InSAR (Interferometric Synthetic Aperture Radar) submillimeter-level high-precision deformation quantity detection method based on initial phase error elimination

Abstract

The invention relates to an InSAR (Interferometric Synthetic Aperture Radar) submillimeter-level high-precision deformation quantity detection method based on initial phase error elimination, and belongs to the field of electronic signal processing. Firstly, millimeter wave radar equipment is used for monitoring and scanning a target area and realizing SAR imaging, and main and auxiliary images are divided according to a radar scanning period; a PS point is selected, and the deformation quantity of the PS point is used for representing the deformation state of the whole target area; then taking the distance from the PS point to the original point of the image as a reference, combining with the radar instantaneous coordinate corresponding to each pulse signal, extracting phase information of the corresponding PS point from all the pulse signals subjected to distance compression, and calculating a phase mean value to eliminate an initial phase error of a transmitted signal; and finally, for the same PS point, calculating a phase difference between two monitoring processing results to obtain a real deformation quantity. According to the method, the error generated by the random fluctuation of the initial phase of the transmitted signal is reduced, the traditional calculation method is simplified, and the calculation efficiency and the calculation precision are both improved.

Description

Technical Field

[0001] The present invention belongs to the field of electronic signal processing, and relates to a method for detecting InSAR sub-millimeter high-precision deformation quantity based on initial phase error elimination. Background Art

[0002] In the modern urbanization process, the stability of buildings and the ground surface has become increasingly important. Monitoring and warning the deformation of the ground surface and buildings can effectively prevent and reduce the risks brought by natural disasters or human factors. The Ground-based Interferometric Synthetic Aperture Radar (GB-InSAR) technology is the most common method for detecting deformation quantity at present, which has the advantages of small atmospheric delay error and noise error, and no various errors caused by baseline decoherence. However, this method requires image registration, phase filtering, and phase unwrapping steps, which are cumbersome and have a large amount of computation.

[0003] The SAR image is a complex image, and each pixel point is a complex number. Its phase information can be used to obtain the elevation or deformation information of the target area. The principle of InSAR deformation quantity detection is that after obtaining the differential phase of the two transmitted signals, there is a "half-wavelength" relationship between the differential phase and the deformation quantity, that is, a whole-cycle fringe corresponds to the deformation of half the SAR signal wavelength in the radar signal propagation direction. However, in the actual test process, there is a problem of random fluctuation of the initial phase of the linear frequency modulation (LFM) signal emitted by the radar, which has a great impact on the calculation of the deformation quantity. According to the deformation quantity calculation formula, assuming that the wavelength of the test transmitted signal is 4.8 mm, when the initial phase difference between the two transmitted LFM signals is π, the calculated deformation quantity will have an error of 1.2 mm, reducing the deformation detection accuracy.

[0004] Aiming at the existing GB-InSAR deformation detection method and the problem of random fluctuation of the initial phase of the transmitted signal in the actual test, a method for detecting InSAR high-precision deformation quantity based on initial phase error elimination is proposed. This method arranges radar equipment in front of the target area, continuously emits LFM signals, repeats the measurement and images separately; then selects the main and auxiliary images and PS points, and directly processes the phase of the LFM signal at the echo signal level through the distance mapping of the selected PS points, reducing the calculation error caused by the random fluctuation of the initial phase of the transmitted signal, and realizing the deformation quantity detection with high resolution. At the same time, compared with the common traditional GB-InSAR deformation quantity detection technology, it omits the steps with large amounts of computation such as image registration, phase filtering, and phase unwrapping, improves the detection efficiency, and is more suitable for real-time monitoring. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a method for detecting sub-millimeter high-precision deformation quantity based on initial phase error elimination in InSAR, which reduces the deformation calculation error caused by the random fluctuation of the initial phase of the LFM signal transmitted by the radar in multiple repeated tests, and at the same time reduces the amount of computation.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The method for detecting high-precision deformation quantity in InSAR based on initial phase error elimination includes the following steps:

[0008] Step 1: System parameter setting. Install and fix the ground-based radar equipment in front of the target area to ensure that the angle of the radar transmitted signal is aligned with the target area. Conduct continuous tests on the target area. During the test, it is necessary to ensure that the equipment does not move, and the movement trajectory, running speed, and movement direction of the radar should all remain consistent. The test schematic diagram is as Figure 1 shown.

[0009] Assume a target point P. At any slow time moment t m The slant range from the antenna phase center to point P is R(t m ), and the complex envelope of the radar transmitted signal is γ represents the frequency modulation slope of the LFM signal, represents the fast time in the range direction. Then the mathematical model of all received echo signals can be expressed as:

[0010]

[0011] In the formula, σ(x,y) is the point target scattering coefficient, a r (·) and a a (·) are the window function and azimuth window function of the LFM signal.

