Settlement process automatic identification method based on adaptive iteration seasonal decomposition and trend analysis

Abstract

The invention discloses a settlement process automatic identification method based on adaptive iteration seasonal decomposition and trend analysis. The method comprises the following steps: acquiring a multi-temporal synthetic aperture radar interference data set; the method comprises the following steps: preprocessing a time sequence in a multi-temporal synthetic aperture radar interference data set based on an iterative seasonal decomposition method to remove seasonal components and noise in the time sequence; carrying out pixel-by-pixel analysis on the preprocessed time sequence based on an absolute difference settlement automatic identification method with a sliding window, and quantifying the settlement trend of each settlement point in the research area; and based on the settlement trend of each settlement point, calculating and outputting a total settlement ratio and an annual settlement ratio corresponding to each settlement point. According to the method, the seasonal components are automatically decomposed from the time sequence data through the iterative seasonal trend decomposition method, and the settlement period in the whole time sequence is quantified through the automatic settlement identification method, so that the time-space granularity quantification of the settlement point is realized, and the detection accuracy of deformation outside a spatial domain is improved.

Description

Technical Field

[0001] The invention relates to the field of surveying and mapping technology, and in particular to an automatic identification method for a sedimentation process with adaptive iterative seasonal decomposition and trend analysis. Background Art

[0002] As the construction of subway lines accelerates, construction accidents frequently occur due to technical defects, insufficient engineering monitoring, and complex geological conditions. Compared with traditional subway monitoring technology, multi-temporal InSAR (MT-InSAR, i.e. multi-temporal synthetic aperture radar interferometry) can achieve large-scale, long-time series, and high-precision deformation monitoring with an accuracy of millimeters. Since subway lines are usually built in urban environments, there are a large number of permanent scatterers in such environments, which makes the point-target-based MT-InSAR processing algorithm more suitable for exploring the evolution of spatiotemporal deformation.

[0003] There are two main types of existing MT-InSAR methods for defining dangerous areas: the first type of method classifies permanent scatterers (PS) into types related to subway structures and uses hazard assessment methods to characterize the degree of deformation hazard. The second type of method combines MT-InSAR with machine learning to assess the level of ground subsidence risk and predict dynamic deformation patterns along the subway network. Although these methods provide a new way to identify dangerous areas, most existing technologies are highly dependent on the calculation of deformation rates, which limits MT-InSAR's ability to detect deformation risks outside the spatial domain, resulting in reduced accuracy in detecting dangerous areas.

[0004] Therefore, the prior art needs to be improved. Summary of the invention

[0005] The technical problem to be solved by the present invention is that, in view of the defects of the prior art, the present invention provides an automatic identification method of sedimentation process with adaptive iterative seasonal decomposition and trend analysis to solve the problem of low accuracy of the existing MT-InSAR method for defining dangerous areas in detecting deformation outside the spatial domain.

[0006] The technical solution adopted by the present invention to solve the technical problem is as follows: In a first aspect, the present invention provides a method for automatically identifying a sedimentation process by adaptive iterative seasonal decomposition and trend analysis, comprising: Acquire multi-temporal synthetic aperture radar interferometry datasets; Preprocessing the time series in the multi-temporal synthetic aperture radar interferometry dataset based on an iterative seasonal decomposition method to remove seasonal components and noise in the time series; The preprocessed time series were analyzed pixel by pixel based on the absolute difference settlement automatic identification method with sliding window to quantify the settlement trend of each settlement point in the study area. Based on the settlement trend of each settlement point, the overall settlement ratio and annual settlement ratio corresponding to each settlement point are calculated and output.

[0007] In one implementation, the preprocessing of the time series in the multi-temporal synthetic aperture radar interferometry dataset based on the iterative seasonal decomposition method includes: Using a local weighted regression scatter point smoothing algorithm to smooth and fit the time series, decomposing the time series into a seasonal component, a trend component and a residual component; According to the seasonal component, the trend component and the residual component, the time series is analyzed pixel by pixel to determine the optimal period within the preset period interval: ; in, is the optimal period, is the preset period interval, is the standard deviation corresponding to the residual component.

