RU2597144C1 - Method for remote earth probing - Google Patents

Abstract

FIELD: solar engineering.

SUBSTANCE: method for remote Earth probing involves obtaining a optical flux of the Sun radiation, reflected from the probed area of the Earth's surface. Then the flux is divided into two beams of equal intensity, predetector adaptive compensation of random wavefront inclinations is performed at one of said beam, due to turbulent atmosphere, and accumulation of adaptively stabilized short expositional images is performed at the second beam. Images are accumulated at quadratic detection during recording, that is larger than time correlation interval of atmospheric fluctuations, and the average short expositional image is recorded, said image is transmitted to the Earth, where its space is filtered and the high resolution image of the probed area of the Earth's surface is restored.

EFFECT: technical result is faster process of obtaining high quality images of the Earth.

1 cl, 3 dwg

Description

The claimed invention relates to the field of optical instrumentation and is intended to promptly obtain high-quality (high resolution) images of probed sections of the earth's surface observed from outer space through a turbulent atmosphere.

Known methods for remote sensing of the Earth, implemented in the Russian satellite Remote Sensing "Resource-DK1" described in the article by G. Petri "Russian satellite" Resource-DK1: an alternative source of ultra-high resolution data ", (Geomatics, No. 4, 2010,
p. 38 ÷ 42 [1]), which has been in orbit since June 15, 2006, and in the recently launched (June 26, 2013) Russian satellite “Resource-P”, described in the article by A.N. Kirilin et al. “Resurs-P spacecraft,” (Geomatics, No. 4, 2010, pp. 23–26).

) и наблюдаемого из Космоса через турбулентную атмосферу, спектральной фильтрации и квадратичном детектировании сформированного изображения, его длинно-экспозиционной регистрации и передаче на Землю по радиолинии для последующей цифровой обработки и тематического анализа.These methods of remote sensing of the Earth are based on the formation of an optical image of the probed portion of the earth’s surface, illuminated by the radiation of the Sun (

) and observed from outer space through a turbulent atmosphere, spectral filtering and quadratic detection of the formed image, its long-exposure registration and transmission to the Earth via a radio line for subsequent digital processing and thematic analysis.

The disadvantages of these methods, considered here as analogues of the proposed method, is that they do not take into account and correct the effect of atmospheric turbulence, which leads to a limitation of the resolution attainable in them.

Thus, the presence of a spectral filtering band Δλ ≈ 1000 A ⁰ in them leads to the frequency averaging of atmospheric fluctuations of light radiation in the generated image, and the WZN technology (time delay and accumulation) used in them leads to the detection and recording of long-exposure images, averaged over atmospheric distortion. The noted frequency and time averaging of atmospheric fluctuations in the generated and detected image limits the resolution of the probed portion of the Earth’s surface in a refractor telescope with an aperture diameter D = 0.5 m of the order of ~ 1 m.

Known methods for remote sensing of the Earth from Space, implemented in foreign remote sensing satellites, such as QuickBird, WorldView 1 and 2 and GeoEye-1 manufactured in the USA, described in the article by V.V. Lavrova “Space-based filming systems of super-high resolution”, (see GIS-Association Geoinformation portal, No. 2, 2010 [3]). These remote sensing spacecraft, like the Russian ones, are based on obtaining a long-exposure image of the probed area of the earth’s surface and its subsequent thematic processing without taking into account and compensating for the effect of atmospheric turbulence. Here, as in the aforementioned Russian methods of remote sensing of the Earth, long exposure images are recorded when implementing the technology of time delay and accumulation (TDI).

The difference between these remote sensing methods from Russian is the use of mirror optics to form an image instead of a lens and increasing the size of the forming optics to an aperture diameter of D = 1.1 m. This allowed us to increase the image quality of the probed area of the earth’s surface by reducing the spatial resolution element: in QuickBird to 0.6 m, in WorldView up to 0.5 m and in GeoEye-1 up to 0.4 m. These methods are also considered here as analogues.

As a prototype of the proposed method, we took the "Remote Sensing Method of the Earth (Remote Sensing)" [4], described by the patent of the Russian Federation No. 2531024 from 08.20.2014. Patent holder - Russian Space Systems OJSC, author K.N. Sviridov.

