A parallel and vertical insect discrimination method based on multi-frequency differential eigenvalues

Abstract

This invention discloses a method for distinguishing parallel and vertical insects based on multi-frequency differential eigenvalues. First, based on simulations and microwave anechoic chamber measurements of insect multi-frequency differential eigenvalue characteristics, the relationship between insect multi-frequency differential eigenvalues ​​and polarization patterns, as well as insect category (parallel or vertical), was discovered. Then, based on three conditions—"differential eigenvalue relationship," "polarization pattern type," and "whether the maximum RCS direction is consistent"—the insect categories at 9.5 GHz and 11.5 GHz were classified and discussed in five cases. Finally, based on the above classification and discussion results, a new method for distinguishing parallel and vertical insects is proposed.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of insect radar, and particularly relates to a parallel and vertical insect discrimination method based on multi-frequency differential eigenvalues. BACKGROUND

[0002] Insect radar is an effective tool for studying migratory insects. Based on the measurement results of insect radar, the body length, body weight, body width, wing beat frequency, horizontal speed, orientation, etc. of insects can be inverted. According to the inverted insect parameters, the species of migratory insects can be identified and the trajectory can be predicted, which is of great significance for studying the behavior of migratory insects and preventing migratory pests.

[0003] Insect orientation is one of the most important behavioral parameters of migratory insects, and accurate orientation measurement results are of great significance for accurately predicting the migration route of migratory pests.

[0004] Based on the assumption that the maximum radar cross-section (RCS) of insects appears when the polarization direction is parallel to the body axis, the vertical insect radar can measure the head orientation of individual insects. However, this assumption is only valid for small insects (hereinafter referred to as parallel insects). For some large insects (hereinafter referred to as vertical insects), the maximum RCS of insects appears when the polarization direction is perpendicular to the body axis, and the extracted head orientation will have a 90° error. Correctly distinguishing between parallel and vertical insects can effectively compensate for the 90-degree error in the extraction of the head orientation of vertical insects, thereby achieving high-precision inversion of insect orientation. SUMMARY

[0005] Therefore, the present application provides a parallel and vertical insect discrimination method based on multi-frequency differential eigenvalues, which can distinguish between parallel and vertical insects, solve the problem of 90-degree error in the extraction of the head orientation of vertical insects under the existing orientation inversion method, and improve the accuracy of insect orientation inversion. This helps to accurately predict the migration route of migratory pests.

[0006] A parallel and vertical insect discrimination method based on multi-frequency differential eigenvalues, which is used to distinguish between parallel and vertical insects at 9.5 GHz and 11.5 GHz, includes the following steps:

[0007] Step one, obtain the differential eigenvalue, polarization direction diagram and maximum RCS direction of the 9.5 GHz (hereinafter referred to as X1 frequency point) insect;

[0008] Step two, obtain the differential eigenvalue, polarization direction diagram and maximum RCS direction of the 11.5 GHz (hereinafter referred to as X2 frequency point) insect;

[0009] Step three, classify the insect according to the results obtained in step one and the results obtained in step two, and complete the parallel and vertical insect discrimination based on multi-frequency differential eigenvalues.

[0010] The method for obtaining the differential eigenvalue of the insect in steps one and two is:

[0011] Suppose the polarization scattering matrix of the insect is

[0012]

[0013] where s 11 , s 12 , s 21 and s 22 are the amplitudes of the HH, HV, VH and VV polarization channels respectively, β, β' and γ are the phases of the HV, VH and VV polarization channels respectively, and for a single base station radar, s 12 =s 21 , β=β'.

