Remote sensing dynamic monitoring method for sensitive ecological region and embankment along river

Abstract

The invention relates to the technical field of remote sensing monitoring, and discloses a remote sensing dynamic monitoring method for a sensitive ecological region and an embankment along a river, which comprises the following steps: dividing the sensitive ecological region and the embankment along the river, and collecting long-time-sequence multi-source remote sensing data and ground actual measurement data; preprocessing the Landsat series satellite image data, calculating an ecological index, performing regression analysis, and constructing a comprehensive ecological index monitoring model; the Landsat series satellite image data and the actually measured water level data are fused through a spatial interpolation method, and a water level safety threshold model is constructed; the method comprises the following steps: preprocessing Sentinel-1 / 3 satellite data, screening image pairs subjected to interference processing, and calculating a differential interference phase to solve terrain deformation rates of a sensitive ecological region and an embankment along a river; performing weight assignment and scoring grade setting on the ecological state, the water level state and the terrain state according to an importance principle, and performing summation through a weighted average method to obtain an overall ecological environment safety condition; the method improves the monitoring precision of the ecological environment.

Description

Technical Field

[0001] The invention relates to the technical field of remote sensing monitoring, and in particular to a remote sensing dynamic monitoring method for sensitive ecological areas and river banks. Background Art

[0002] With the intensification of global climate change and human activities, the stability of ecologically sensitive areas and riverbanks faces severe challenges. Traditional monitoring methods have problems such as limited coverage, long data acquisition cycle, and insufficient accuracy, which can hardly meet the needs of modern environmental monitoring. Summary of the invention

[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides a remote sensing dynamic monitoring method for sensitive ecological areas and river embankments, which is used to solve the problem of low accuracy in monitoring the ecological environmental safety status of ecologically sensitive areas and river embankments in traditional monitoring methods, thereby achieving accuracy in the assessment of the overall ecological environmental safety status and improving monitoring accuracy.

[0004] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0005] The remote sensing dynamic monitoring method for sensitive ecological areas and river banks includes the following steps:

[0006] S1. Delineate sensitive ecological areas and riverbanks, and collect long-term multi-source remote sensing data and ground-based measured data;

[0007] Among them, the long-term multi-source remote sensing data include Landsat series satellite image data and Sentinel-1 / 3 satellite data; the ground measured data are measured water level data;

[0008] S2. After preprocessing the Landsat satellite image data, the ecological index is calculated and regression analysis is performed to build a comprehensive ecological index monitoring model for monitoring the ecological status of sensitive ecological areas and riverbanks;

[0009] S3. Use spatial interpolation method to fuse Landsat satellite image data with measured water level data to build a water level safety threshold model for monitoring water level status in sensitive ecological areas and along riverbanks;

[0010] S4. Preprocess the Sentinel-1 / 3 satellite data and select the image pairs for interferometric processing. By calculating the differential interferometric phase, the terrain deformation rate of sensitive ecological areas and river banks is solved to monitor the terrain status of sensitive ecological areas and river banks.

[0011] S5. The ecological status, water level status and terrain status of sensitive ecological areas and river banks are weighted and rated according to the principle of importance, and the weighted average method is used to sum them up to obtain the overall ecological environmental safety status of sensitive ecological areas and river banks.

[0012] The present invention has the following beneficial effects:

[0013] The remote sensing dynamic monitoring method for sensitive ecological areas and riverbanks proposed in the present invention combines long-term multi-source remote sensing data with ground-measured data to obtain the ecological status, water level status and terrain status of sensitive ecological areas and riverbanks respectively, which can accurately evaluate the overall ecological environmental safety status of sensitive ecological areas and riverbanks, and improve the accuracy of ecological environmental monitoring in the study area. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The figure is a flow chart of the remote sensing dynamic monitoring method for sensitive ecological areas and river banks according to the present invention. DETAILED DESCRIPTION

[0015] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0016] like Figure 1 As shown, the remote sensing dynamic monitoring method for sensitive ecological areas and river banks includes the following steps S1-S5:

[0017] S1. Delineate sensitive ecological areas and riverbanks, and collect long-term multi-source remote sensing data and ground-measured data; among them, long-term multi-source remote sensing data include Landsat series satellite image data and Sentinel-1 / 3 satellite data; ground-measured data are measured water level data.