[0008] In one implementation, the expression of the local weighted regression scatter point smoothing algorithm is: ; in, is the length of the time series, is the input time series, , , They are respectively represented as the seasonal component, the trend component and the residual component.

[0009] In one implementation, the automatic identification method of settlement based on absolute difference with sliding window performs pixel-by-pixel analysis on the preprocessed time series to quantify the settlement trend of each settlement point in the study area, including: Define the time series as a vector ; Wherein, the vector Contains N displacement values; Calculate the median of adjacent windows at each time point; The absolute difference of each time point is calculated according to the median of the adjacent windows, and the settlement trend of the corresponding settlement point is determined according to the absolute difference, the central tendency and dispersion index and the preset threshold.

[0010] In one implementation, the median of the adjacent windows is expressed as: ; ; in, and are the data obtained from the current sliding window and the previous sliding window respectively, Indicates the window size.

[0011] In one implementation, the absolute difference of each time point is calculated based on the median of the adjacent windows, and the settlement trend of the corresponding settlement point is determined based on the absolute difference, the central tendency and dispersion index, and a preset threshold, including: Calculate the absolute difference at each time point: ; According to the absolute difference, the central tendency and dispersion index and the preset threshold, the settlement trend of the corresponding settlement point is determined using the following conditions: ; ; ; in, is the absolute difference; are the central tendency and dispersion indices; is the preset threshold.

[0012] In one implementation, the calculation and output of the overall settlement ratio and annual settlement ratio corresponding to each settlement point based on the settlement trend of each settlement point includes: Based on the settlement trend of each settlement point, the ratio of the settlement trend segment corresponding to each settlement point to the total time series is calculated to obtain the overall settlement ratio corresponding to each settlement point; Calculate the ratio of the subsidence trend fragment corresponding to each subsidence point compared to the annual time series to obtain the annual subsidence ratio corresponding to each subsidence point; Output the overall settlement ratio and annual settlement ratio corresponding to each settlement point.

[0013] In a second aspect, the present invention provides an automatic identification system for sedimentation process with adaptive iterative seasonal decomposition and trend analysis, comprising: A data acquisition module, used to acquire a multi-temporal synthetic aperture radar interferometry data set; An iterative seasonal decomposition module, used for preprocessing the time series in the multi-phase synthetic aperture radar interferometry data set based on an iterative seasonal decomposition method to remove seasonal components and noise in the time series; The settlement automatic identification module is used to analyze the preprocessed time series pixel by pixel based on the absolute difference settlement automatic identification method with a sliding window to quantify the settlement trend of each settlement point in the study area; The indicator output module is used to calculate and output the overall settlement ratio and annual settlement ratio corresponding to each settlement point based on the settlement trend of each settlement point.

[0014] In a third aspect, the present invention provides a terminal, comprising: a processor and a memory, wherein the memory stores a program for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis, and when the program for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis is executed by the processor, it is used to implement the operation of the method for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis as described in the first aspect.

[0015] In a fourth aspect, the present invention further provides a medium, which is a computer-readable storage medium, and which stores a program for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis. When the program for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis is executed by a processor, it is used to implement the operation of the method for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis as described in the first aspect.

[0016] The present invention adopts the above technical solution to achieve the following effects: The present invention obtains a multi-phase synthetic aperture radar interferometer data set, and can pre-process the time series in the multi-phase synthetic aperture radar interferometer data set based on an iterative seasonal decomposition method to remove seasonal components and noise in the time series; and the pre-processed time series is analyzed pixel by pixel through an absolute difference settlement automatic identification method with a sliding window to quantify the settlement trend of each settlement point in the study area; based on the settlement trend of each settlement point, the overall settlement ratio and annual settlement ratio corresponding to each settlement point can be calculated and output. The present invention automatically decomposes seasonal components from time series data through an iterative seasonal trend decomposition method, and quantifies the settlement period in the entire time series through a settlement automatic identification method, thereby realizing the spatiotemporal granularity quantification of settlement points and improving the detection accuracy of deformation outside the spatial domain. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0018] Figure 1It is a flow chart of the automatic identification method of the sedimentation process of the adaptive iterative seasonal decomposition and trend analysis in the present invention.