In this method, unlike analogs, the effect of a turbulent atmosphere on remote sensing images is taken into account, its compensation is carried out algorithmically by using a long-exposure remote sensing image and a series of N spectrally-filtered short-exposure images that are independent of each other by atmospheric distortions.

(∆λ≈250 A⁰) ограничивает энергетику изображений ДЗЗ, восстанавливаемых с высоким разрешением.The sequence of operations proposed in the prototype allows to improve the quality (resolution) of the average short-exposure image of the probed portion of the Earth's surface in comparison with the average long-exposure image of analogues. The technology for obtaining high-resolution images proposed in the prototype is lengthy, due to the inertia of the quadratic detector and the application of the “selective” frame strategy to obtain independent short-exposure images in a series, and the statistical processing of short-exposure images in isoplanatic regions also requires significant time costs, except Moreover, limited by the required absence of frequency averaging of atmospheric distortions, the narrow spectral filtering band ∆ _A = Δλ

(∆λ≈250 A ⁰ ) limits the energy of remote sensing images reconstructed with high resolution.

The improvement in the quality of long-exposure images of remote sensing with an increase in the size of the receiving equipment D of the telescope forming the image in foreign spacecraft (D = 1.1 m) compared to domestic (D = 0.5 m) is easily explained by the work of remote sensing in the near zone (zone Fresnel) when a diverging spherical wave of reflected backlight radiation propagates from each point of the probed portion of the earth's surface.

In FIG. Figure 1 shows the geometry of the propagation of a spherical wave from the earth's surface to an ERS telescope through a turbulent atmosphere. This shows that when a spherical wave propagates from the Earth’s surface to an ERS spacecraft to a height H, all spatial characteristics of the spherical wave, including the distortions acquired by it in a turbulent atmosphere (lower L = 10 km, near the Earth’s surface) will increase as it propagates to heights H and can turn out to be significantly larger than the aperture D of the remote sensing telescope.

Indeed, this is confirmed by our studies (see, for example, K.N. Sviridov “On the Ultimate Resolution of Aerospace Remote Sensing Earth Systems (ERS)” in the scientific and technical journal “Rocket and Space Instrumentation and Information Systems” Volume 1, Issue 1, 2014, p. 34–40) [5].

H) увеличивается с высотой H до космического аппарата ДЗЗ и определяется как According to [5], the magnitude of radius of the spatial fluctuations of atmospheric correlation r ₀ (

H) increases with height H to the remote sensing spacecraft and is defined as

,H)≈

, (1)r ₀ (

, H) ≈

, (one)

=0,1 м на границе турбулентного слоя L = 10 км. При этом на высоте H величина r₀(

,H) оказывается большей величины апертуры D. При этом атмосферные искажения волнового фронта на приемной апертуре телескопа ДЗЗ (Рис.1) представлены только случайными наклонами волнового фронта, вызывающими случайные сдвиги изображения ДЗЗ в процессе его длинно-экспозиционной регистрации. Это соображение и лежит в основе предлагаемого здесь способа.Where

= 0.1 m at the boundary of the turbulent layer L = 10 km. Moreover, at a height H, the quantity r ₀ (

, H) it turns out that the aperture D is larger. At the same time, the atmospheric distortions of the wavefront at the receiving aperture of the remote sensing telescope (Fig. 1) are represented only by random tilts of the wavefront, which cause random shifts of the image of the remote sensing during its long-exposure registration. This consideration underlies the method proposed here.

The proposed method for remote sensing of the Earth, in contrast to the prototype, is based not on an algorithmic, but on a hardware solution to the problem of achieving high quality (resolution) of remote sensing images.

The technical result of the proposed method for remote sensing of the Earth is to accelerate the process of compensating for atmospheric distortions to achieve high resolution images of remote sensing.