[0018] In this embodiment, the purpose of acquiring long-term multi-source remote sensing data and ground-measured data is to provide comprehensive and accurate data support for the ecological and environmental safety status of sensitive ecological areas and riverbanks.

[0019] S2. After data preprocessing of Landsat series satellite image data, the ecological index is calculated and regression analysis is performed to construct a comprehensive ecological index monitoring model for monitoring the ecological status of sensitive ecological areas and riverbanks.

[0020] Specifically, step S2 specifically includes S21-S24:

[0021] S21. Perform data preprocessing including radiation correction, atmospheric correction and geometric correction on Landsat series satellite image data to obtain corrected Landsat series satellite image data.

[0022] In this embodiment, the purpose of data preprocessing is to make satellite image data closer to the actual reflectivity or radiation brightness value of the surface by eliminating the influence of factors such as atmosphere, sensor characteristics and solar altitude angle on remote sensing data, thereby improving the accuracy of the data.

[0023] S22. Based on the corrected Landsat satellite image data, the vegetation index, improved water index and temperature vegetation drought index are calculated, namely:

[0024]

[0025] Among them, DNVI represents the vegetation index, which is used to evaluate the vegetation coverage and growth status, NIR represents the near-infrared light band of the calibrated Landsat series satellite image data, Red represents the red light band of the calibrated Landsat series satellite image data, MNDWI represents the improved water body index, which is used to identify water bodies in vegetation-covered areas, SWIR represents the short infrared light band of the calibrated Landsat series satellite image data, TVDI represents the temperature vegetation drought index, which is used to evaluate soil moisture, T S The surface temperature of pixels representing sensitive ecological areas and riverbanks, T Smin Indicates the minimum surface temperature corresponding to the vegetation index NDVI, T Smax It represents the maximum surface temperature corresponding to the vegetation index NDVI, a and b represent the dry edge fitting coefficients, and a′ and b′ represent the wet edge fitting coefficients.

[0026] In this embodiment, the vegetation index is used to evaluate the vegetation coverage and growth status, that is, the value range of NDVI is from -1 to 1. When red light is absorbed in large quantities and NIR is reflected in large quantities, NDVI will be closer and closer to 1, indicating more healthy and dense vegetation; on the contrary, when the vegetation is unhealthy, more red light is reflected and less NIR is reflected, and NDVI is closer and closer to 0. In addition, in places without vegetation, red light will be reflected in large quantities, while NIR will be basically absorbed, and the NDVI value at this time will be close to -1.

[0027] At the same time, since the chlorophyll in the vegetation has a strong reflective effect on light in the green light band, the reflectivity of the water body in the green light band is relatively low, while the reflectivity of the vegetation in this band is relatively high; at the same time, the absorption characteristics of the water body in the short infrared band are very strong and it almost does not reflect the light in this band, so the reflectivity of the vegetation and soil in this band is relatively high; therefore, this embodiment proposes to introduce the short infrared light band to calculate the improved water body index for identifying the water bodies in the vegetation and building areas, so as to effectively reduce the interference of vegetation and buildings on water body extraction and improve the accuracy of water body detection.

[0028] In addition, in the two-dimensional feature space composed of vegetation index (NDVI) and land surface temperature (LST), there is a dry edge and a wet edge; the dry edge represents the set of pixels with the same vegetation cover but the lowest soil moisture (i.e., the highest surface temperature), while the wet edge represents the set of pixels with the same vegetation cover but the highest soil moisture (i.e., the lowest surface temperature). Therefore, by fitting the linear relationship between the dry edge and the wet edge, the dry edge fitting coefficient and the wet edge fitting coefficient can be obtained.

[0029] S23, linear regression analysis was performed on vegetation index, improved water index and temperature vegetation drought index to construct a comprehensive ecological index monitoring model, namely:

[0030] Y=α0NDVI+α1MNDWI+α2TVDI+ε

[0031] Among them, Y represents the comprehensive ecological index, ε represents the error term, and α0, α1, α2, and α3 all represent regression coefficients.