[0019] Figure 2 It is a schematic diagram of the optimal cycle selection process in the present invention.

[0020] Figure 3 It is a rendering of the automatic settlement identification method of the present invention.

[0021] Figure 4 It is a flow chart of automatic identification and processing of the sedimentation process in the present invention.

[0022] Figure 5 It is a functional principle diagram of a terminal in one implementation of the present invention.

[0023] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the present invention clearer and more specific, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0025] Exemplary Methods There are two main types of existing MT-InSAR methods for defining dangerous areas: the first type of method classifies permanent scatterers (PS) into types related to subway structures and uses hazard assessment methods to characterize the degree of deformation hazard. The second type of method combines MT-InSAR with machine learning to assess the level of ground subsidence risk and predict dynamic deformation patterns along the subway network. Although these methods provide a new way to identify dangerous areas, most existing technologies are highly dependent on the calculation of deformation rates, which limits MT-InSAR's ability to detect deformation risks outside the spatial domain, resulting in reduced accuracy in detecting dangerous areas.

[0026] In view of the above technical problems, an automatic identification method for the settlement process of adaptive iterative seasonal decomposition and trend analysis is provided in an embodiment of the present invention. The method can preprocess the time series in the multi-phase synthetic aperture radar interferometer data set based on the iterative seasonal decomposition method to remove the seasonal components and noise in the time series by acquiring a multi-phase synthetic aperture radar interferometer data set; and perform pixel-by-pixel analysis on the preprocessed time series by an absolute difference settlement automatic identification method with a sliding window to quantify the settlement trend of each settlement point in the study area; and calculate and output the overall settlement ratio and annual settlement ratio corresponding to each settlement point based on the settlement trend of each settlement point. In the embodiment of the present invention, the seasonal components are automatically decomposed from the time series data by the iterative seasonal trend decomposition method, and the settlement period in the entire time series is quantified by the settlement automatic identification method, so as to realize the spatiotemporal granularity quantification of the settlement points and improve the detection accuracy of deformation outside the spatial domain.

[0027] like Figure 1 As shown, an embodiment of the present invention provides a method for automatically identifying a sedimentation process by adaptive iterative seasonal decomposition and trend analysis, comprising the following steps: Step S100: Acquire a multi-temporal synthetic aperture radar interferometry data set.

[0028] Multi-temporal InSAR (MT-InSAR) technology has unique advantages in monitoring deformation and structural stability along subway lines. However, a large number of deformation points with complex features can be obtained through MT-InSAR, with the number reaching millions or even more, which makes it challenging to automatically identify hot spots in deformation time series. In addition, as the observation period continues to increase, the deformation rate of long time series is difficult to accurately reflect the local settlement risk conditions that occur at specific time periods during the entire monitoring period, such as deformation caused by construction along the subway line. Therefore, it is crucial to pay close attention to hot spots that show continuous settlement and faster settlement rates throughout the observation period and in specific time periods. Compared with relying on rate to classify deformation characteristics, it is more feasible to develop an automated method to comprehensively quantify settlement from massive MT-InSAR results for evaluating deformation caused by construction.

[0029] In this paper, a new method is proposed for quantitative evaluation of deformation results from MT-InSAR. First, an iterative seasonal decomposition (STL) method is introduced to determine the best period to separate seasonal components. Then, an absolute difference detector with a sliding window is proposed to quantify the proportion of subsidence within the time series, making deformation hotspots more visible.

[0030] In order to achieve quantitative evaluation of the deformation results of MT-InSAR, in this embodiment, it is necessary to obtain a multi-temporal synthetic aperture radar interferometry dataset corresponding to the study area, that is, to obtain an MT-InSAR deformation dataset; wherein, the MT-InSAR deformation dataset can be obtained from a public remote sensing data platform, or from a high-resolution and high-precision deformation monitoring application, or by using MT-InSAR software to process public SAR data.