[0012] Step 2: SAR imaging. Perform SAR imaging on the data obtained from the tests respectively. Select the SAR image obtained from the first test as the main image, and one SAR image at other times as the auxiliary image.

[0013] Step 3: PS point identification. Find 3 PS points with stable phase information and high coherence. PS points have the characteristic of maintaining stable scattering characteristics in the time series images for a long time. Therefore, the deformation quantities of these three PS points are used to represent the deformation quantity of the entire target area.

[0014] Common PS point recognition methods include the amplitude deviation index threshold method, the coherence coefficient threshold method, the intensity threshold method, etc. The coherence coefficient threshold method uses a sliding window to estimate the overall coherence. The size of the window will affect the accuracy of the selection result. If the window is set too large, it will increase the calculation time and cause some individually existing points to not be selected. If the window is set too small, the credibility of the selection result is not high. On high signal-to-noise ratio resolution cells, the amplitude deviation index can effectively reflect the phase noise information in the time series image, overcoming the limitation that the coherence coefficient cannot measure the phase noise level. The intensity threshold method directly uses the pixel intensity information as an index for selecting PS points, with simple calculation and fast operation speed.

[0015] The method of selecting PS points individually only considers the scattering characteristics or coherence of pixel points. When setting the threshold, if the threshold is set too low, it may lead to over-screening and failure to select PS points in the deformation area. If the threshold is set too high, it will lead to an increase in candidate PS points and a decrease in the reliability of the selected PS points. In practical applications, the intensity, noise phase, and coherence of pixel points are usually considered comprehensively, and multiple methods are combined to improve the accuracy and reliability of PS point recognition. The present invention uses the above three methods to jointly select PS points. Denote the thresholds of the amplitude deviation index, coherence coefficient, and intensity mean as th 1 、th 2 、th 3 respectively. Then the pixel points that simultaneously satisfy D A <th 1 、γ>th 2 、 are candidate PS points.

[0016] When using a radar device for testing, the initial phase of the transmitted signal will undergo random fluctuations, resulting in a certain error between the phase of each group of data and the true phase. The essence of InSAR deformation amount calculation is to solve the deformation amount based on the interference phase of two groups of data. Therefore, the error generated by the initial phase has a greater impact on the calculation accuracy and cannot be ignored. In step four, after performing motion phase compensation on the echo signal, the range coordinate of the PS point is mapped to the phase index of the echo signal, and the corresponding position phase is screened out. Step five is to calculate the phase mean to reduce or eliminate the impact of the initial phase error on the interference phase.

[0017] Step four, motion phase compensation and range mapping. When performing InSAR testing, the radar moves uniformly along the orbit each time. Denote the distance between the radar at the center position of the orbit and the target as R 0 and the distance between the radar and the target corresponding to the slow time t as R(t). Then the echo signal is as shown in equation (2). First, compensate the phase -j4πf 0 ΔR(t) / c generated by the radar motion.

[0018] S = Aexp{-j4πf 0 R 0 / c}·exp{-j4πf 0 ΔR(t) / c} (2)

[0019] Then, according to the range bin coordinates corresponding to the selected PS points, the echo signal phases corresponding to the range bin coordinates in each compensated LFM signal are extracted. Denote the maximum range d that the selected radar can image, the range bin coordinate of the selected PS point as r, and the number of sampling points as n. Then, the index m of the corresponding LFM signal phase can be calculated according to Equation (3), and the signal phase corresponding to the m-th sampling point in all LFM signals is taken.

[0020]

[0021] Step Five: Initial phase error elimination. The phases of the obtained LFM signals are averaged, and the difference between the average phases of the two test data is used as the interferometric phase, thereby reducing the phase error caused by the random fluctuation of the initial phase of the radar transmitted signal and improving the detection accuracy.

[0022] When taking the phase mean value, the truncated mean method is adopted, with the FFT amplitude as the truncation criterion. The phases of the obtained LFM signals are sorted according to the corresponding FFT amplitude values, and with a truncation ratio of 5%, the first 5% and the last 5% of the LFM signal phases are deleted, and then the remaining LFM signal phases are averaged. Compared with the ordinary arithmetic mean, the truncated mean can obtain a more robust measure of central tendency. Using the truncated mean in signal processing can achieve a simple filtering effect, removing noise and smoothing the signal while retaining the important features of the signal.