), отраженного от зондируемого участка земной поверхности, и разделяют его на два пучка равной интенсивности, по одному из которых осуществляют преддетекторную адаптивную компенсацию случайных наклонов волнового фронта, обусловленных турбулентной атмосферой, а по другому пучку осуществляют накопление адаптивно стабилизированных коротко-экспозиционных изображений, для чего по первому пучку измеряют величину и направление случайных наклонов волнового фронта и адаптивно компенсируют их в принимаемом потоке светового излучения за время τ_КЭ, меньшее интервала временной корреляции атмосферных флуктуаций τ_A, (так называемого времени «замороженности» турбулентностей атмосферы), одновременно по второму пучку осуществляют спектральную фильтрацию адаптивно скорректированного по первому пучку принимаемого светового излучения в полосе ∆λ=

/

_θ*_,D (где

_θ*_,D- среднеквадратичное отклонение атмосферных флуктуаций фазы θ волнового фронта светового излучения на приемной апертуре D при наличии адаптивной компенсации случайных наклонов волнового фронта) и, фокусируя его, формируют адаптивно стабилизированные коротко-экспозиционные изображения ДЗЗ, накапливают их при квадратичном детектировании за время регистрации τ_Э большее интервала временной корреляции атмосферных флуктуаций τ_А, и регистрируют среднее коротко-экспозиционное изображение, которое передают на Землю, где его пространственно фильтруют и восстанавливают изображение зондируемого участка земной поверхности, обладающее высоким разрешением.The technical result is achieved by collimating the solar light flux received by the telescope (

) reflected from the probed portion of the earth’s surface and divide it into two beams of equal intensity, one of which carries out pre-detector adaptive compensation of random wavefront tilts caused by the turbulent atmosphere, and the other beam carries out the accumulation of adaptively stabilized short-exposure images, for which the first beam measures the magnitude and direction of random tilts of the wavefront and adaptively compensate them in the received stream of light radiation over time I τ _TBE minimal interval time atmospheric fluctuations correlation τ _A, (the so-called time "frozen" atmospheric turbulence), while the second beam is carried out spectral filtering is adaptively adjusted by the first beam received in the light emission band Δλ =

/

_θ * _{, D} (where

_θ * _{, D} is the standard deviation of atmospheric fluctuations of the phase θ of the wavefront of light radiation at the receiving aperture D in the presence of adaptive compensation for random slopes of the wavefront) and, focusing it, form adaptively stabilized short-exposure images of remote sensing, accumulate them during quadratic detection during registration τ _E larger correlation time interval τ _A of atmospheric fluctuation is detected and the average short-exposure image that is transmitted to Earth, where its space Twain filtered and reduced image of the probed part of the earth surface, having a high resolution.

The features and essence of the claimed invention are explained in the following detailed description, illustrated by drawings, which shows the following:

In FIG. 2 presents a variant of the implementation scheme of the claimed method, which shows:

FIG. 2a is a block diagram of an adaptive remote sensing imaging channel, where:

1 - a mirror telescope of a remote sensing spacecraft;

2 - collimating optics;

3 - flat adaptive oscillating mirror - compensator for random tilts of the wavefront;

4 - translucent beam splitting mirror;

5 - detector of random tilts of the wavefront — shear interferometer;

6 - block adaptive control swinging mirror;

7 - light filters;

8 - focusing optics;

9 - the focal plane of the image;

FIG. 2b is a structural diagram of a channel for detecting and recording remote sensing images, where:

10 - electromechanical shutter;

11 - image intensifier;

12 - transfer optics;

13 - quadratic panoramic detector;

14 - digital video processing system;

15 - on-board computer;

16 - encoding device;

17 - airborne radar;

FIG. 2c is a structural diagram of a remote sensing image processing channel, where:

18 - ground radar;

19 - decoding device;

20 - computing means;

21 - software;

22 - algorithmic support for the processing of remote sensing images;

23 - operator workstation;

24 - consumers of high resolution remote sensing images.

In FIG. 3 shows the dependence of the normalized resolution R / R _max on the normalized diameter of the telescope D / r ₀ for the average long-exposure R _D-E / R _max and the average short-exposure image R _K-E / R _max [4].

It should be noted that the remote sensing imaging channel and the remote sensing image detection and registration channel are located onboard the remote sensing spacecraft, and the remote sensing image processing channel is located on Earth and is connected to them via a radio link.

Consider the implementation of the proposed method in accordance with the circuit shown in FIG. 2.

The telescope of the DZZ-1 spacecraft receives light from the sun reflected from the probed portion of the earth’s surface and forms a distorted image of the probed portion of the earth’s surface observed from space through the turbulent atmosphere.

The collimating optics -2 creates a parallel beam of received light radiation necessary for the proper operation of subsequent optical elements.