[0031] In this embodiment, by obtaining the deformation results of MT-InSAR, the iterative seasonal decomposition (STL) method can be used to determine the optimal period for separating the seasonal components, and then the absolute difference detector with a sliding window is used within the optimal period to quantify the settlement ratio in the time series, thereby making the deformation hotspots more obvious.

[0032] like Figure 1 As shown, an embodiment of the present invention provides a method for automatically identifying a sedimentation process by adaptive iterative seasonal decomposition and trend analysis, comprising the following steps: Step S200: preprocessing the time series in the multi-temporal synthetic aperture radar interferometry dataset based on an iterative seasonal decomposition method to remove seasonal components and noise in the time series.

[0033] In this embodiment, the iterative seasonal decomposition method mainly utilizes the local weighted regression scatter point smoothing method (Loess method) to smooth and fit the time series, decompose the time series into seasonal, trend and residual parts, and obtain the optimal period for quantifying the sedimentation ratio in the time series, thereby removing the seasonal components and noise in the time series from these components.

[0034] Specifically, in an implementation of this embodiment, step S200 includes the following steps: Step S201, using a local weighted regression scatter point smoothing algorithm to perform smoothing and fitting processing on the time series, and decomposing the time series into a seasonal component, a trend component and a residual component; Step S202: performing pixel-by-pixel analysis on the time series according to the seasonal component, the trend component and the residual component to determine the optimal period within a preset period interval.

[0035] In this embodiment, in the process of smoothing and fitting the time series using the local weighted regression scatter point smoothing method (Loess method), the expression of the model processed by the local weighted regression scatter point smoothing algorithm is: (1); in, is the length of the time series, is the input time series, , , They are respectively represented as the seasonal component, the trend component and the residual component.

[0036] In the process of seasonal decomposition method processing, the seasonal period input parameter is very important for decomposing time series, but it is usually selected as a fixed value based on experience. In most studies using Sentinel-1A (Sentinel-1A is a high-resolution ground observation data provided by the European Space Agency), the period is fixed to 30 due to the inability to automatically estimate seasonality and collection intervals. However, due to the uneven thermal expansion effect in the study area and the lack of field observation data, the traditional seasonal decomposition method cannot decompose the ideal seasonal component with a fixed period for the entire study area. To solve this problem, an iterative seasonal decomposition method is proposed in this embodiment, which can automatically determine the optimal period from a given range pixel by pixel: (2); in, is the optimal period, is the preset period interval (the preset period interval is a user-defined period interval), is the standard deviation corresponding to the residual component.

[0037] As an example, in this embodiment, a wider cycle range (for example, from 2 to 80 cycles, equivalent to 24 to 960 days) is initially used to determine the optimal cycle. Figure 2 As shown, Figure 2 (a) shows the residual standard deviation and Pearson correlation coefficient obtained from 2 to 80 periods. Figure 2 In (a), three optimal periods (i.e., 2, 28, and 59) are identified, and their corresponding residuals have the smallest standard deviations.

[0038] Figure 2 (b)-(d) are the decomposition results corresponding to the optimal periods 2, 59, and 28 in different intervals. Correspondingly, the decomposition results of the three selected optimal periods are shown in Figure 2 In (b)-(d). Among them, from Figure 2 As can be seen in (b), the optimal period of 2 is too short to capture seasonal changes, and the extracted deformation shows excessive fluctuations. Figure 2 As can be seen in (c), the optimal period of 59 clearly exceeds the actual duration of seasonal changes; Figure 2 As shown in the recorded temperatures in (d), this results in a time series displacement that is too smooth and fails to capture nonlinear deformations.

[0039] In view of the trade-off between smoothness and residual standard deviation, in this embodiment, the search period range of each pixel is narrowed to 20 to 40 (i.e., the preset period interval). Then, 28 is selected as the optimal period for a given timing, such as Figure 2 As shown in (d). In addition, in this embodiment, the correlation between the seasonal component obtained by decomposition and the temperature data recorded by the meteorological station (King's Park Station) is compared. For this optimal period, its Pearson correlation coefficient is 0.831, which proves the effectiveness of the optimal period decomposed by the iterative seasonal decomposition method in this embodiment.

