A remote sensing detection method and system for sand dune movement and rapid change, electronic equipment, and storage medium

Abstract

The present invention provides a remote sensing detection method for sand dune movement and speed variation, which can be applied to the field of remote sensing image processing technology. The method includes: extracting dune edges from dune remote sensing observation time-series images to obtain a dune edge image sequence; preprocessing the dune edge image sequence to obtain a sand ridge line image sequence; constructing a velocity vector field from the sand ridge line image data of two adjacent frames in the sand ridge line image sequence using a dense optical flow method to obtain a space-time cube array of the sand ridge line movement velocity field; performing pixel-by-pixel time-series analysis on the space-time cube array of the sand ridge line movement velocity field using an exponentially weighted moving average control chart to obtain a time-series rate anomaly judgment result; and obtaining a dune movement and speed variation detection result by superimposing the time-series rate anomaly judgment result onto the dune remote sensing observation time-series images for identification. The present invention also provides a sand dune movement and speed variation remote sensing detection system, electronic equipment, and storage medium.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing image processing, and in particular to a remote sensing detection method and system for sand dune movement and rapid change, electronic equipment, and storage medium. Background Art

[0002] Dune movement velocity refers to the dramatic change in dune velocity during dune movement. Currently, dune movement monitoring relies primarily on ground-based surveying techniques and remote sensing imaging. Ground-based surveying techniques primarily employ methods such as the plunger method, total stations, theodolites, and real-time dynamic differential GPS. While these techniques offer high accuracy, they suffer from the harsh field environment and limited efficiency, resulting in significant workload, high costs, and a limited scope. Remote sensing monitoring, on the other hand, primarily utilizes satellite remote sensing imagery and drone aerial imagery. These techniques offer advantages such as rapid acquisition of dune movement parameters, a wide range, and the ability to monitor a large number of dunes. However, dune movement measurement is primarily based on visual interpretation, using geographic information processing software to manually outline dune images and select characteristic points (e.g., five-point measurement or centroid measurement) to reveal macroscopic dune movement.

[0003] Existing techniques for extracting dune movement parameters from remote sensing images primarily rely on visual interpretation and manual calibration of feature points using geographic information processing software. Most studies only select a few representative feature points to describe the macroscopic movement of dunes, making it difficult to characterize the detailed changes in the movement of mobile dunes. Detecting more detailed and rapid changes in dune movement and morphology requires more time-series satellite imagery with shorter time intervals. Manually annotating dune movement measurements is inherently time-consuming. Furthermore, existing techniques lack a solution for detecting rapid changes in dune movement. Summary of the Invention

[0004] In view of the above problems, the present invention provides a method and system for remote sensing detection of sand dune movement and rapid change, an electronic device, and a storage medium, which are used to solve at least one of the problems in the prior art.

[0005] According to a first aspect of the present invention, a method for remote sensing detection of sand dune movement and rapid change is provided, comprising:

[0006] The deep learning method is used to extract the dune edge from the dune remote sensing observation time series image to obtain the dune edge image sequence, and the dune edge image sequence is preprocessed to obtain the sand ridge line image sequence;

[0007] The dense optical flow method is used to construct the velocity vector field of the sand ridge line image data of two adjacent frames in the sand ridge line image sequence, and the space-time cube array of the sand ridge line moving velocity field is obtained;

[0008] The exponentially weighted moving average control chart is used to perform pixel-by-pixel time series analysis on the space-time cube array of the sand ridge line movement velocity field to obtain the time series rate anomaly judgment results. The judgment results of the time series rate anomaly points are then superimposed on the dune remote sensing observation time series image for position-time identification to obtain the dune movement speed change detection results.

[0009] According to an embodiment of the present invention, the preprocessing of the dune edge image sequence to obtain the sand ridge line image sequence includes:

[0010] The erosion filtering method is used to filter the noise of the dune edge image sequence to obtain the denoised dune edge image sequence;

[0011] Based on the dune edge image sequence, the denoised dune edge image sequence is resampled using the bilinear interpolation method to obtain the sand ridge line image sequence.

[0012] According to an embodiment of the present invention, the above-mentioned dense optical flow method is used to construct the velocity vector field of the sand ridge line image data of two adjacent frames in the sand ridge line image sequence to obtain the sand ridge line moving velocity field space-time cube array including:

[0013] The Farneback dense optical flow model is used to calculate the horizontal and vertical displacement of each pixel in the sand ridge image data of two adjacent frames to obtain the time series of the displacement vector field.