[0023] Step Six: Deformation calculation. After obtaining the interferometric phase, calculate the deformation of the PS points. The principle of InSAR for calculating deformation is as Figure 2 shown. When measuring, the position of the radar device and the azimuthal orbit remain unchanged, that is, the spatial baseline of the two sets of data obtained is zero. The InSAR deformation calculation formula is as follows.

[0024]

[0025] Among them, c represents the speed of light, f c represents the carrier frequency, represents the interferometric phase. The deformation ΔP Los calculated by this method is only the deformation in the signal propagation direction, and the true deformation still needs to be coordinate-transformed according to the actual terrain.

[0026] Step 7, Deformation Warning. Compare the calculated deformation amount with the pre-set maximum allowable deformation amount threshold. If the calculation result is greater than the threshold, send an alarm to the upper-level supervision system to remind the relevant responsible personnel to take necessary protective measures in a timely manner to prevent possible natural or man-made disasters and reduce potential economic and property losses.

[0027] The beneficial effects of the present invention are as follows:

[0028] (1) The present invention proposes an InSAR high-precision deformation amount detection method based on initial phase error elimination, which has broad application prospects in the fields of geological terrain landslides, deformation monitoring of large buildings, etc., and the deformation amount detection accuracy can reach the sub-millimeter level.

[0029] (2) The InSAR deformation detection method based on initial phase error elimination proposed by the present invention takes into account the error problem caused by the random fluctuation of the initial phase of the radar transmission signal in actual tests, and effectively reduces the initial phase error by mapping the range coordinate of the PS point and taking the phase mean value of the corresponding echo signal, thereby reducing the calculation error caused by the random fluctuation of the initial phase of the transmission signal and improving the detection accuracy.

[0030] (3) Compared with the traditional GB-InSAR technology, the InSAR deformation calculation method based on initial phase error elimination proposed by the present invention omits steps such as phase filtering and phase unwrapping of the traditional algorithm, and the calculation efficiency is greatly improved, making it more suitable for real-time monitoring.

[0031] Other advantages, objectives and features of the present invention will be described to some extent in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the following specification. Description of the Drawings

[0032] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be described in detail preferably with reference to the accompanying drawings, where:

[0033] Figure 1 is a schematic diagram of the GB-InSAR test geometric structure;

[0034] Figure 2 is a schematic diagram of the InSAR deformation amount detection principle;

[0035] Figure 3 is a schematic diagram of the PS point distance mapping;

[0036] Figure 4 is a flowchart of the InSAR high-precision deformation amount detection method based on initial phase error elimination. Detailed implementation manners

[0037] The following specific examples are used to illustrate the implementation manners of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the drawings provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0038] Among them, the drawings are only for illustrative purposes, showing only schematic diagrams, not physical diagrams, and should not be construed as a limitation on the present invention; in order to better illustrate the embodiments of the present invention, some components in the drawings will be omitted, enlarged or reduced, which do not represent the dimensions of the actual product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0039] In the drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", "front", "rear", etc. indicating the orientation or positional relationship, they are based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, the terms describing the positional relationship in the drawings are only for illustrative purposes and should not be construed as a limitation on the present invention. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific circumstances.

[0040] The InSAR sub-millimeter-level high-precision deformation detection method based on initial phase error elimination includes the following steps:

[0041] The first step: System parameter setting. A radar device is arranged in front of the detection area. The height of the radar arrangement is the same as that of the target area, and the straight-line distance between the target and the device should be based on the parameters set by the radar. The present invention uses a millimeter-wave signal with a signal frequency of 60 - 70 GHz as the transmitted signal and performs frequency linear modulation to obtain a chirp signal. Taking a swing-arm scanning radar as an example, the sampling rate is set to f s , the frequency modulation slope is γ, and the number of sampling points is n. Then, according to Equation (5), the maximum measurable distance d of the device can be calculated. When arranging the device, it should be ensured that the distance between the device and the area to be measured is less than this maximum distance.

[0042]

[0043] When repeating the test, the carrier frequency f of the radarc parameters such as the frequency modulation slope γ and the sampling rate f s remain unchanged. It is also necessary to ensure that the device does not move or deform during the repeated tests. Taking the swing-arm scanning radar as an example, it is necessary to ensure that the rotation speed and rotation angle of the radar rotation axis are consistent during the test, thereby reducing or eliminating the errors caused by the device itself or human factors.

[0044] Step 2: SAR imaging. Perform SAR imaging on the data obtained from the tests respectively, select the SAR image obtained from the first test as the main image, and one SAR image at other times as the auxiliary image.