[0040] In this embodiment, the MT-InSAR deformation dataset is decomposed by an iterative seasonal decomposition method, and the ideal seasonal components are decomposed with a fixed period in the entire study area in combination with the smoothness and residual standard deviation of the time series of the corresponding dataset, thereby removing the seasonal components and noise in the time series, thereby ensuring that the subsequent analysis is not affected by seasonal factors and outliers.

[0041] like Figure 1 As shown, an embodiment of the present invention provides a method for automatically identifying a sedimentation process by adaptive iterative seasonal decomposition and trend analysis, comprising the following steps: Step S300, based on the absolute difference settlement automatic identification method with a sliding window, the preprocessed time series is analyzed pixel by pixel to quantify the settlement trend of each settlement point in the study area.

[0042] Generally, a larger downward trend in the time series of the deformation data set represents a greater potential danger. The downward interval in the time series may mean that settlement caused by construction or emergencies occurred during this period, and these phenomena require immediate attention. Therefore, within the optimal period, based on the trend component separated by the iterative seasonal decomposition method, an absolute difference settlement automatic identification method with a sliding window is used in this embodiment to systematically detect the settlement trend in the deformation time series.

[0043] Specifically, in an implementation of this embodiment, step S300 includes the following steps: Step S301, defining the time series as a vector ; Wherein, the vector Contains N displacement values; Step S302, calculating the median of adjacent windows at each time point; Step S303, calculating the absolute difference of each time point according to the median of the adjacent windows, and determining the settlement trend of the corresponding settlement point according to the absolute difference, the central tendency and dispersion index and the preset threshold.

[0044] In this embodiment, when the absolute difference settlement automatic identification method with a sliding window detects the settlement trend in the deformation time series, the trend component separated based on the iterative seasonal decomposition method is analyzed section by section; during the analysis, the time series is defined as a vector, , contains N displacement values, the median of adjacent windows can be expressed as: (3); (4); in, and Represented as the data from the current sliding window and the previous sliding window, respectively. Indicates the window size.

[0045] Specifically, in an implementation of the present embodiment, the absolute difference of each time point is calculated based on the median of adjacent windows, and the settlement trend of the corresponding settlement point is determined based on the absolute difference, the central tendency and dispersion index and the preset threshold. The method includes: calculating the absolute difference of each time point; determining the settlement trend of the corresponding settlement point based on the absolute difference, the central tendency and dispersion index and the preset threshold.

[0046] In this embodiment, in formula (5), the absolute difference is calculated by comparing the medians ( ).if Exceeding the historical interquartile range ( )and If the threshold is greater than , it is considered that there is a sedimentation trend, as shown in formula (6): (5); (6); (7); (8); in, is the absolute difference; are the central tendency and dispersion indices; is the preset threshold.

[0047] In formula (5), represents the absolute difference between the medians, It is an index that describes the central tendency and dispersion of data distribution, covering the range of the middle 50% of the data, that is, the distance between the 25th percentile (first quartile, Q1) and the 75th percentile (third quartile, Q3). Represents a user-defined threshold for determining whether a sedimentation trend exists.

[0048] like Figure 3 As shown, Figure 3 This is the effect diagram of the automatic settlement identification method. Figure 3 The ideal result shown in (a) is preprocessed in this example by extending the beginning and end of the time series to relax the criteria for identifying the sedimentation trend.

[0049] The larger research area determines the diversity of deformation time series, making fixed parameters no longer applicable to all situations. In order to dynamically identify the downward trend in the time series, a multi-stage parameter adjustment strategy is designed in this embodiment, that is, using the standard deviation to distribute and Parameters, as shown in formula (7) and formula (8). A smaller standard deviation indicates a more stable time series; on the contrary, a larger standard deviation indicates a greater volatility of the time series.

[0050] After determining the best period to separate seasonal components based on the iterative seasonal decomposition (STL) method, this paper proposes an absolute difference detector with a sliding window to quantify the settlement proportion within the time series, thereby making the deformation hotspots more visible.