[0014] Two eigenvalues μ1 and μ2 of the polarization scattering matrix can be calculated from equation (1). Without loss of generality, suppose |μ1|≥|μ2|, then

[0015]

[0016]

[0017] The physical meaning of the two eigenvalues is the RCS when the polarization direction is parallel and perpendicular to the axis of the insect;

[0018] The differential eigenvalue of the insect is defined as the ratio of the large eigenvalue to the small eigenvalue of the insect, that is

[0019]

[0020] The polarization pattern of the insect is represented in steps one and two as:

[0021]

[0022] where h(α) represents the normalized effective antenna length when the polarization direction is α;

[0023] The classification method in step three is:

[0024] The first type is that the maximum RCS direction of the X1 frequency point is inconsistent with that of the X2 frequency point, and the insect is determined to be parallel at the X1 frequency point and perpendicular at the X2 frequency point;

[0025] The second type is that the maximum RCS direction of the X1 frequency point is consistent with that of the X2 frequency point, and the amplitude of the differential eigenvalue of the X2 frequency point is smaller than that of the X1 frequency point, and the insect is determined to be parallel at both the X1 frequency point and the X2 frequency point;

[0026] The third type, the X1 frequency point and the X2 frequency point maximum RCS direction is consistent, the X2 frequency point differential eigenvalue amplitude is greater than the X1 frequency point differential eigenvalue amplitude, the polarization pattern of the insect at the X1 frequency point is "cross type", and the insect is determined to be vertical insect at the X1 frequency point and the X2 frequency point;

[0027] The fourth type, the X1 frequency point and the X2 frequency point maximum RCS direction is consistent, the X2 frequency point differential eigenvalue amplitude is greater than the X1 frequency point differential eigenvalue amplitude, the polarization pattern of the insect at the X1 frequency point and the X2 frequency point is "8 type", if the X1 frequency point differential eigenvalue amplitude is less than 6dB, the insect is determined to be vertical insect at the X1 frequency point and the X2 frequency point, otherwise, if the X1 frequency point differential eigenvalue amplitude is not less than 6dB, the insect is determined to be parallel insect at the X1 frequency point and the X2 frequency point;

[0028] The fifth type, the X1 frequency point and the X2 frequency point maximum RCS direction is consistent, the X2 frequency point differential eigenvalue amplitude is greater than the X1 frequency point differential eigenvalue amplitude, the polarization pattern of the insect at the X1 frequency point is "8 type", and the polarization pattern of the insect at the X2 frequency point is "cross type", and the situation does not exist.

[0029] Beneficial effects

[0030] (1) The application discloses a parallel and vertical insect discrimination method based on multi-frequency differential eigenvalues.

[0031] (2) The application provides an effective means for parallel and vertical insect discrimination, which is helpful to improve the observation ability of insect radar and promote the research on insect migration behavior.

[0032] (3) The application is a parallel and vertical insect discrimination method based on multi-frequency differential eigenvalues, which provides an effective means for parallel and vertical insect discrimination. BRIEF DESCRIPTION OF DRAWINGS

[0033] Fig. 1 is a curve of the differential eigenvalue amplitude of the simulated insects versus frequency: Fig. 1(a) simulates the differential eigenvalue amplitude of the insects with a body length to body width ratio of 2 and a body length of 5 mm, 10 mm, 15 mm, 20 mm and 25 mm versus frequency; Fig. 1(b) is a high-precision result of the multi-frequency differential eigenvalue amplitude of the insect with a body length of 10 mm in Fig. 1(a) versus frequency after frequency interpolation, wherein the Nth maximum value on the curve is defined as the Nth maximum point, and the Nth minimum value is defined as the Nth minimum point;

[0034] Fig. 2 is a polarized pattern of the simulated insects in Fig. 1 at different frequency points: Fig. 2(a) is a change of the polarized pattern of the insects before the first maximum point, Fig. 2(b) is a change of the polarized pattern of the insects between the first maximum point and the first minimum point, and Fig. 2(c) is a change of the polarized pattern of the insects between the first minimum point and the second maximum point;

[0035] Fig. 3 is a curve of the differential eigenvalue amplitude and phase of four insects with different body sizes versus frequency in the darkroom and the polarized pattern of the insects at the X1 and X2 frequency points, wherein the blue hollow circle dotted line represents the differential eigenvalue phase, and the red solid circle dotted line represents the differential eigenvalue amplitude;