[0032] In this embodiment, the comprehensive ecological index monitoring model can accurately reflect the status of the ecosystem and is used to identify and evaluate the characteristics of ecological spatiotemporal evolution. At the same time, it can also comprehensively consider information on vegetation coverage, water distribution, soil moisture, etc., providing a scientific and effective method for the comprehensive assessment of the ecosystem. Compared with a single indicator, the comprehensive ecological index monitoring model can more comprehensively reflect the complexity and dynamics of the ecosystem, thereby improving the precision and accuracy of ecological monitoring in sensitive ecological areas and along river banks.

[0033] S24. Determine whether the comprehensive ecological index is greater than the ecological preset value. If so, the ecological status of the sensitive ecological area and the riverbank is good; otherwise, the ecological status of the sensitive ecological area and the riverbank is poor.

[0034] Among them, the ecological preset values ​​are set based on the historical vegetation coverage, water conditions and soil moisture conditions of sensitive ecological areas and river banks.

[0035] S3. Use spatial interpolation method to fuse Landsat satellite image data with measured water level data to construct a water level safety threshold model for monitoring water level status in sensitive ecological areas and along riverbanks.

[0036] Specifically, step S3 specifically includes S31-S34:

[0037] S31. Perform data preprocessing including radiation correction, atmospheric correction and geometric correction on the Landsat series satellite image data to obtain corrected Landsat series satellite image data. Simultaneously, use a spatial feature extraction method to extract water body features from the corrected Landsat series satellite image data to obtain water body area data.

[0038] In this embodiment, the process of water body feature extraction is as follows: (1) Selecting a suitable band: According to the spectral characteristics of the water body, a suitable band is selected for water body extraction. (2) Calculating the water body index: Using the selected band, the water body index is calculated; (3) Threshold segmentation: According to the value of the water body index, a suitable threshold is set for segmentation; pixels above the threshold are considered to be water body areas, while pixels below the threshold are considered to be other ground objects.

[0039] S32, performing data preprocessing on the measured water level data, including removing outliers and interpolating missing values, to obtain corrected measured water level data, and at the same time aligning the time of the corrected measured water level data with the water body area data to obtain aligned measured water level data.

[0040] In this embodiment, by removing outliers and filling missing values, errors and incomplete parts in the data are eliminated, and the accuracy and reliability of the data set are improved. At the same time, the method of removing outliers and interpolating missing values ​​is not limited. For example, the cubic spline interpolation method can be used to interpolate missing values, and the standard deviation method can be used to remove outliers.

[0041] S33. Obtain the spatial distribution characteristics of sensitive ecological areas and riverbanks, and use the Kriging interpolation method to interpolate the aligned measured water level data to obtain interpolated measured water level data.

[0042] Among them, the calculation formula of Kriging interpolation is:

[0043]

[0044] Among them, Z(x0) represents the water level value of the interpolation point x0, n represents the total number of known points, and λ i represents the weight coefficient calculated using the semivariogram function, Z(x i ) represents a known point x i water level value.

[0045] In this embodiment, the Kriging interpolation method can generate more accurate and reliable water level interpolation data by considering the spatial correlation between known points, so as to facilitate data fusion in subsequent steps, thereby providing strong support for water resource management, flood prevention and disaster reduction in sensitive ecological areas and along riverbanks.

[0046] S34. Superimpose the water body area data with the interpolated measured water level data to obtain the water level distribution map of the sensitive ecological area and the river embankment, and construct a water level safety threshold model to obtain the water level status of the sensitive ecological area and the river embankment.

[0047] In this embodiment, GIS software (such as ArcGIS, QGIS, etc.) is used to superimpose the water body area data with the interpolated water level data. This can be achieved through the layer overlay function, with the water level data used as the base map and the water body area data displayed as the upper layer, thereby obtaining a water level distribution map of sensitive ecological areas and river banks, so as to more intuitively understand the changes in water levels.