[0051] like Figure 1 As shown, an embodiment of the present invention provides a method for automatically identifying a sedimentation process by adaptive iterative seasonal decomposition and trend analysis, comprising the following steps: Step S400, based on the settlement trend of each settlement point, calculate and output the overall settlement ratio and annual settlement ratio corresponding to each settlement point.

[0052] Specifically, in an implementation of this embodiment, step S400 includes the following steps: Step S401, based on the settlement trend of each settlement point, calculate the ratio of the settlement trend segment corresponding to each settlement point compared to the total time series to obtain the overall settlement ratio corresponding to each settlement point; Step S402, calculating the ratio of the settlement trend segment corresponding to each settlement point to the annual time series, and obtaining the annual settlement ratio corresponding to each settlement point; Step S403, outputting the overall settlement ratio and annual settlement ratio corresponding to each settlement point.

[0053] In this embodiment, after the subsidence identification is performed, the ratio of the subsidence trend segment of each point to the total time series and the annual time series can be determined, which are called the overall subsidence ratio and the annual subsidence ratio, respectively. Subsequently, the relevant spatiotemporal data can be visualized on a map, and the visualization results are as follows: Figure 3 As shown in (b).

[0054] In this embodiment, the iterative seasonal decomposition method and the absolute difference settlement automatic identification method with a sliding window are integrated into the MT-InSAR post-processing module, such as Figure 4 As shown, the processing process of the MT-InSAR post-processing module is as follows: 1) Use iterative seasonal decomposition method to preprocess the deformation time series to remove seasonal components and noise in the time series; In this process, for the deformation time series dataset, first, precise alignment, removal of horizon phase, differential interference and geocoding processing are performed; then, spatial correlation errors and atmospheric delay contributions are removed through phase noise estimation and phase stability detection; finally, the deformation time series dataset is verified for accuracy to obtain the deformation rate and the corresponding time series.

[0055] 2) The absolute difference settlement automatic identification method with a sliding window was used to quantify the settlement trend pixel by pixel, and the results were expressed as overall settlement ratio and annual settlement ratio indicators.

[0056] This process smoothes and fits the deformation rate and the corresponding time series based on the optimal period, decomposing the time series into seasonal components, trend components and residual components; then, the absolute difference sedimentation automatic identification method with a sliding window is used to quantify the sedimentation trend pixel by pixel, and the results are expressed as overall sedimentation ratio and annual sedimentation ratio indicators.

[0057] This embodiment achieves the following technical effects through the above technical solution: This embodiment obtains a multi-phase synthetic aperture radar interferometer data set, and can pre-process the time series in the multi-phase synthetic aperture radar interferometer data set based on the iterative seasonal decomposition method to remove the seasonal components and noise in the time series; and the pre-processed time series is analyzed pixel by pixel through the absolute difference settlement automatic identification method with a sliding window to quantify the settlement trend of each settlement point in the study area; based on the settlement trend of each settlement point, the overall settlement ratio and annual settlement ratio corresponding to each settlement point can be calculated and output. This embodiment automatically decomposes the seasonal components from the time series data through the iterative seasonal trend decomposition method, and quantifies the settlement period in the entire time series through the settlement automatic identification method, thereby realizing the spatiotemporal granularity quantification of the settlement points and improving the detection accuracy of deformation outside the spatial domain.

[0058] Exemplary Devices Based on the above embodiments, the present invention further provides a sedimentation process automatic identification system with adaptive iterative seasonal decomposition and trend analysis, comprising: A data acquisition module, used to acquire a multi-temporal synthetic aperture radar interferometry data set; An iterative seasonal decomposition module, used for preprocessing the time series in the multi-phase synthetic aperture radar interferometry data set based on an iterative seasonal decomposition method to remove seasonal components and noise in the time series; The settlement automatic identification module is used to analyze the preprocessed time series pixel by pixel based on the absolute difference settlement automatic identification method with a sliding window to quantify the settlement trend of each settlement point in the study area; The indicator output module is used to calculate and output the overall settlement ratio and annual settlement ratio corresponding to each settlement point based on the settlement trend of each settlement point.