[0036] Figure 4 Fig. 4 is a distribution of the first maximum point, the first minimum point, the second maximum point and the second minimum point of 198 insects in the darkroom (not shown beyond 8-18 GHz);

[0037] Figure 5 Fig. 5 is a comparison of the RCS distribution of the target with the dumbbell-shaped polarized pattern at the X1 and X2 frequency points and the increased differential amplitude;

[0038] Figure 6 Fig. 6 is a flowchart of the parallel and vertical insect discrimination method;

[0039] Figure 7 Fig. 7 is a comparison of the correct rate of the parallel and vertical discrimination of 198 insects based on the existing method and the new method under different signal-to-noise ratios. DETAILED DESCRIPTION

[0040] This invention provides a method for distinguishing parallel and vertical insects based on multi-frequency differential eigenvalues. First, based on simulation and microwave anechoic chamber measurements of insect multi-frequency differential eigenvalue characteristics, the relationship between insect multi-frequency differential eigenvalues ​​and polarization patterns, as well as insect category (parallel or vertical insects), is discovered. Then, based on three conditions—"differential eigenvalue relationship," "polarization pattern type," and "whether the maximum RCS direction is consistent"—the insect categories at 9.5 GHz and 11.5 GHz are classified and discussed in five cases. Finally, based on the above classification and discussion results, a new method for distinguishing parallel and vertical insects is proposed.

[0041] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0042] Assume the polarization scattering matrix (PSM) of the insect is as follows:

[0043]

[0044] Among them, s 11 s 12 s 21 and s 22 σH, σV, σH, and σV, respectively, represent the amplitudes of the HH, HV, VH, and VV polarization channels, while β, β′, and γ represent the phases of the HV, VH, and VV polarization channels, respectively. For a monostatic radar, sH... 12 =s 21 , β=β′;

[0045] The two eigenvalues ​​μ1 and μ2 of the polarization scattering matrix can be calculated from equation (1). Without loss of generality, assuming |μ1|≥|μ2|, then we have

[0046]

[0047]

[0048] The physical meaning of these two eigenvalues ​​is the RCS when the polarization direction is parallel and perpendicular to the insect's body axis.

[0049] The difference eigenvalue of an insect is defined as the ratio of its largest eigenvalue to its smallest eigenvalue.

[0050]

[0051] Fig. 1(a) is a multi-frequency differential eigenvalue curve of different body length insects with a body length-width ratio of 2, the density of the simulated insects is consistent, the larger the body length, the larger the surface insect size, it can be found from Fig. 1(a) that the differential eigenvalue amplitude curve exists periodic fluctuation, all increase first and then decrease, the maximum value point and the minimum value point appear alternately, the larger the insect size, the lower the frequency point of the first maximum value point and the first minimum value point. In order to further analyze the characteristics of the multi-frequency differential eigenvalue of the insect, the multi-frequency differential eigenvalue curve of the insect with a body length of 10 mm is interpolated and redrawn, that is, as shown in Fig. 1(b).

[0052] Fig. 2 shows the polarization pattern of the insect at different frequency points in Fig. 1(b). The radar echo intensity of the insect at different polarization directions (i.e. the polarization pattern) can be expressed as

[0053]

[0054] wherein, wherein h(α) represents the normalized effective antenna length when the polarization direction is α;

[0055] The polarization pattern reflects the size of the RCS of the insect at different polarization directions, based on the assumption that the maximum RCS of the insect appears at the polarization direction parallel to the body axis, the direction of the maximum RCS of the insect polarization pattern can be used to determine the head direction of the insect, but this assumption is only valid for parallel insects, that is, the maximum RCS of the parallel and vertical insect appears at the polarization direction parallel to the body axis, and for vertical insects, the assumption is changed to "the maximum RCS of the vertical insect appears at the polarization direction perpendicular to the body axis". The current orientation inversion algorithm determines the orientation by extracting the direction of the maximum RCS, therefore, accurate discrimination of parallel and vertical insects is necessary to solve the 90-degree error in the extraction of the orientation of vertical insects.