[0045] Step 3: Selection of PS points. Select three PS points to represent the deformation state of the entire target area. The method of selecting PS points alone only considers the scattering characteristics or coherence of pixel points. In practical applications, multiple methods are usually combined to improve the accuracy and reliability of PS point recognition. The present invention uses three methods, namely the amplitude deviation index threshold method, the coherence coefficient threshold method, and the intensity threshold method, to jointly select PS points. The amplitude deviation index D A the coherence coefficient γ, and the intensity mean The calculation formulas are as follows:

[0046]

[0047]

[0048] In the formula, σ A and m A represent the amplitude variance and the amplitude mean respectively. M(i, j) and S(i, j) represent two SAR images, and M represents the number of SAR images. Denote the thresholds of the amplitude deviation index, the coherence coefficient, and the intensity mean as th 1 th 2 th 3 , then the points that simultaneously satisfy D A <th 1 , γ>th 2 , are candidate PS points.

[0049] Among them, when calculating the amplitude deviation index and the intensity mean, radiometric calibration needs to be performed first to normalize the amplitudes of all SAR images to the same scale. The main steps of radiometric calibration are to calculate the amplitude mean I m of a single SAR image respectively, calculate the mean of the temporal amplitudes calculate the correction factor K m =a / I m of each SAR image respectively, and correct the amplitudes A' m =A m ×K m of all SAR images respectively.

[0050] Step 4: In actual tests, the initial phase of the radar transmitted signal will randomly fluctuate, affecting the calculation of the interference phase and reducing the accuracy of deformation quantity calculation. After compensating for the motion phase in Step 4, map the range coordinate of the PS point to the phase index of the LFM signal, and screen out the phase at the corresponding position. The specific steps are as follows:

[0051] Step 4 (1): Motion phase compensation. It has been verified that when the radar is stationary, the deformation quantity can be correctly calculated, and the maximum error from the theoretical deformation quantity is about 0.2 mm, with the calculation accuracy reaching the sub-millimeter level. When the radar is tested in a stationary state, the straight-line distance between the radar and the target is denoted as R 0 , and the output signal after range compression of the echo signal is:

[0052] S 0 =Aexp{-j4πf 0 R 0 / c} (9)

[0053] where A represents the amplitude and f 0 represents the carrier frequency. When the radar moves along a certain orbit, for example, when using a swing-arm SAR, denote the distance between the radar and the target at the center position of the orbit as R 0 , and the distance between the radar and the target corresponding to the slow time t is R(t). Then the echo signal is:

[0054]

[0055] where, ΔR(t)=R(t)-R 0 , and the phase generated due to radar motion is -j4πf 0 ΔR(t) / c. Taking the calculation of the deformation quantity of a PS point as an example, calculate the motion phase generated by the radar relative to this PS point when moving along the orbit according to Equation (10), and perform motion phase compensation on all LFM signals.

[0056] Step 4 (2): Range mapping. Determine the range coordinate corresponding to this PS point, find the echo signal corresponding to this range coordinate among all LFM signals, and calculate the phase of this signal. As Figure 3 shown, denote the coordinates of the selected PS point as (x, r), that is, the range coordinate is r, the number of sampling points is n, and the maximum range of the selected radar device for imaging is d. Then the LFM signal corresponding position index m can be calculated according to the following formula, and take the signal phase corresponding to the mth sampling point of all LFM signals.

[0057]

[0058] Step 5: Initial phase error elimination. The phases of all the selected LFM signals are averaged using the method of truncated mean. The FFT amplitude at the corresponding position of the LFM signal is used as the standard for truncated mean, and the signals are sorted according to the amplitude size. The truncation ratio is set to 5%. The 5% data with the largest and smallest FFT amplitudes are removed respectively, and the LFM signals with relatively stable intensity are selected. Finally, the phases of the remaining LFM signals are averaged again. The difference between the average phases obtained from the two measurements is used as the interference phase of this PS point.

[0059] ph 1 = mean{p 1 (5%·l + 1:l - 5%·l, 1)} (12)

[0060] In the formula, p 1 represents a one-dimensional array of the signal phases after the fourth-step screening and sorted according to the FFT amplitude size. l represents the number of selected signal phases, and mean{·} represents taking the arithmetic mean. ph 1 represents the mean of the signal phases of a PS point in the main image calculated. Similarly, the mean of the signal phases of the same PS point in the auxiliary image, ph 2 , ph 1 - ph 2 is the interference phase corresponding to this PS point.

[0061] Step 6: Deformation calculation. After obtaining the interference phase, calculate the deformation of the PS point. The formula for calculating the InSAR deformation is as follows:

[0062]

[0063] Among them, c represents the speed of light, f c represents the carrier frequency, represents the interference phase. The deformation calculated by this method is only the deformation in the signal propagation direction. If the actual deformation is required, terrain coordinate conversion needs to be carried out, and the actual deformation is calculated according to the terrain geometric relationship.