3 - adaptive flat mirror, compensating for random tilts of the wavefront of the received light radiation;

4 - a beam splitting mirror dividing the received light radiation into two beams of equal intensity.

_θ,D атмосферных искажений фазы θ волнового фронта светового излучения и определяется выражениемUsing the first beam in the detector of random tilts of the wavefront — the shear interferometer — 5 measure the magnitude and direction of the random tilts of the wavefront and, using the control unit — 6, according to information received from the interferometer 5, the slopes of the adaptive plane mirror 3 controlled by the tilts compensate for the random tilts of the wavefront of the received light radiation for time τ _KE less than the interval of time correlation of atmospheric fluctuations τ _A. In accordance with the studies (see, for example, KN Sviridov, “Technologies for achieving high angular resolution of optical atmospheric vision systems,“ Knowledge ”publication, M. 2005) [6], the value of τ _A varies over a wide range of τ _A = (1 ÷ 100) ms and has the minimum values of τ _Аmin = 1 ms. Given this, it is necessary to adaptively compensate for random tilts of the wavefront for a time τ _K-E ≤ τ _A = 1 ms. Next, a beam 2 with an adaptively corrected wavefront is spectrally filtered by a turret with light filters 7. In this case, the frequency band of spectral filtering ∆λ is selected based on the standard deviation

_{θ, D of} atmospheric distortions of the phase θ of the wavefront of light radiation and is determined by the expression

/

_θ,D (2)Δλ =

/

_{θ, D} (2)

 (see, for example, KN Sviridov et al. “Static Evaluation of the Spectral Band of the Spot Interferometry Method” Optics and Spectroscopy, vol. 54, issue 5, p. 890, 1983) [7].

The presence of spectral filtering in this band ensures the coherence of atmospheric phase fluctuations in the received light radiation.

_θ,D атмосферных искажений фазы волнового фронта определяется в радианах как [6]In the absence of adaptive compensation for random slopes of the wavefront (in the prototype), the standard deviation

_{θ, D of} atmospheric distortions of the wavefront phase is defined in radians as [6]

_θ,D= 2π 0,16 (

^5/6[рад] (3)

_{θ, D} = 2π 0.16 (

^5/6 [rad] (3)

А⁰ равняется величине ∆λ ≤ 250 A⁰.and has a maximum value of the order of 20 radians. In this case, the required value of spectral filtering in the prototype when

And ⁰ is equal to ∆λ ≤ 250 A ⁰ .

_θ*_,D уменьшается и определяется выражениемWith adaptive compensation of random wavefront tilts, the quantity

_θ * _{, D} decreases and is determined by the expression

_θ*_,D = 2π 0,06 (

^5/6[рад] (4)

_θ * _{, D} = 2π 0.06 (

^5/6 [rad] (4)

In this case, the permissible value of spectral filtering in the proposed method increases in comparison with the prototype 2.66 times and turns out to be equal to Δλ ≤ 667 A ⁰ . This extension of the required spectral filtering band in the proposed method allows to increase the energy of the formed image.

Focusing optics 8 forms filtered and adaptively stabilized, by compensating for random tilts, short-exposure images of the probed portion of the earth's surface in the focal plane 9, containing a shutter 10 of the channel for detecting and recording remote sensing images.

The electromechanical shutter 10, synchronized with the frame rate of the quadratic panoramic detector 13, provides the required accumulation time of adaptively stabilized short-exposure images to obtain an average short-exposure image.

The brightness amplifier of the image 11 should ensure the operation of the quadratic detector 13 in a quantum-limited mode, when quantum noise prevails over other noise of the detection process. As an amplifier for image brightness on board the spacecraft, it is advisable to use microchannel plates (MCPs), which have less weight and dimensions than electron-optical converters (ICs).

The transfer optics 12 projects the image of the object from the output of the image intensifier 11 to the input of the quadratic panoramic detector 13 without changing the angle. As the transfer optics, you can use, for example, a set of lenses or fiber optics with less light loss during image transmission.

A quadratic panoramic detector 13 (for example, a CCD) is designed for quadratic detection of adaptively stabilized short-exposure images and their accumulation during registration to obtain an average short-exposure remote sensing image.