[0048] Specifically, in step S34, the water level safety threshold model is constructed to obtain the water level status of sensitive ecological areas and riverbanks as follows:

[0049] According to the water level distribution map of sensitive ecological areas and river banks, the maximum, minimum and average water level values ​​are extracted.

[0050] In this embodiment, the maximum value, minimum value and average value of the water level value reflect the basic distribution of the water level data; wherein the maximum value and the minimum value represent the highest and lowest levels of the water level during the observation period, respectively, and the average value reflects the average level of the water level.

[0051] Based on the maximum, minimum and average values ​​of the water level, a water level safety threshold model is constructed, namely:

[0052] F=IF(ρ1,Avg_Z∈[MIN_Safe_Z,MAX_Safe_Z],IF(ρ2,Max_Z>MAX_Safe_Z),IF(ρ3,Min_Z>MIN_Safe_Z))

[0053] Among them, F represents the water level safety threshold model, IF represents the judgment function, ρ1 represents that the current water level is in a safe range, Avg_Z represents the average value of the water level, MIN_Safe_Z represents the minimum safe water level value, MAX_Safe_Z represents the maximum safe water level value, ρ2 represents that the current water level has a flood risk, Max_Z represents the maximum value of the water level, ρ3 represents that the current water level has a water resource shortage risk, and Min_Z represents the minimum value of the water level.

[0054] In this embodiment, different levels of water level safety status are defined based on the statistical characteristics of the water level value (average value, maximum value, minimum value); by setting the minimum safe water level value (usually higher than the minimum value but lower than the average value) and the maximum water level safety value (usually lower than the maximum value but higher than the average value), three main water level safety intervals can be divided: safety range, flood risk area and water resource shortage risk area. At the same time, by constructing a water level safety threshold model, the water level safety status of different areas can be quantitatively evaluated, which helps decision makers to more accurately understand the water level safety status of different areas, so as to formulate targeted management measures.

[0055] S4. Preprocess the Sentinel-1 / 3 satellite data and select the image pairs for interferometric processing. By calculating the differential interferometric phase, the terrain deformation rate of sensitive ecological areas and river banks is solved to monitor the terrain status of sensitive ecological areas and river banks.

[0056] Specifically, step S4 specifically includes S41-S49:

[0057] S41. Perform data preprocessing including orbit correction, radiation correction, thermal noise removal and filtering on the Sentinel-1 / 3 satellite data to obtain corrected SAR image time series data.

[0058] S42. Based on the corrected SAR image time series data, the influence pairs for interference processing are screened according to the spatial baseline length, so as to obtain the interference pairs.

[0059] Among them, the interference pair selection conditions are:

[0060]

[0061] Where N represents the SAR image time series data, and M represents the interferometric pair.

[0062] In this embodiment, the spatial baseline length and angle between the image pairs are calculated according to the orbital parameters of the SAR satellite, and the image pairs with a moderate spatial baseline length are selected to ensure good interferometric coherence.

[0063] S43, if for the jth interferogram, the main image acquisition time of pixel x in the SAR image time series data is t A , auxiliary image acquisition time is t B , then the differential interference phase of pixel x is:

[0064]

[0065] Among them, Δφ diff_j (x) represents the differential interference phase of pixel x in the jth interference pattern, φ(t B,x) represents the pixel x at time t B The phase value at φ(t A ,x) represents the pixel x at time t A The phase value at ω represents the radar wavelength, d(t B ,x) represents the pixel x at time t B The deformation along the radar line of sight, d(t A ,x) represents the pixel x at time t A The deformation along the radar line of sight.

[0066] In this embodiment, in SAR interferometry (InSAR), the differential interferometry phase is a key parameter reflecting surface deformation, that is, it includes the contribution of multiple factors such as surface deformation, atmospheric delay, terrain undulation, etc.; therefore, when two SAR images (primary image and auxiliary image) are acquired at almost the same viewing angle but at different times, the phase difference between them (i.e., differential interferometry phase) can reveal the deformation information of the surface during this period of time, where deformation refers to the displacement along the radar line of sight.