[0064] Step 7: Deformation warning. After calculating the deformation of the PS point, compare it with the allowable maximum deformation threshold set by the system. If the calculated deformation is greater than the threshold, an alarm is sent to the upper-layer management system to remind the relevant staff that deformation has occurred in this area.

[0065] Figure 4 is the flow chart of the InSAR high-precision deformation detection method based on initial phase error elimination.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the present technical solution, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An InSAR sub-millimeter-level high-precision deformation detection method based on initial phase error elimination, characterized in that: The following steps are involved: Use millimeter wave radar equipment to monitor and scan the target area and obtain two SAR images respectively; Select at least one PS point and use the deformation value of the PS point to represent the deformation state of the entire target area; Map the distance from the PS point to the image origin into the phase index of the echo signal, and filter out the phase at the corresponding position; The screened phases are averaged, and the difference between the average phases of the two test data is taken as the interference phase; The deformation amount of the PS point is calculated based on the interference phase.

2. The InSAR submillimeter-level high-precision deformation detection method based on initial phase error elimination according to claim 1 is characterized in that: The millimeter wave radar device uses a linear frequency modulation signal for transmission.

3. The InSAR submillimeter-level high-precision deformation detection method based on initial phase error elimination according to claim 1 is characterized in that: The PS points are selected by combining the amplitude deviation index threshold method, the coherence coefficient threshold method and the intensity threshold method.

4. The InSAR submillimeter-level high-precision deformation detection method based on initial phase error elimination according to claim 1 is characterized in that: The averaging process adopts a truncated mean method, with the FFT amplitude as the truncation standard.

5. The InSAR submillimeter-level high-precision deformation detection method based on initial phase error elimination according to claim 1 is characterized in that: The calculation formula of the deformation variable is: Where c is the speed of light, f c Indicates the carrier frequency, represents the interference phase; The calculated value is the deformation variable ΔP in the direction of signal propagation Los , the real shape needs to be transformed according to the actual terrain.

6. The InSAR submillimeter-level high-precision deformation detection method based on initial phase error elimination according to claim 1 is characterized in that: The detection method further comprises the following steps: After the deformation amount is obtained, it is compared with a preset maximum allowable deformation amount threshold; If the deformation amount is greater than the threshold, an alarm is issued.

7. An InSAR sub-millimeter-level high-precision deformation detection system based on initial phase error elimination, characterized in that: include: Millimeter wave radar equipment is used to monitor and scan the target area and obtain two SAR images respectively; A PS point selection module is used to select at least one PS point from the SAR image; The phase mapping module is used to map the distance from the PS point to the image origin into the phase index of the echo signal and filter out the phase at the corresponding position; The phase averaging module is used to average the screened phases and take the difference between the average phases of two test data as the interference phase; A deformation variable calculation module is used to calculate the deformation variable of the PS point according to the interference phase; The millimeter wave radar equipment is responsible for monitoring and scanning the target area and generating SAR image data; The PS point selection module receives the SAR image data from the millimeter wave radar equipment, and uses the amplitude deviation index threshold method, the coherence coefficient threshold method and the intensity threshold method to identify and select the PS points that meet the conditions from the image; the phase mapping module receives the PS point information identified by the PS point selection module and the SAR image data, maps the distance from the PS point to the image origin into the phase index of the echo signal, and extracts the phase information of the corresponding PS point from all the pulse signals that have undergone distance compression; the phase averaging module receives the phase information extracted by the phase mapping module, uses the truncated mean method to average the phase to eliminate the initial phase error of the transmitted signal, and calculates the interference phase; the shape variable calculation module receives the interference phase calculated by the phase averaging module, and calculates the shape variable of the PS point according to the shape variable calculation formula.

8. The InSAR sub-millimeter-level high-precision deformation detection system based on initial phase error elimination according to claim 7 is characterized in that: The millimeter wave radar device uses a linear frequency modulation signal for transmission.

9. The InSAR sub-millimeter-level high-precision deformation detection system based on initial phase error elimination according to claim 7, characterized in that: The PS point selection module adopts the amplitude deviation index threshold method, the coherence coefficient threshold method and the intensity threshold method to jointly select PS points.

10. The InSAR sub-millimeter-level high-precision deformation detection system based on initial phase error elimination according to claim 7, characterized in that: The detection system also includes an alarm module, which is used to issue an alarm when the deformation amount is greater than a threshold value.

Images