In the proposed method, in contrast to the prototype, the technology of "selective" frame is not used to obtain an average short-exposure image, which formed short-exposure images 14, independent of each other by atmospheric distortion, and eliminating the influence of inertia of the quadratic detector. In the proposed method for the accumulation of adaptive stabilized short-exposure images, the technology of detection and accumulation of each frame is used.

In this case, the gain in time of obtaining each short-exposure image here for a single frame time τ _K-E is equal to the interval τ _p = 7 · τ _K between the independent short-exposure images formed in the prototype.

This gain in time of obtaining the average short-exposure remote sensing image in the proposed method compared with the prototype is even greater, since in the proposed method there is no need for time-consuming post-detector processing of independent short-exposure images in their isoplanatic areas.

The digital video processing system 14 is designed to digitize the average short-exposure image coming from the quadratic detector 13. The video signals from the quadratic detector 13 digitized in the video processing system 14 are recorded in the digital memory of the on-board computer 15.

Next, the digitized average short-exposure images from the computer 15 are transferred to the encoding device 16 and transmitted via the radar - 17 to the Earth via a radio link for further processing.

Processing average short-exposure images of the probed portion of the earth’s surface obtained on board the remote sensing spacecraft in accordance with FIG. 2a and FIG. 2b and transmitted over the radio link to the Earth, carried out according to the scheme shown in FIG. 2c.

The ground-based radar 18 receives radio information from the airborne radar 17.

The decoding device 19 converts the signals from the ground radar 18 to a form convenient for writing to the memory of computing means 20.

Computing tools 20 are designed to implement an algorithm for processing medium short-exposure images of remote sensing images 22.

Software (software) 21 computing tools 20, is intended to organize the process of computing tools 20 to implement the image processing algorithm of remote sensing 22.

Algorithmic support 22 for processing average short-exposure images of remote sensing represents a sequence of operations on the registered average short-exposure image, ensuring the achievement of the goal - improving the quality (resolution) of processed images of the probed area of the earth's surface.

AWP of the operator 23 is designed to control the processing process and analyze the quality (resolution) of processed images in the process of implementing the image processing algorithm of remote sensing images 22.

In accordance with the algorithm for processing medium short exposure images proposed here, the following sequence of operations is performed:

1) transform the recorded average short-exposure remote sensing image according to Fourier into the region of the spatial spectrum

d

= <

(

>_к-э (5)

d

= <

(

> _ke (5)

2) spatially filter the spectrum of the formed medium short-exposure image of the average short-exposure optical transfer function of the atmosphere-telescope system;

(

>_к-э·

= <

(

(6)<

(

> _k-e

= <

(

(6)

3) with the opposite of item 1) Fourier transform from the filtered spatial spectrum of item 2) restore the image of the probed portion of the earth's surface, free from the influence of atmospheric turbulence and having high resolution.

(

d

=

(7)

(

d

=

(7)

The image processing algorithm proposed here 22 is the last three operations of the image processing algorithm 18 in the prototype. These three operations represent one third of the nine operations that process the images in the prototype, and although the operations are different in complexity and execution time, we can conditionally assume that the processing proposed here will be carried out three times faster than in the prototype.

The mathematical justification of the proposed method ERS similar mathematical basis prototype [4], in which the analytical function and the graphs (Fig. 3) depending on the normalized resolution R / R _max of the normalized diameter telescope D / r ₀ for the average long-exposure image R _{D -E} / R _max (remote sensing system without atmospheric distortion compensation) and the average short-exposure image R _К-Э / R _max (remote sensing system with adaptive compensation for random wavefront tilts). Analytically, the dependence for the average short-exposure image ([4] formula (19) was obtained by eliminating the instantaneous wavefront slope by averaging. This corresponds to the adaptive pre-detector compensation of random wavefront slopes proposed here, which indicates that a high resolution is achieved in the proposed method for significantly less time than in the prototype.

Indeed, adaptive compensation of random wavefront tilts and obtaining an average short exposure image during accumulation of adaptively stabilized instantaneous images using the detection strategy of each frame, in contrast to the strategy of a “selective” frame in the prototype, can reduce the time to obtain the average short exposure image in comparison with the prototype 7 times due to the elimination of the procedure for obtaining short-exposure independent from each other by atmospheric distortions images in every eighth sample frame. When processing images in the proposed method compared with the prototype there is an additional reduction in time by more than 3 times. Thus, the proposed method for obtaining high-resolution remote sensing images in comparison with the prototype allows to accelerate the process of obtaining high-quality remote sensing images by ~ 21 times.