[0067] S44, if the main image acquisition time of the i1th interference pair is t IEi1 , auxiliary image acquisition time is t ISi1 , and i1∈M, t IEi1 >t ISi1 , then the differential interference phase of the i1th interference pair is expressed as:

[0068] Δφ i1 =φ(t iSi1 )-φ(t IEi1 )

[0069] Among them, Δφ i1 represents the differential interference phase of the i1th interference pair, φ(t ISi1 ) indicates that the i1th interference pair has an auxiliary image acquisition time of t ISi1 The phase, φ(t IEi1 ) indicates that the i1th interference pair is acquired at the main image acquisition time t IEi1 phase.

[0070] S45, combining the differential interference phases of all interference pairs into a differential interference phase vector, and constructing an M×N coefficient matrix A so as to satisfy the following relationship:

[0071] Δφ=AΔR

[0072] Among them, Δφ represents the differential interference phase vector, and ΔR represents the deformation parameter vector.

[0073] Among them, the deformation parameter vector ΔR is the deformation at different time points.

[0074] In this embodiment, steps S44-S45, by obtaining the differential interference phase of all interference pairs to form a differential interference phase vector, are used to construct a matrix relationship, that is, Δφ=AΔR, so that ΔR can be solved in subsequent steps to obtain the deformation of the surface at different time points extracted from the differential interference phase vector Δφ, thereby realizing quantitative monitoring of surface deformation.

[0075] S46. Use the least square method to solve Δφ=AΔR, and we have:

[0076] ΔR=(A T A) -1 A T Δφ

[0077] Where T stands for transpose.

[0078] S47. Based on the deformation variable parameter vector, the deformation rate of the pixel is calculated, that is:

[0079]

[0080] Among them, v represents the deformation rate of the pixel, Δt represents the adjacent time interval, that is, Δt = t A -t B .

[0081] S48. Process the deformation rates of all pixels to obtain the terrain deformation rate.

[0082] In this embodiment, steps S46-S48 calculate the deformation rate of the pixel from the deformation variable parameter vector obtained by differential interferometry, and further process it to obtain a terrain deformation rate map of the entire study area, providing important data support for geological disaster warning, environmental monitoring, scientific research and other fields.

[0083] S49. Determine whether the terrain deformation rate is greater than a preset value of the terrain deformation rate. If so, there are geological risks in the sensitive ecological area and the river embankments. Otherwise, the sensitive ecological area and the river embankments are in a safe state.

[0084] Among them, the preset value of terrain deformation rate is set according to the historical terrain deformation rate conditions of sensitive ecological areas and river embankments.

[0085] S5. The ecological status, water level status and terrain status of sensitive ecological areas and river banks are weighted and rated according to the principle of importance, and the weighted average method is used to sum them up to obtain the overall ecological environmental safety status of sensitive ecological areas and river banks.

[0086] Specifically, step S5 includes S51-S53:

[0087] S51. Weights are assigned to the ecological status, water level status and terrain status of sensitive ecological areas and riverbanks according to the principle of importance. The specific process is as follows:

[0088] If the ecological status of sensitive ecological areas and riverbanks is good, their weight is assigned to u1.

[0089] If the ecological status of sensitive ecological areas and riverbanks is poor, their weight is assigned to y2.

[0090] If the current water level is within the safe range, its weight is assigned to z1.

[0091] If there is a flood risk at the current water level, its weight is assigned to z2.

[0092] If there is a risk of water shortage at the current water level, its weight is assigned to z3.

[0093] If there are geological risks in sensitive ecological areas and riverbanks, their weights are assigned to τ1.

[0094] If the sensitive ecological areas and river banks are in a safe state, their weights are assigned to τ2.

[0095] In this embodiment, u1>u2 and z1>z3>z2 and τ1<τ2, and u1=z1=τ2=0.5, z3=0.3, u2=z2=τ1=0.2.

[0096] S52. Set scoring levels for the ecological status, water level status and terrain status of sensitive ecological areas and river banks, specifically:

[0097] When the ecological status of sensitive ecological areas and river banks is good, the scoring level is set as excellent.