[0056] In simulation, the real orientation of the insect is 0 degrees (or 180 degrees, the head direction of the insect has a 180-degree ambiguity), that is, if the insect is a parallel insect, the direction of its maximum RCS should appear at 0 degrees (or 180 degrees, only 0 degrees is used instead of the real head direction of the insect), and for a vertical insect, the direction of its maximum RCS appears at 90 degrees (or 270 degrees, 90 degrees represents the vertical axis direction of the insect). If the maximum value of the polarization pattern of the insect at a certain frequency point appears at 0 degrees, the insect is a parallel insect, otherwise, if the maximum value of the polarization pattern of the insect at a certain frequency point appears at 90 degrees, the insect is a vertical insect. That is, for a specific insect, its parallel and vertical insect properties at different frequencies are not constant.

[0057] Next, see the change of the polarized pattern of the insect at different frequencies in Fig. 1(b). Fig. 2(a) is the polarized pattern of the first four frequencies (3GHz, 4GHz, 5GHz, 6GHz) before the first maximum point in Fig. 1(b). The polarized patterns of the four frequencies are all "8-shaped", and the maximum RCS direction appears at 0 degree, that is, the insect is parallel insect at this time, and as the frequency increases, the differential eigenvalue amplitude increases continuously; Fig. 2(b) is the polarized pattern of the four frequencies (11GHz, 13GHz, 15GHz, 17GHz) between the first maximum point and the first minimum point. As the frequency increases, the polarized pattern changes from "8-shaped" to "cross-shaped", and the differential eigenvalue amplitude decreases continuously, but the maximum RCS direction is still at 0 degree, and the insect is still parallel insect at this stage; Fig. 2(c) is the polarized pattern of the four frequencies (18GHz, 20GHz, 21GHz, 23GHz) between the first minimum point and the first maximum point. As the frequency increases, the differential eigenvalue gradually increases, and the polarized pattern gradually changes from "cross-shaped" to "8-shaped", and the maximum RCS direction appears at 90 degrees, that is, the insect is vertical insect. That is, at the minimum point of the multi-frequency differential eigenvalue curve, the "parallel" and "vertical" properties of the insect will change. According to the characteristics of the multi-frequency differential eigenvalue of the insect and the change of the polarized pattern of the insect, we can distinguish the parallel and vertical insects.

[0058] Fig. 3 shows the multi-frequency differential characteristic amplitude and phase curves of four different sized insects in the range of 8-18 GHz and the polarization patterns at the X1 and X2 frequencies in the darkroom measurement. In the darkroom measurement, the real head of the insect is oriented at 0 degrees, which is consistent with the simulation. It can be seen that the multi-frequency differential characteristic amplitude of the four insects varies similarly to the simulation results. Fig. 3(a) is for a 114.9 mg insect, the first maximum point is between 9-10 GHz, the first minimum point is between 16-17 GHz, the X1 and X2 frequencies are between the first maximum point and the first minimum point, the differential characteristic amplitude decreases, the polarization patterns at the X1 and X2 frequencies are both "8-shaped", the maximum RCS direction is at 0 degrees, and at the X1 and X2 frequencies, the insect is parallel; Fig. 3(b) is for a 160.5 mg insect, which is larger than the first one, the first maximum point is around 8 GHz, the first minimum point is between 15-16 GHz, the first maximum point and the first minimum point are smaller than the first insect, the X1 and X2 frequencies are between the first maximum point and the second minimum point, the differential characteristic amplitude decreases, the polarization pattern at the X1 frequency is "8-shaped", and at the X2 frequency, it is "cross-shaped", the maximum RCS direction is at 0 degrees, and at the X1 and X2 frequencies, the insect is parallel; Fig. 3(c) is for a 935.3 mg large-sized insect, the first maximum point and the first minimum point are both smaller than 8 GHz, the X1 and X2 frequencies are between the first minimum point and the second maximum point, the differential characteristic amplitude increases, the polarization patterns are both "cross-shaped", the maximum RCS direction is at 90 degrees, and at the X1 and X2 frequencies, the insect is perpendicular; Fig. 3(d) is for a 842 mg large-sized insect, the first maximum point is smaller than 8 GHz, the second minimum point is between 10-11 GHz, the X1 frequency is between the first maximum point and the first minimum point, the X2 frequency is between the first minimum point and the second maximum point, the polarization patterns at the X1 and X2 frequencies are both "cross-shaped", at the X1 frequency, the maximum RCS direction is at 0 degrees, and the insect is parallel, at the X2 frequency, the maximum RCS direction is at 90 degrees, and the insect is perpendicular.