[0059] This embodiment achieves the following technical effects through the above technical solution: This embodiment obtains a multi-phase synthetic aperture radar interferometer data set, and can pre-process the time series in the multi-phase synthetic aperture radar interferometer data set based on the iterative seasonal decomposition method to remove the seasonal components and noise in the time series; and the pre-processed time series is analyzed pixel by pixel through the absolute difference settlement automatic identification method with a sliding window to quantify the settlement trend of each settlement point in the study area; based on the settlement trend of each settlement point, the overall settlement ratio and annual settlement ratio corresponding to each settlement point can be calculated and output. This embodiment automatically decomposes the seasonal components from the time series data through the iterative seasonal trend decomposition method, and quantifies the settlement period in the entire time series through the settlement automatic identification method, thereby realizing the spatiotemporal granularity quantification of the settlement points and improving the detection accuracy of deformation outside the spatial domain.

[0060] Based on the above embodiment, the present invention further provides a terminal, whose principle block diagram can be as follows: Figure 5 shown.

[0061] The terminal includes: a processor, a memory, an interface, a display screen and a communication module connected via a system bus; wherein the processor of the terminal is used to provide computing and control capabilities; the memory of the terminal includes a storage medium and an internal memory; the storage medium stores an operating system and a computer program; the internal memory provides an environment for the operation of the operating system and the computer program in the storage medium; the interface is used to connect to external devices; the display screen is used to display corresponding information; and the communication module is used to communicate with a cloud server or other devices.

[0062] When the computer program is executed by a processor, it is used to implement the operation of the automatic identification method of sedimentation process with adaptive iterative seasonal decomposition and trend analysis.

[0063] It can be understood by those skilled in the art that Figure 5The principle block diagram shown in the figure is only a block diagram of a partial structure related to the scheme of the present invention, and does not constitute a limitation on the terminal to which the scheme of the present invention is applied. The specific terminal may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0064] In one embodiment, a terminal is provided, which includes: a processor and a memory, wherein the memory stores a program for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis, and when the program for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis is executed by the processor, it is used to implement the operation of the method for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis as described above.

[0065] In one embodiment, a storage medium is provided, wherein the storage medium stores an automatic identification program for sedimentation processes of adaptive iterative seasonal decomposition and trend analysis, and when the automatic identification program for sedimentation processes of adaptive iterative seasonal decomposition and trend analysis is executed by a processor, it is used to implement the operation of the automatic identification method for sedimentation processes of adaptive iterative seasonal decomposition and trend analysis as described above.

[0066] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and volatile memory.

[0067] In summary, the present invention provides an automatic identification method for settlement process with adaptive iterative seasonal decomposition and trend analysis, including: obtaining a multi-phase synthetic aperture radar interferometer data set; preprocessing the time series in the multi-phase synthetic aperture radar interferometer data set based on an iterative seasonal decomposition method to remove seasonal components and noise in the time series; performing pixel-by-pixel analysis on the preprocessed time series based on an absolute difference settlement automatic identification method with a sliding window to quantify the settlement trend of each settlement point in the study area; and calculating and outputting the overall settlement ratio and annual settlement ratio corresponding to each settlement point based on the settlement trend of each settlement point. The present invention automatically decomposes seasonal components from time series data through an iterative seasonal trend decomposition method, and quantifies the settlement period in the entire time series through a settlement automatic identification method, thereby realizing the spatiotemporal granularity quantification of settlement points and improving the detection accuracy of deformation outside the spatial domain.

[0068] It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. An automatic identification method for sedimentation process with adaptive iterative seasonal decomposition and trend analysis, characterized in that: include: Acquire multi-temporal synthetic aperture radar interferometry datasets; Preprocessing the time series in the multi-temporal synthetic aperture radar interferometry dataset based on an iterative seasonal decomposition method to remove seasonal components and noise in the time series; The preprocessed time series were analyzed pixel by pixel based on the absolute difference settlement automatic identification method with sliding window to quantify the settlement trend of each settlement point in the study area. Based on the settlement trend of each settlement point, the overall settlement ratio and annual settlement ratio corresponding to each settlement point are calculated and output.

