CN106772362A - Rail SAR high is vertical to non-homogeneous Vegetation canopy backscattering coefficient analogy method - Google Patents

Abstract

It is vertical to non-homogeneous Vegetation canopy backscattering coefficient analogy method the invention relates to a kind of rail SAR high, comprise the following steps：Determine the species of Vegetation canopy component；Determine the vertical to distribution function of various types of vegetation component volume density；Calculate the thin layer back scattering matrix in double-matrix algorithm；Calculate the backscattering coefficient of Vegetation canopy.Vegetation canopy assembly body dense vertical is introduced double-matrix algorithm and carries out backscattering coefficient simulation by the present invention to distribution function, for the vertical influence research caused to backscattering coefficient to heterogeneity of simulation Vegetation canopy provides approach, it is particularly suited for rail SAR observations high.

Description

Simulation Method of Vertical Backscattering Coefficient of Non-uniform Vegetation Canopy with High Orbit SAR

technical field

The invention relates to the research field of vegetation radar remote sensing observation, in particular to a method for simulating the backscatter coefficient of vertical non-uniform vegetation canopy by high-orbit SAR.

Background technique

Vegetation canopy backscatter coefficient simulation technology is used to explain the interaction between radar electromagnetic waves and vegetation canopy, which can assist in the development of vegetation canopy parameter inversion technology and carry out research on vegetation space observation technology based on radar remote sensing. The current numerical calculation model of the vegetation canopy backscatter coefficient mostly assumes that the vegetation canopy is vertically uniform, and lacks the description of the vertical non-uniformity of the vegetation canopy. However, there is extensive vertical heterogeneity in the vegetation canopy in nature. The assumption of uniformity limits the ability of the vegetation backscatter coefficient model to be used for theoretical analysis, and it is impossible to make accurate predictions on the interaction mechanism between microwave electromagnetic waves and vegetation. Explanation.

It can be seen that the above-mentioned existing vegetation canopy backscatter coefficient simulation method obviously still has inconvenience and defects in the method and use, and needs to be further improved, especially the use of high-orbit synthetic aperture radar (SAR) ) observations, since it is located in a geosynchronous orbit at an altitude of 360,000 km, higher requirements are put forward for the observation of vegetation vertical inhomogeneity. Therefore, how to create a new numerical simulation method for the backscatter coefficient that can describe the vertical non-uniformity of the vegetation canopy is one of the important research and development topics in this field.

Contents of the invention

The technical problem to be solved by the present invention is to provide a high-orbit SAR vertically non-uniform vegetation canopy backscatter coefficient simulation method, so that it can correct, remove or analyze the vertically non-uniform vegetation when using high-orbit SAR for vegetation observation. This provides a way to overcome the shortcomings of the existing simulation methods for vegetation canopy backscatter coefficients.

In order to solve the above-mentioned technical problems, a method for simulating the backscattering coefficient of vertical non-uniform vegetation canopy by high-orbit SAR of the present invention comprises the following steps: determining the type of vegetation canopy components; determining the vertical direction of the volume density of various types of vegetation components distribution function N _j (h); calculating the backscattering matrix of the thin layer in the double-matrix algorithm; calculating the backscattering coefficient of the vegetation canopy based on the double-matrix algorithm. Among them, the calculation formulas of the thin-layer backscattering matrix S and forward scattering matrix T in the double-matrix algorithm are:

In the formula: θ _i is the incident angle, θ _s is the scattering angle, is the incident azimuth angle, is the scattering azimuth, U ^-1 is a diagonal matrix, the matrix elements are the cosines of the scattering directions, P is the bi-station scattering phase matrix, which is determined by the incident electromagnetic field and the morphological structure and dielectric properties of the vegetation components, h is the vegetation component in The position in the canopy, Δz is the thickness of each thin layer, N is the density of vegetation canopy components in the unit volume of the thin layer, n is the number of types of vegetation canopy components, and j represents the jth type of vegetation canopy components.

The types of the vegetation canopy components include leaves, stems, branches, fruits or flowers and the like.