[0098] When the ecological status of sensitive ecological areas and river banks is poor, the score level is set to poor.

[0099] When the current water level is within a safe range, the rating level is set to excellent.

[0100] When there is a flood risk at the current water level, the score level is set to poor.

[0101] When the current water level poses a risk of water shortage, the score is set to medium.

[0102] When there are geological risks in sensitive ecological areas and river banks, the score level is set to poor.

[0103] When sensitive ecological areas and riverbanks are in a safe state, their rating level is set as excellent;

[0104] Among them, the score corresponding to the excellent rating is The corresponding score for the rating level is The score corresponding to the poor rating is

[0105] S53. The ecological status, water level status and terrain status of the sensitive ecological area and the river bank are multiplied by the corresponding weights and then weighted averaged to obtain the safety value of the sensitive ecological area and the river bank, that is:

[0106]

[0107] in, represents the safety value of sensitive ecological areas and riverbanks, and choic(·||·) represents the selection function, which is used to indicate that only one set of product values ​​is selected.

[0108] Among them, the larger the safety value, the better the overall ecological and environmental safety of sensitive ecological areas and river banks.

[0109] To sum up, the remote sensing dynamic monitoring method for sensitive ecological areas and riverbanks proposed in the present invention combines long-term multi-source remote sensing data with ground-measured data to obtain the ecological status, water level status and terrain status of sensitive ecological areas and riverbanks respectively, which can effectively evaluate the overall ecological environmental safety status of sensitive ecological areas and riverbanks and improve the accuracy of ecological environmental monitoring in the study area.

[0110] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0111] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A remote sensing dynamic monitoring method for sensitive ecological areas and river banks, characterized in that: The following steps are involved: S1. Delineate sensitive ecological areas and riverbanks, and collect long-term multi-source remote sensing data and ground-based measured data; Among them, the long-term multi-source remote sensing data include Landsat series satellite image data and Sentinel-1 / 3 satellite data; the ground measured data are measured water level data; S2. After preprocessing the Landsat satellite image data, the ecological index is calculated and regression analysis is performed to build a comprehensive ecological index monitoring model for monitoring the ecological status of sensitive ecological areas and riverbanks; S3. Use spatial interpolation method to fuse Landsat satellite image data with measured water level data to build a water level safety threshold model for monitoring water level status in sensitive ecological areas and along riverbanks; S4. Preprocess the Sentinel-1 / 3 satellite data and select the image pairs for interferometric processing. By calculating the differential interferometric phase, the terrain deformation rate of sensitive ecological areas and river banks is solved to monitor the terrain status of sensitive ecological areas and river banks. S5. The ecological status, water level status and terrain status of sensitive ecological areas and river banks are weighted and rated according to the principle of importance, and the weighted average method is used to sum them up to obtain the overall ecological environmental safety status of sensitive ecological areas and river banks.

2. The remote sensing dynamic monitoring method for sensitive ecological areas and riverbanks according to claim 1 is characterized in that: Step S2 specifically includes: S21, performing data preprocessing including radiation correction, atmospheric correction and geometric correction on the Landsat series satellite image data to obtain corrected Landsat series satellite image data; S22. Based on the corrected Landsat satellite image data, the vegetation index, improved water index and temperature vegetation drought index are calculated, namely: Among them, NDVI represents the vegetation index, which is used to evaluate the vegetation coverage and growth status, NIR represents the near-infrared light band of the calibrated Landsat series satellite image data, Red represents the red light band of the calibrated Landsat series satellite image data, MNDWI represents the improved water body index, which is used to identify water bodies in vegetation-covered areas, SWIR represents the short infrared light band of the calibrated Landsat series satellite image data, TVDI represents the temperature vegetation drought index, which is used to evaluate soil moisture, T S The surface temperature of pixels representing sensitive ecological areas and along riverbanks, T Smin Indicates the minimum surface temperature corresponding to the vegetation index NDVI, T Smax It represents the maximum surface temperature corresponding to the vegetation index NDVI, a and b represent the dry edge fitting coefficients, and a′ and b′ represent the wet edge fitting coefficients; S23, linear regression analysis was performed on vegetation index, improved water index and temperature vegetation drought index to construct a comprehensive ecological index monitoring model, namely: Y=α0NDVI+α1MNDWI+α2TVDI+ε Among them, Y represents the comprehensive ecological index, ε represents the error term, and α0, α1, α2, and α3 all represent regression coefficients; S24. Determine whether the comprehensive ecological index is greater than the ecological preset value. If so, the ecological status of the sensitive ecological area and the riverbank is good; otherwise, the ecological status of the sensitive ecological area and the riverbank is poor; Among them, the ecological preset values ​​are set based on the historical vegetation coverage, water conditions and soil moisture conditions of sensitive ecological areas and river banks.