[0014] Splicing the time series of the displacement vector fields of all two adjacent frames of sand ridge line image data to obtain a splicing result;

[0015] Based on the splicing results, the displacement vector of each two adjacent frames of sand ridge line image data and the time interval of each two adjacent frames of sand ridge line image data are calculated to obtain the space-time cube array of the sand ridge line movement velocity field.

[0016] According to an embodiment of the present invention, the above-mentioned sand ridge line movement velocity field space-time cube array includes time axis data, image horizontal axis data, image vertical axis data, pixel horizontal axis displacement data, pixel vertical axis displacement data, pixel horizontal axis velocity data and pixel vertical axis velocity data. The sand ridge line movement velocity field space-time cube array is used to characterize the velocity information of pixels in the sand ridge line image sequence.

[0017] According to an embodiment of the present invention, the above-mentioned use of the exponentially weighted moving average control chart to perform pixel-by-pixel time series analysis on the space-time cube array of the sand ridge line movement velocity field, and the time series rate anomaly point judgment results include:

[0018] Extract the displacement sequence and movement rate of each image coordinate on the sand ridge line from the spatiotemporal cube array of the sand ridge line movement velocity field to obtain the spatiotemporal sequence of the sand ridge line movement rate;

[0019] The exponentially weighted moving average control chart is used to identify the time series rate anomaly points of the sand ridge line movement rate time and space series, and the time series rate anomaly point judgment results are obtained.

[0020] According to an embodiment of the present invention, the above-mentioned extraction of the displacement sequence and movement rate of each image coordinate on the sand ridge line from the spatiotemporal cube array of the sand ridge line movement velocity field to obtain the spatiotemporal sequence of the sand ridge line movement rate includes:

[0021] The space-time cube array of the sand ridge line moving velocity field at each moment is marked pixel by pixel to obtain the sand ridge line pixel marking points at each moment;

[0022] The horizontal velocity vector and the vertical velocity vector corresponding to the pixel marker point of the sand ridge line at each moment are added together and then modulo operation is performed to obtain the spatiotemporal sequence of the sand ridge line movement rate.

[0023] According to an embodiment of the present invention, the above-mentioned use of the exponentially weighted moving average control chart to perform time series rate anomaly point discrimination on the spatiotemporal sequence of the sand ridge line movement rate obtains the time series rate anomaly point discrimination result including:

[0024] The recursive formula of the statistic of the exponentially weighted moving average control chart is constructed using the spatiotemporal sequence of the sand ridge moving rate. The recursive formula of the statistic is expanded into the form of the exponentially weighted average of all time rate sequences to obtain the expanded form of the statistic.

[0025] Based on the maximum likelihood estimation method, the standard deviation estimate of the statistic is constructed, and the standard deviation estimate of the statistic is used to calculate the expanded form of the statistic to obtain the upper and lower control limits and center line of the exponentially weighted moving average control chart;

[0026] The upper and lower control limits and center line of the exponentially weighted moving average control chart are used to identify the time series rate anomalies of the sand ridge line movement rate on a pixel-by-pixel basis. The pixels whose time series rate exceeds the upper and lower control limits of the exponentially weighted moving average control chart are identified as time series rate anomalies.

[0027] A second aspect of the present invention provides a remote sensing detection system for sand dune movement and rapid change, comprising:

[0028] The remote sensing image preprocessing module is used to extract the dune edge from the dune remote sensing observation time series image using a deep learning method to obtain a dune edge image sequence, and preprocess the dune edge image sequence to obtain a sand ridge line image sequence;

[0029] The velocity vector field construction module is used to construct the velocity vector field of the sand ridge line image data of two adjacent frames in the sand ridge line image sequence using the dense optical flow method, and obtain the spatiotemporal cube array of the sand ridge line moving velocity field;

[0030] The dune rapid change detection module is used to perform pixel-by-pixel time series analysis on the space-time cube array of the sand ridge line movement velocity field using the exponentially weighted moving average control chart to obtain the time series rate anomaly judgment result. The judgment result of the time series rate anomaly point is superimposed on the dune remote sensing observation time series image to perform position-time identification and obtain the dune movement rapid change detection result.