The backscattering coefficient of described calculation vegetation canopy comprises the following steps: calculating the backscattering matrix S' and the forward scattering matrix T' of the thin layer that the thickness that two tiny thin layers constitute up and down is 2Δz; Repeat above-mentioned calculation to obtain The backscattering matrix S ⁰ of the medium layer with the required thickness; calculate the backscattering coefficient σ ⁰ ; where, the calculation formulas of the backscattering matrix S' and the forward scattering matrix T' are:

The formula for calculating the backscattering coefficient is: σ ⁰ =4πS ⁰

In the formula, S1 is the backscattering matrix of the _first layer, T1 is the forward scattering matrix of the _first layer, S2 is the backscattering matrix of the _second layer, and T2 is the forward scattering matrix of the _second layer , * indicates the back and forward scattering matrix when the incident direction is reverse incident.

After adopting such a design, the present invention introduces the volume density vertical distribution function of various types of vegetation components in the vegetation canopy into the double matrix algorithm to simulate the backscattering coefficient. The research on the impact of coefficients provides a way, especially for the observation of vegetation targets by SAR satellites in high-orbit mode.

Description of drawings

The above is only an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

Fig. 1 is a schematic flow chart of the steps of the method for simulating the backscatter coefficient of the vertical non-uniform vegetation canopy in the high-orbit SAR of the present invention.

Figure 2 is a schematic diagram of the volume density N _j defined in the modeling space and the volume density μ _j defined in the interior space of the vegetation canopy in the high-orbit SAR observation mode.

Fig. 3 is the calculation of the multiple scattering process of two adjacent thin layers for unit incident energy by using the double matrix algorithm in the present invention.

Fig. 4 is a schematic diagram of a cone canopy for a simulation experiment of the present invention.

Fig. 5 is the simulated results of the VV polarization backscattering coefficients of the same ideal canopy under the assumptions of vertical uniformity and vertical non-uniformity when the simulation experiment is carried out in the present invention.

detailed description

Please refer to shown in Fig. 1, the main steps of a kind of high-orbit SAR vertical non-uniform vegetation canopy backscattering coefficient simulation method of the present invention comprise determining the kind of vegetation canopy component, determine the vertical direction distribution function of vegetation component volume density, according to The determined vertical distribution function calculates the thin-layer scattering matrix in the double-matrix algorithm, and calculates the backscatter coefficient of the vegetation canopy.

Specifically, first, the types of vegetation components in the vegetation canopy are determined.

The vegetation component refers to the smallest unit for calculating the scattering matrix in the discrete microwave scattering model, and the types include leaves, stems, branches, fruits, flowers, etc.

Secondly, determine the vertical distribution function of the first type of volume density of various types of vegetation components.

The density of various types of vegetation components refers to the number of vegetation components of this type in a unit volume, and the unit is "one per cubic meter" or "one per cubic decimeter". Since the volume density will change due to different measurement spaces, it is necessary to distinguish between the two volume densities N _j and μ _j involved in this patent. Please refer to Figure 2, where the former N _j represents the volume density of the jth type of canopy component in the scattering model modeling space in the SAR satellite observation mode, and the latter μ _j represents the jth type of vegetation component in the canopy space body density. For simplicity, N _j is called the first type of volume density in this patent, and μ _j is called the second type of volume density. The "volume density" in "determining the vertical distribution function of the volume density of various types of vegetation components" in the embodiment of this step refers to the first type of volume density N _j .

Note that the vertical distribution function of the first type of volume density of the jth type of vegetation component in the canopy is N _j (h), where h represents the position variable of the vegetation component in the canopy in the height upward direction, h=0 at the bottom of the canopy, and record the canopy The layer thickness is H, then the highest point of the canopy has h=H.

N _j (h) can be determined in four ways:

(1) Field measurement. By measuring the quantity of various types of vegetation components in the canopy at each height and dividing by the modeling space volume, the vertical distribution function of the first type of volume density of various types of vegetation components in the canopy is obtained;

(2) Growth model. The virtual canopy of the same plant can be obtained through the three-dimensional plant growth model, and the vertical distribution function of the first type of volume density of various vegetation components in the canopy can be obtained according to the position of various vegetation components in the virtual canopy;

(3) Equivalent replacement. Replace the vertical distribution function of the volume density of various vegetation components in the canopy with the vertical distribution function of other variables, such as NDVI (Normalized Differential Vegetation Index, normalized vegetation index), etc.;

(4) Theoretical assumptions. Sometimes in order to easily obtain the vertical distribution function of the first type of volume density of various vegetation components in the canopy, the specific functional form of N _j (h) can also be given directly based on experience, for example, N _j (h)=1000* (10-h)h∈[0,10].