[0059] According to the above analysis results, the insect categories at the X1 and X2 frequencies in the five cases can be classified and discussed according to the three conditions of the "differential characteristic amplitude relationship", "polarization pattern type", and "whether the maximum RCS direction is consistent" of the insects at the X1 and X2 frequencies. The classification and discussion are shown in the following table:

[0060]

[0061] Next, each case and the discrimination result will be described.

[0062] The first case: the maximum RCS directions at the X1 and X2 frequencies are not consistent, which indicates that there is a minimum point between the X1 and X2 frequencies, according to the "differential characteristic amplitude relationship", the polarization pattern type, and whether the maximum RCS direction is consistent, the insect at the X1 and X2 frequencies can be classified into three categories as follows: Figure 4The statistics of the first maximum point, the first minimum point, the second maximum point and the second minimum point of the insects in the darkroom 198 found that the frequency of the second minimum point of all insects was in Ku band (more than 12 GHz), that is, if there is a minimum point between the X1 and X2 frequency points, it must be a minimum point, and before the minimum point, the insects are parallel insects, and after the minimum point, the insects are vertical insects, so in this case, it can be determined that the insects at the X1 frequency point are parallel insects, and the insects at the X2 frequency point are vertical insects.

[0063] The second case: the maximum RCS directions of the X1 and X2 bands are consistent, and the differential eigenvalue amplitude of X2 is smaller than that of X1. The consistent maximum RCS directions of the X1 and X2 bands indicate that the X1 and X2 frequency points are on the same side of the first minimum point. Under this premise, if both X1 and X2 are greater than the first minimum frequency, there can be three cases for the positions of X1 and X2: ① both X1 and X2 are located on the left side of the second maximum point; ② both X1 and X2 are located on the right side of the second maximum point; ③ X1 and X2 are located on both sides of the second maximum point. For case ①, the differential eigenvalue amplitude of X2 is greater than that of X1, which contradicts the assumption premise that the differential eigenvalue amplitude of X2 is smaller than that of X1, so this possibility is excluded; for case ②, according to Figure 4 , the smallest second maximum point of the insects in the darkroom 198 is located at more than 11 GHz, that is, the X1 frequency point must be located on the left side of the second maximum, so this possibility is excluded; for case ③, according to Figure 4 , the smallest second maximum point is located between 11-11.5 GHz, which means that the X2 frequency point is very close to the second maximum point, in other words, the differential eigenvalue amplitude of the X2 frequency point is very close to the second maximum of the differential eigenvalue amplitude, while the frequency interval between the X1 frequency point and the second minimum point exceeds 1.5 GHz, that is, the differential eigenvalue amplitude of the X1 frequency point must be smaller than that of the X2 frequency point, which contradicts the assumption premise that the differential eigenvalue amplitude of X2 is smaller than that of X1, so this possibility is excluded. Thus, all possibilities of the X1 and X2 bands being located on the right side of the first minimum point are excluded, so in this case, the X1 and X2 frequency points must be located on the left side of the first minimum point, so the insects at the X1 and X2 frequency points are parallel insects.

