewing angle regions. in which the partial recordings are respectively filtered out of the total recording, wherein - proportions of the synthetic aperture which, at certain viewing angles, have a signal-to-noise ratio which is smaller than a predefinable limit value can be utilized by a signal-to-noise dependent summation of the individual looks, wherein the summation is derived either directly from the azimuth antenna diagram, regardless of location, by a normalized correction of the antenna pattern in the individual looks, first of all correcting the antenna pattern and then performing normalization in each look For example, at a certain azimuth angle or, if the operation is performed in the spectrum, at a certain azimuth frequency specific to each look, the product of inverse antenna pattern and normalization will give a value of one in each look, or, depending on location, one at each pixel position different signal-to-noise-dependent weighting of the looks is performed, wherein the weighting function is created according to desired optimization criteria. A method according to claim 2, characterized in that with location-dependent weighting of the looks, the optimization criterion for the creation of the weighting function - the best possible radiometric resolution and / or - the best possible signal-to-noise ratio in the multi-look image and / or - is the highest possible number of Equivalent Number of Looks. Method according to one of the preceding claims, characterized in that the recording in the spotlight mode and thus by extending the illumination of the male region due to the expansion of the azimuthal swivel angle range, or in TOPSAR mode by extending the azimuthal swivel angle range, or bi-directional SAR mode or in forward / backward looking SAR mode.

Images