Again, the thin-layer scattering matrix in the double-matrix algorithm is calculated according to the determined vertical distribution function.

The determined vertical distribution function N _j (h) of the volume density of various vegetation components in the canopy is to be incorporated into the framework of the double matrix algorithm.

In the double-matrix algorithm, firstly, the canopy is divided into a series of thin layers, and the thickness of each thin layer is Δz, and P is the single scattering phase matrix of the thin layer. The forward scattering matrix S and T can be expressed as:

S(θ _s ,θ _i ,φ _s -φ _i )＝U ^-1 P(θ _s ,θ _i ,φ _s -φ _i )Δz

T(θ _t ,θ _i ,φ _t -φ _i )＝U ^-1 P(θ _t ,θ _i ,φ _t -φ _i )Δz

Among them, θ _i is the incident angle; θ _s is the scattering angle; is the incident azimuth; is the scattering azimuth; U ^-1 is a diagonal matrix, and the elements are the cosines of the scattering direction; n is the number of types of plant components; is the first-class average volume density of the jth vegetation component in the thin layer at h in the vegetation canopy, the unit is unit, and it is a constant in the vertically uniform canopy. The measurement of is relatively simple. It can measure the number M of plant components of the jth type in the unit volume of the canopy, so that where V is the total volume of the canopy. It can also be calculated from N _j (h), the relationship between the two is:

The calculation formula of the back and forward scattering matrix of the rewritten vertical non-uniform canopy is:

Through the rewritten equation, the vertical distribution function N _j (h) of the number of various types of vegetation canopy components is included in the framework of the double matrix algorithm, and the relationship with the vegetation canopy scattering matrix is established.

Finally, the backscatter coefficient of the vegetation canopy is calculated based on the double matrix algorithm.

This calculation process can be realized through the existing vegetation backscatter coefficient model.

Please refer to Figure 3, in the process of multiple scattering between two equal-thickness thin layers, the back and forward scattering matrices of the first layer are S ₁ and T ₁ , and the backward and forward scattering matrices of the second layer are The scattering matrices are S ₂ and T ₂ . The superscript "*" indicates the backward and forward scattering matrices when the incident angle is reversed (that is, the incident angle is rotated by 180 degrees in the incident plane). Using the double-matrix algorithm, the back and forward scattering matrices S', T' of a thin layer with a thickness of 2Δz composed of upper and lower two tiny thin layers can be obtained.

Repeat this process to get the scattering matrix of any thickness dielectric layer. The relationship between the backscattering matrix S ⁰ and the backscattering coefficient σ ⁰ is:

σ ⁰ =4πS ⁰ .

The following is a comparative simulation experiment of modeling the scattering characteristics of the ideal cone tree canopy shown in Figure 2 based on vertical uniformity and vertical non-uniformity.

Please refer to Figure 4, the characteristics of the ideal cone canopy constructed in this example are as follows:

(1) The canopy is constructed according to the pine canopy, including two types of vegetation components: needles and branches;

(2) The cone angle α of the cone canopy is set to 30°, the thickness of the canopy is H, and the radius of the bottom layer is R=H·tanα, it is easy to get Ω(h)=π·(Hh) ² ·tan ² α;

(3) Without losing the generality of the orientation of the vegetation components, the leaves are assumed to be placed horizontally, and the azimuth is symmetrical; the branches are assumed to be placed vertically;

(4) Vegetation components are uniformly distributed in the canopy envelope space, that is, the second type volume density μ _j is a constant, which is set to 8×10 ^-5 cm ^-3 in this example;

(5) Other canopy parameters and incident electromagnetic field parameters are shown in the following table:

According to the steps established in the detailed description:

First, identify the species of vegetation components in the vegetation canopy. In this example, the canopy is an ideal canopy constructed artificially, and the types of vegetation components are needles and branches.

Second, determine the vertical distribution function of the volume density of various types of vegetation components. In this example, the canopy is artificially constructed, which is equivalent to giving a definite growth model, so that the precise position of each component can be obtained, and the vertical distribution function N _j (h) of the first type of volume density of various types of vegetation components can be further determined .