2. The method for automatic identification of sedimentation process by adaptive iterative seasonal decomposition and trend analysis according to claim 1 is characterized in that: The method of preprocessing the time series in the multi-temporal synthetic aperture radar interference data set based on the iterative seasonal decomposition method includes: Using a local weighted regression scatter point smoothing algorithm to smooth and fit the time series, decomposing the time series into a seasonal component, a trend component and a residual component; According to the seasonal component, the trend component and the residual component, the time series is analyzed pixel by pixel to determine the optimal period within the preset period interval: ； in, is the optimal period, is the preset period interval, is the standard deviation corresponding to the residual component.

3. The method for automatic identification of sedimentation process by adaptive iterative seasonal decomposition and trend analysis according to claim 2 is characterized in that: The expression of the local weighted regression scatter point smoothing algorithm is: ； in, is the length of the time series, is the input time series, , , They are respectively represented as the seasonal component, the trend component and the residual component.

4. The method for automatic identification of sedimentation process by adaptive iterative seasonal decomposition and trend analysis according to claim 1, characterized in that: The automatic identification method of settlement based on absolute difference with sliding window analyzes the preprocessed time series pixel by pixel to quantify the settlement trend of each settlement point in the study area, including: Define the time series as a vector ; Wherein, the vector Contains N displacement values; Calculate the median of adjacent windows at each time point; The absolute difference of each time point is calculated according to the median of the adjacent windows, and the settlement trend of the corresponding settlement point is determined according to the absolute difference, the central tendency and dispersion index and the preset threshold.

5. The method for automatic identification of sedimentation process by adaptive iterative seasonal decomposition and trend analysis according to claim 4 is characterized in that: The median of the adjacent windows is expressed as: ； ； in, and are the data obtained from the current sliding window and the previous sliding window respectively, Indicates the window size.

6. The method for automatic identification of sedimentation process by adaptive iterative seasonal decomposition and trend analysis according to claim 4, characterized in that: The absolute difference of each time point is obtained by calculating the median of adjacent windows, and the settlement trend of the corresponding settlement point is determined according to the absolute difference, the central tendency and dispersion index and the preset threshold, including: Calculate the absolute difference at each time point: ； According to the absolute difference, the central tendency and dispersion index and the preset threshold, the settlement trend of the corresponding settlement point is determined using the following conditions: ； ； ； in, is the absolute difference; are the central tendency and dispersion indices; is the preset threshold.

7. The method for automatic identification of sedimentation process by adaptive iterative seasonal decomposition and trend analysis according to claim 1, characterized in that: The calculation and output of the overall settlement ratio and annual settlement ratio corresponding to each settlement point based on the settlement trend of each settlement point includes: Based on the settlement trend of each settlement point, the ratio of the settlement trend segment corresponding to each settlement point to the total time series is calculated to obtain the overall settlement ratio corresponding to each settlement point; Calculate the ratio of the subsidence trend fragment corresponding to each subsidence point compared to the annual time series to obtain the annual subsidence ratio corresponding to each subsidence point; Output the overall settlement ratio and annual settlement ratio corresponding to each settlement point.

8. An automatic identification system for sedimentation process with adaptive iterative seasonal decomposition and trend analysis, characterized in that: include: A data acquisition module, used to acquire a multi-temporal synthetic aperture radar interferometry data set; An iterative seasonal decomposition module, used for preprocessing the time series in the multi-phase synthetic aperture radar interferometry data set based on an iterative seasonal decomposition method to remove seasonal components and noise in the time series; The settlement automatic identification module is used to analyze the preprocessed time series pixel by pixel based on the absolute difference settlement automatic identification method with a sliding window to quantify the settlement trend of each settlement point in the study area; The indicator output module is used to calculate and output the overall settlement ratio and annual settlement ratio corresponding to each settlement point based on the settlement trend of each settlement point.

9. A terminal, characterized in that: include: A processor and a memory, wherein the memory stores a program for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis, and when the program for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis is executed by the processor, it is used to implement the operation of the method for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis, which, when executed by a processor, is used to implement the operation of the method for automatic identification of sedimentation processes of adaptive iterative seasonal decomposition and trend analysis as described in any one of claims 1 to 7.

Images