3. The remote sensing dynamic monitoring method for sensitive ecological areas and riverbanks according to claim 2 is characterized in that: Step S3 specifically includes: S31, performing data preprocessing including radiation correction, atmospheric correction and geometric correction on the Landsat series satellite image data to obtain corrected Landsat series satellite image data, and simultaneously using a spatial feature extraction method to extract water body features from the corrected Landsat series satellite image data to obtain water body area data; S32, performing data preprocessing on the measured water level data, including removing outliers and interpolating missing values, to obtain corrected measured water level data, and simultaneously aligning the time of the corrected measured water level data with the water body area data to obtain aligned measured water level data; S33, obtaining the spatial distribution characteristics of sensitive ecological areas and riverbanks, and interpolating the aligned measured water level data using the Kriging interpolation method to obtain interpolated measured water level data; Among them, the calculation formula of Kriging interpolation is: Among them, Z(x0) represents the water level value of the interpolation point x0, n represents the total number of known points, and λ i represents the weight coefficient calculated using the semivariogram function, Z(x i ) represents a known point x i The water level value; S34. Superimpose the water body area data with the interpolated measured water level data to obtain the water level distribution map of the sensitive ecological area and the river embankment, and construct a water level safety threshold model to obtain the water level status of the sensitive ecological area and the river embankment.

4. The remote sensing dynamic monitoring method for sensitive ecological areas and riverbanks according to claim 3 is characterized in that: In step S34, the specific process of constructing a water level safety threshold model and obtaining the water level status of sensitive ecological areas and riverbanks is as follows: According to the water level distribution map of sensitive ecological areas and river banks, the maximum, minimum and average water level values ​​are extracted; Based on the maximum, minimum and average values ​​of the water level, a water level safety threshold model is constructed, namely: F=IF(ρ1,Avg_Z ∈[MIN_safe_Z,MAX_Safe_Z],IF(ρ2,Max_Z>MAX_Safe_Z),IF(ρ3,Min_Z>MIN_Safe_Z)) Among them, F represents the water level safety threshold model, IF represents the judgment function, ρ1 represents that the current water level is in a safe range, Avg_Z represents the average value of the water level, MIN_Safe_Z represents the minimum safe water level value, MAX_Safe_Z represents the maximum safe water level value, ρ2 represents that the current water level has a flood risk, Max_Z represents the maximum value of the water level, ρ3 represents that the current water level has a water resource shortage risk, and Min_Z represents the minimum value of the water level.