[0064] The third case: the maximum RCS directions of the X1 and X2 frequencies are consistent, the differential eigenvalue amplitude of the X2 frequency is greater than that of the X1 frequency, and the polarization pattern of the X1 frequency is "cross-shaped". The fact that the maximum RCS directions of the X1 and X2 frequencies are consistent indicates that the X1 and X2 frequencies are on the same side of the first minimum value, and the "cross-shaped" polarization pattern of the X1 frequency indicates that the X1 frequency is close to the first minimum value point. The differential eigenvalue amplitude of the X2 frequency is greater than that of the X1 frequency, which eliminates the possibility that the X1 and X2 frequencies are on the left side of the first minimum value. Therefore, the X1 and X2 frequencies must be on the right side of the first minimum value, and thus the X1 and X2 frequencies are both vertical insects.

[0065] The fourth case: the maximum RCS directions of the X1 and X2 frequencies are consistent, the differential eigenvalue amplitude of the X2 frequency is greater than that of the X1 frequency, and the polarization patterns of the X1 and X2 frequencies are both "8-shaped". In this case, the X1 and X2 frequencies are either near the first maximum value or near the second maximum value. If they are near the first maximum value, the X1 and X2 frequencies are both parallel insects, and if they are near the second maximum value, the X1 and X2 frequencies are both vertical insects. Figure 5 The X1 frequency RCS and the X1 frequency differential eigenvalue amplitude scatter plot of the insects meeting the fourth case are given, where the blue solid circles represent parallel insects and the red hollow circles represent vertical insects. It can be found that although the two types of insects meet the fourth condition, they are completely distinguishable in the two-dimensional plane of RCS and differential eigenvalue amplitude. Here, we use the X1 frequency differential eigenvalue amplitude as a threshold to distinguish the two types of insects. In the fourth case, if the X1 frequency differential eigenvalue amplitude is greater than 6 dB, the insects are considered to be parallel insects at the X1 and X2 frequencies, and if the X1 frequency differential eigenvalue amplitude is less than 6 dB, the insects are considered to be vertical insects at the X1 and X2 frequencies.

[0066] The fifth case: if the X1 frequency pattern is "8-shaped" and the X2 frequency pattern is "cross-shaped", and the maximum RCS directions are consistent, it means that the X1 and X2 frequencies are between the first maximum value and the second minimum value, and the differential eigenvalue amplitude of the X2 frequency must be less than that of the X1 frequency, which contradicts the assumption that the differential eigenvalue amplitude of the X2 frequency is greater than that of the X1 frequency. Therefore, the fifth case does not exist.

[0067] According to the above analysis results, the results in the table are correct and complete.

[0068] Based on the above analysis, the parallel and vertical insect discrimination method based on the differential eigenvalues of the X1 and X2 frequencies can be represented as the flowchart shown in FIG. Figure 6

[0069] ​In order to verify the parallel and vertical insect discrimination method described above, based on the microwave darkroom measurement data of 198 insects, the proposed parallel and vertical insect discrimination method is verified, and compared with the existing parallel and vertical insect discrimination method (a parallel and vertical insect discrimination method based on feature phase, patent number: ZL201911203473.0). The steps are as follows:

[0070] Step one, set a series of SNR (13dB, 14dB, …, 50dB), take the measured PSM of 198 insects as the true value, take the power of the first element in PSM (i.e. ) as the signal power, according to the set SNR, respectively simulate four complex Gaussian white noises added to the four values of PSM.

[0071] Step two, respectively adopt the parallel and vertical insect discrimination method based on multi-frequency band difference feature value described in the present application and the existing parallel and vertical insect discrimination method (a parallel and vertical insect discrimination method based on feature phase, patent number: ZL201911203473.0) to discriminate insects, and calculate the parallel and vertical insect discrimination accuracy.

[0072] Step three, repeat steps one to two 500 times, and calculate the average of the parallel and vertical insect discrimination accuracy of the two methods.