_θ*_,D атмосферных искажений фазы θ волнового фронта светового излучения в 2,66 раза (0,16/0,06), что позволяет увеличить допустимую величину полосы спектральной фильтрации ∆λ =

/

_θ*_,D также в 2,66 раза и расширить ее с величины 250А⁰ в прототипе до величины ∆λ = 667А⁰ в предлагаемом способе. Это увеличение полосы спектральной фильтрации позволяет улучшить энергетические характеристики формируемого изображения зондируемого участка земной поверхности.At the same time, adaptive compensation of random wavefront tilts reduces the standard deviation

_θ * _{, D of} atmospheric distortions of the phase θ of the wavefront of light radiation by 2.66 times (0.16 / 0.06), which allows us to increase the allowable value of the spectral filtering band Δλ =

/

_θ * _{, D is} also 2.66 times and expand it from a value of 250A ⁰ in the prototype to a value Δλ = 667A ⁰ in the proposed method. This increase in the spectral filtering band allows to improve the energy characteristics of the generated image of the probed portion of the earth's surface.

Here, as in the prototype, the maximum gain in resolution compared to analogs is about 8 times (4 times for pre-detector adaptive compensation of random wavefront tilts and about 2 times for post-detector filtering of the average short-exposure image).

The main goal achieved in this invention, the acceleration of the process of obtaining high-quality remote sensing images (resolution), is extremely important in the conditions of motion of a spaceborne remote sensing spacecraft.

Information sources

1. G. Petry “Russian satellite“ Resource-DK1 ”: an alternative source of ultra-high resolution data”, Geomatika, No. 4, 2010, p. 38-42, (analogue).

2. A.N. Kirilin et al. “Resource-P Spacecraft, Geomatics, No. 4, 2010, pp. 23-26, (analogue).

3. VV Lavrov “Space-based filming systems of ultra-high resolution” Geoinformation portal of the GIS-Association, No. 2, 2010, (analogues).

4. KN Sviridov “Remote Sensing Method of the Earth (ERS)”, author's application for invention No. 2013125540 dated 06/03/2013, RF Patent No. 2531024 dated 08/20/2014, patent holder of Russian Space Systems OJSC, (prototype).

5. KN Sviridov “On the Ultimate Resolution of Aerospace Remote Sensing Earth (ERS) Systems”, Rocket and Space Instrument Making and Information Systems, Volume 1, Issue 1, 2014, p. 34-40.

6. KN Sviridov “Technologies for achieving high angular resolution of optical systems for atmospheric vision”, ed. “Knowledge”, M., 2005.

7. K. N. Sviridov et al. “Statistical estimation of the spectral band of the spot interferometry method”, Optics and Spectroscopy, vol. 54, issue 5, p. 890, 1983.

Claims (1)

The method of remote sensing of the Earth, which consists in collimating the light flux of the sun received by the telescope (

) reflected from the probed portion of the earth’s surface and divide it into two beams of equal intensity, one of which carries out pre-detector adaptive compensation of random wavefront tilts caused by the turbulent atmosphere, and the other beam carries out the accumulation of adaptively stabilized short-exposure images, for which the first beam measures the magnitude and direction of random tilts of the wavefront and adaptively compensate them in the received stream of light radiation over time I τ _K-E minimal interval time correlation atmospheric fluctuations - τ _A, while the second beam is carried out spectral filtering is adaptively adjusted by the first beam of the received light radiation in the band Δλ =

/

_θ * _{, D} (where

_θ * _{, D} is the standard deviation of atmospheric fluctuations of the phase θ of the wavefront of light radiation at the receiving aperture D in the presence of adaptive compensation for random slopes of the wavefront) and, focusing it, form adaptively stabilized short-exposure images of remote sensing, accumulate them during quadratic detection during registration τ _E, greater than the time correlation interval of atmospheric fluctuations τ _A , and record the average short-exposure image, which is transmitted to the Earth, where its space They have a high resolution and filter and restore the image of the probed portion of the earth's surface.

Images