5. The remote sensing dynamic monitoring method for sensitive ecological areas and riverbanks according to claim 4 is characterized in that: Step S4 specifically includes: S41, performing data preprocessing including orbit correction, radiation correction, thermal noise removal and filtering on the Sentinel-1 / 3 satellite data to obtain corrected SAR image time series data; S42, based on the corrected SAR image time series data, screening the influence pairs for interferometric processing according to the spatial baseline length, that is, obtaining the interferometric pairs; Among them, the interference pair selection conditions are: Among them, N represents the SAR image time series data, and M represents the interferometric pair; S43, if for the jth interferogram, the main image acquisition time of pixel x in the SAR image time series data is t A , auxiliary image acquisition time is t B , then the differential interference phase of pixel x is: Among them, Δφ diff_j (x) represents the differential interference phase of pixel x in the jth interference pattern, φ(t B ,x) represents the pixel x at time t B The phase value at φ(t A ,x) represents the pixel x at time t A The phase value at ω represents the radar wavelength, d(t B ,x) represents the pixel x at time t B The deformation along the radar line of sight, d(t A ,x) represents the pixel x at time t A The deformation along the radar line of sight; S44, if the main image acquisition time of the i1th interference pair is t IEi1 , auxiliary image acquisition time is t ISi1 , and i1∈M, t IEi1 >t ISi1 , then the differential interference phase of the i1th interference pair is expressed as: Df i1 =φ(t ISi1 )-φ(t IEi1 ) Among them, Δφ i1 represents the differential interference phase of the i1th interference pair, φ(t ISi1 ) indicates that the i1th interference pair has an auxiliary image acquisition time of t ISi1 The phase, φ(t IEi1 ) indicates that the i1th interference pair is acquired at the main image acquisition time t IEi1 The phase of S45, combining the differential interference phases of all interference pairs into a differential interference phase vector, and constructing an M×N coefficient matrix A so as to satisfy the following relationship: Δφ=AΔR Among them, Δφ represents the differential interference phase vector, ΔR represents the deformation parameter vector; Among them, the deformation parameter vector ΔR is the deformation at different time points; S46. Use the least square method to solve Δφ=AΔR, and we have: ΔR=(A T A) -1 A T Df Where T represents transpose; S47. Based on the deformation variable parameter vector, the deformation rate of the pixel is calculated, that is: Among them, v represents the deformation rate of the pixel, and Δt represents the time interval between adjacent pixels; S48, performing data processing on the deformation rates of all pixels to obtain the terrain deformation rate; S49, determining whether the terrain deformation rate is greater than a preset value of the terrain deformation rate, if so, there are geological risks in the sensitive ecological area and the river bank, otherwise, the sensitive ecological area and the river bank are in a safe state; Among them, the preset value of terrain deformation rate is set according to the historical terrain deformation rate conditions of sensitive ecological areas and river embankments.

6. The remote sensing dynamic monitoring method for sensitive ecological areas and riverbanks according to claim 5 is characterized in that: Step S5 specifically includes: S51. Weights are assigned to the ecological status, water level status and terrain status of sensitive ecological areas and riverbanks according to the principle of importance. The specific process is as follows: If the ecological status of sensitive ecological areas and riverbanks is good, their weight is assigned to u1; If the ecological status of sensitive ecological areas and riverbanks is poor, their weight is assigned to u2; If the current water level is within the safe range, its weight is assigned to z1; If there is a flood risk at the current water level, its weight is assigned to z2; If there is a risk of water shortage at the current water level, its weight is assigned to z3; If there are geological risks in sensitive ecological areas and riverbanks, their weights are assigned to τ1; If the sensitive ecological area and the riverbank are in a safe state, their weight is assigned to τ2; S52. Set scoring levels for the ecological status, water level status and terrain status of sensitive ecological areas and river banks, specifically: When the ecological status of sensitive ecological areas and riverbanks is good, the rating level is set as excellent; When the ecological status of sensitive ecological areas and riverbanks is poor, the rating level is set as poor; When the current water level is within a safe range, the rating is set to be excellent; When there is a flood risk at the current water level, the score is set to poor; When the current water level poses a risk of water shortage, the score is set to medium; When there are geological risks in sensitive ecological areas and riverbanks, the rating level is set as poor; When sensitive ecological areas and riverbanks are in a safe state, their rating level is set as excellent; Among them, the score corresponding to the excellent rating is θ1=90, the score corresponding to the medium rating is θ2=60, and the score corresponding to the poor rating is θ2=30; S53. The ecological status, water level status and terrain status of the sensitive ecological area and the river bank are multiplied by the corresponding weights and then weighted averaged to obtain the safety value of the sensitive ecological area and the river bank, that is: in, represents the safety value of sensitive ecological areas and riverbanks, choic(·||·) represents the selection function, which is used to indicate that only one set of product values ​​is selected; Among them, the larger the safety value, the better the overall ecological and environmental safety of sensitive ecological areas and river banks.

Images