The vertical distribution function of the first type of volume density N _j (h) can be known from the calculation of the canopy shape cone and the second type of volume density μ _j :

Again, the thin-layer scattering matrix in the double-matrix algorithm is calculated according to the determined vertical distribution function.

Finally, the backscatter coefficient of the vegetation canopy is calculated based on the double matrix algorithm.

The simulation results of the backscatter coefficient of the ideal canopy under the VV polarization mode are shown in Fig. 5. The simulation results show that the difference between the simulation results of the vertically uniform and vertically non-uniform canopy is about -3.4dB, indicating that the vertically non-uniform It has a significant impact on the backscattering coefficient of the vegetation canopy. For SAR vegetation observation, especially in the high-orbit SAR observation mode, it is a factor that must be considered in the vegetation microwave scattering model. The technical solution proposed by the present invention is useful for subsequent Relevant accurate research is crucial.

The above is only a preferred embodiment of the present invention, and does not limit the present invention in any form. Those skilled in the art make some simple modifications, equivalent changes or modifications by using the technical content disclosed above, all of which fall within the scope of the present invention. within the scope of protection of the invention.

Claims (3)

1. A high-orbit SAR vertical non-uniform vegetation canopy backscatter coefficient simulation method is characterized in that comprising the following steps: Identify the species of vegetation canopy components; Determine the vertical distribution function N _j (h) of the volume density of various vegetation components; Calculate the thin layer backscattering matrix S and forward scattering matrix T in the double matrix algorithm; Calculate the backscatter coefficient of the vegetation canopy; Among them, the calculation formulas of the thin-layer backscattering matrix S and forward scattering matrix T in the double-matrix algorithm are: T(θ _t ,θ _i ,φ _t -φ _i )＝U ^-1 P(θ _t ,θ _i ,φ _t -φ _i )Δz In the formula: θ _i is the incident angle; θ _s is the scattering angle; is the incident azimuth; is the scattering azimuth; U ^-1 is a diagonal matrix, and the elements are the cosines of the scattering directions; P is the bistation scattering phase matrix; h is the position of the vegetation component in the canopy; Δz is the thickness of each thin layer; N is the unit The density of vegetation canopy components in the thin layer volume; n is the number of types of vegetation canopy components, and j represents the jth type of vegetation canopy components. 2. The high-orbit SAR vertical non-uniform vegetation canopy backscatter coefficient simulation method according to claim 1, wherein the type of the vegetation canopy component is a blade, a stem, a branch, a fruit or a flower. 3. the high-orbit SAR vertical non-uniform vegetation canopy backscatter coefficient simulation method according to claim 1, is characterized in that the backscatter coefficient of described calculation vegetation canopy comprises the following steps: Calculate the backscattering matrix S' and forward scattering matrix T' of a thin layer with a thickness of 2Δz composed of two tiny thin layers above and below; Repeat the above calculation to obtain the backscattering matrix S ⁰ of the medium layer with the required thickness; Calculate the backscatter coefficient σ ⁰ ; Among them, the calculation formulas of the backscattering matrix S' and the forward scattering matrix T' are: ${\mathrm{<span\; class="notranslate">\; S</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{<span\; class="notranslate">\; *</span>}{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{<span\; class="notranslate">\; *</span>}{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}$ ${\mathrm{<span\; class="notranslate">\; T</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{<span\; class="notranslate">\; *</span>}{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}$ ${\mathrm{<span\; class="notranslate">\; S</span>}}^{<span\; class="notranslate">\; \&prime;</span><span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{<span\; class="notranslate">\; *</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{<span\; class="notranslate">\; *</span>}$ ${\mathrm{<span\; class="notranslate">\; T</span>}}^{<span\; class="notranslate">\; \&prime;</span><span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{<span\; class="notranslate">\; *</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{<span\; class="notranslate">\; *</span>}$ The formula for calculating the backscattering coefficient is: σ ⁰ =4πS ⁰ In the formula, S1 is the backscattering matrix of the _first layer, T1 is the forward scattering matrix of the _first layer, S2 is the backscattering matrix of the _second layer, and T2 is the forward scattering matrix of the _second layer , * indicates the back and forward scattering matrix when the incident direction is reverse incident.