[0073] The comparison results are shown in Figure 7 , the solid circle dash line is the discrimination accuracy of the new method, and the hollow circle dash line is the discrimination accuracy of the existing method. It can be seen that when the signal-to-noise ratio is lower than 33dB, the discrimination accuracy of the new method is always higher than that of the existing method, when the signal-to-noise ratio is higher than 33dB, the performance of the two methods is equivalent, and both achieve 100% accuracy. Therefore, the parallel and vertical insect discrimination method based on multi-frequency difference feature value described in the present application can realize parallel and vertical insect discrimination, and the performance is better than that of the existing method.

[0074] In summary, the above is only an embodiment of the present application based on 198 insects data, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0075] In summary, the above is only a preferred embodiment of the present application, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A parallel and perpendicular insect discrimination method based on multi-frequency differential eigenvalues, characterized in that: The method refers to distinguishing 9.5GHz and 11.5GHz parallel and vertical insects, comprising the following steps: Step one, obtaining the differential characteristic value, polarization pattern and maximum RCS direction of 9.5GHz insects; Step two, obtaining the differential characteristic value, polarization pattern and maximum RCS direction of 11.5GHz insects; Step three, classifying insects according to the results obtained in step one and the results obtained in step two, completing the parallel and vertical insect discrimination based on multi-frequency differential characteristic value; In step three, 9.5GHz is called X1 frequency point, and 11.5GHz is called X2 frequency point, and the classification method is: The maximum RCS directions of X1 frequency point and X2 frequency point are inconsistent, and the insects are judged as parallel insects at X1 frequency point and vertical insects at X2 frequency point; The maximum RCS directions of X1 frequency point and X2 frequency point are consistent, and the differential characteristic value amplitude of X2 frequency point is smaller than that of X1 frequency point, and the insects are judged as parallel insects at X1 frequency point and X2 frequency point; The maximum RCS directions of X1 frequency point and X2 frequency point are consistent, and the differential characteristic value amplitude of X2 frequency point is greater than that of X1 frequency point, and the polarization pattern of insects at X1 frequency point is "cross type", and the insects are judged as vertical insects at X1 frequency point and X2 frequency point; The maximum RCS directions of X1 frequency point and X2 frequency point are consistent, and the differential characteristic value amplitude of X2 frequency point is greater than that of X1 frequency point, and the polarization pattern of insects at X1 frequency point and X2 frequency point is "8 type", and if the differential characteristic value amplitude of X1 frequency point is less than 6dB, the insects are judged as vertical insects at X1 frequency point and X2 frequency point, otherwise, if the differential characteristic value amplitude of X1 frequency point is not less than 6dB, the insects are judged as parallel insects at X1 frequency point and X2 frequency point; The maximum RCS directions of X1 frequency point and X2 frequency point are consistent, and the differential characteristic value amplitude of X2 frequency point is greater than that of X1 frequency point, and the polarization pattern of insects at X1 frequency point is "8 type", and the polarization pattern of insects at X2 frequency point is "cross type", which is a contradictory prerequisite condition, and the situation does not exist.

2. The parallel and vertical insect discrimination method based on multi-frequency differential characteristic value according to claim 1, characterized in that: In steps one and two, the method for obtaining the differential characteristic value of the insect is: Assuming that the polarization scattering matrix of the insect is （1） wherein , , and are the amplitudes of the HH, HV, VH and VV polarized channels, respectively, , , are the phases of the HV, VH and VV polarized channels, respectively, for monostatic radar, , ; The two eigenvalues of the polarized scattering matrix are calculated from (1) and , assuming then （2） （3） The differential characteristic value of the insect is （4）。 3. The parallel and vertical insect discrimination method based on multi-frequency differential characteristic value according to claim 2, characterized in that: In steps one and two, the polarization pattern of the insect is represented as: （5） wherein, wherein represents the normalized effective antenna length when the polarization direction is represents the normalized effective antenna length when the polarization direction is

Images