[0031] A third aspect of the present invention provides an electronic device, comprising: one or more processors; a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0032] The fourth aspect of the present invention further provides a computer-readable storage medium having a computer program or instructions stored thereon, which implements the steps of the above method when the computer program or instructions are executed by a processor.

[0033] The method for remote sensing detection of rapid changes in sand dune movement provided by the present invention constructs the dune movement state through the dense optical flow method, can realize the automatic estimation of sand dune movement in remote sensing images, and can estimate sand dune movement information from a microscopic perspective; in addition, the method for remote sensing detection of rapid changes in sand dune movement provided by the present invention utilizes a simple and efficient detection method based on the optical flow method and the exponentially weighted moving average control map, which can effectively detect rapid changes in sand dune movement in remote sensing image sequences. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The above contents and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0035] Figure 1 2. This is a diagram showing an application scenario of a remote sensing detection method for sand dune movement and rapid change according to an embodiment of the present invention;

[0036] Figure 2 is a flow chart of a remote sensing detection method for sand dune movement and rapid change according to an embodiment of the present invention;

[0037] Figure 3 2. It is a data processing framework diagram of a method for detecting rapid changes in sand dune movement according to an embodiment of the present invention;

[0038] Figure 4 is a schematic diagram of an optical flow field established by two adjacent frames according to an embodiment of the present invention;

[0039] Figure 5 is a schematic diagram of a sand dune movement speed change detection result according to an embodiment of the present invention;

[0040] Figure 6 is a schematic diagram of a basic form of a quality control chart according to an embodiment of the present invention;

[0041] Figure 7 2 is a schematic structural diagram of a remote sensing detection system for sand dune movement and rapid change according to an embodiment of the present invention;

[0042] Figure 8 1 is a schematic diagram of a real scene of two deserts, Sand1 and Sand2, and a schematic diagram of corresponding sand ridge line extraction according to an embodiment of the present invention;

[0043] Figure 9 are an input image and a corresponding output image for Sand1 rapid change detection according to an embodiment of the present invention;

[0044] Figure 10 2020-2021 is a schematic diagram of the rapid change detection results of Sand1 according to an embodiment of the present invention;

[0045] Figure 11 is a schematic diagram of the rapid change detection results of Sand1 between 2010 and 2014 according to an embodiment of the present invention;

[0046] Figure 12 are an input image and a corresponding output image for Sand2 speed change detection according to an embodiment of the present invention;

[0047] Figure 13 is a schematic diagram of Sand2 rapid change detection results from 2010 to 2014 according to an embodiment of the present invention;

[0048] Figure 14 4 is a block diagram of an electronic device suitable for implementing a remote sensing detection method for sand dune movement speed change according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concept of the present invention.

[0050] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise", "include", etc. used herein indicate the presence of the features, steps, operations and / or components, but do not exclude the presence or addition of one or more other features, steps, operations or components.

[0051] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0052] When expressions such as "at least one of A, B, and C, etc." are used, they should generally be interpreted in accordance with the meaning commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0053] Figure 1 2 is a diagram illustrating an application scenario of a remote sensing detection method for rapid changes in sand dune movement according to an embodiment of the present invention.

[0054] like Figure 1 As shown, the application scenario 100 according to this embodiment may include the field of remote sensing image processing technology. A network 104 is used as a medium for providing a communication link between a first terminal device 101, a second terminal device 102, a third terminal device 103, and a server 105. The network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0055] A user may use a first terminal device 101, a second terminal device 102, or a third terminal device 103 to interact with a server 105 via a network 104 to receive or send messages, etc. Various communication client applications may be installed on the first terminal device 101, the second terminal device 102, or the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (for example only).

[0056] The first terminal device 101 , the second terminal device 102 , and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0057] The server 105 may be a server that provides various services, such as a background management server (for example only) that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process received data such as user requests, and feed back processing results (e.g., web pages, information, or data obtained or generated based on user requests) to the terminal devices.

[0058] It should be noted that the remote sensing detection method for sand dune movement speed change provided in the embodiment of the present invention can generally be executed by the server 105. Accordingly, the remote sensing detection system for sand dune movement speed change provided in the embodiment of the present invention can generally be set in the server 105. The remote sensing detection method for sand dune movement speed change provided in the embodiment of the present invention can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the remote sensing detection system for sand dune movement speed change provided in the embodiment of the present invention can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0059] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0060] The following will be based on Figure 1 The scene described by Figures 2 to 6 The method for remote sensing detection of sand dune movement speed change according to the disclosed embodiment is described in detail.

[0061] Figure 2 4 is a flow chart of a method for remote sensing detection of sand dune movement speed changes according to an embodiment of the present invention.

[0062] like Figure 2 As shown, the above-mentioned remote sensing detection method for sand dune movement speed change includes operations S210 to S230.

[0063] In operation S210 , a deep learning method is used to extract dune edges from dune remote sensing observation time series images to obtain a dune edge image sequence, and the dune edge image sequence is preprocessed to obtain a sand ridge line image sequence.

[0064] Since dune remote sensing observation time series images are large in size, they need to be processed before dune movement and rapid change detection: first, the dune edges are extracted using deep learning methods; then the extracted dune edge image sequence is preprocessed, such as filtering and downsampling.

[0065] In operation S220 , a velocity vector field is constructed for the sand ridge line image data of two adjacent frames in the sand ridge line image sequence using a dense optical flow method to obtain a spatiotemporal cube array of the sand ridge line moving velocity field.

[0066] The above dense optical flow method may be the Farneback dense optical flow method.

[0067] In operation S230, the exponentially weighted moving average control chart is used to perform pixel-by-pixel time series analysis on the space-time cube array of the sand ridge line movement velocity field to obtain the time series rate anomaly point judgment result, and the time series rate anomaly point judgment result is superimposed on the dune remote sensing observation time series image to perform position-time identification to obtain the dune movement speed change detection result.

[0068] The method for remote sensing detection of rapid changes in sand dune movement provided by the present invention constructs the dune movement state through the dense optical flow method, can realize the automatic estimation of sand dune movement in remote sensing images, and can estimate sand dune movement information from a microscopic perspective; in addition, the method for remote sensing detection of rapid changes in sand dune movement provided by the present invention utilizes a simple and efficient detection method based on the optical flow method and the exponentially weighted moving average control map, which can effectively detect rapid changes in sand dune movement in remote sensing image sequences.

[0069] The following is a specific implementation method and combined with the attached Figures 3-5 The above-mentioned sand dune movement and rapid change detection method provided by the present invention is further described in detail.

[0070] Figure 3 4 is a data processing framework diagram of a method for detecting rapid changes in sand dune movement according to an embodiment of the present invention.

[0071] Figure 4 Schematic diagram of an optical flow field established by two adjacent frames according to an embodiment of the present invention.

[0072] Figure 5 FIG. 4 is a schematic diagram of a sand dune movement speed change detection result according to an embodiment of the present invention.

[0073] Using a desert as the research object, the optical flow method was used to construct a velocity field time series from the extracted dune ridge data. Time series anomaly detection was then used to detect velocity anomalies, thereby enabling automated detection and analysis of dune movement velocity variations. First, dune images and extracted sand ridges were acquired for a specific time period, for example, from 2010 to 2022. Dense optical flow was then used to model the optical flow variations between ridge image frames to describe the dune motion. Finally, an exponentially weighted moving average (EWMA) control chart was used to detect velocity anomalies and complete the monitoring of velocity variations. The innovative approach provided by this invention lies in combining dense optical flow with EWMA control charts to achieve automated detection of dune movement velocity variations at a microscopic level. This method can quickly and accurately identify sudden accelerations and decelerations in dune movement, as well as the appearance and disappearance of sand ridges, providing objective, efficient, and repeatable analysis results.

[0074] like Figure 3 As shown, Figure 3The left side of the figure shows the optical flow field establishment process, and the right side shows the velocity field anomaly detection process. First, a sequence of dune edge images extracted from a sequence of desert remote sensing observation images using deep learning is used as input. Currently, the algorithm for directly extracting sand ridge lines is not perfect, and it mainly extracts sand dune edges. Due to the small difference between sand dunes and the desert background, the extraction process produces many noise points and discontinuities. Therefore, the corrosion filtering method is used to remove some noise points, making the extracted sand ridge lines smoother and with less background noise. In addition, since remote sensing images often have too large pixel widths, such as Figure 3 The pixel width of the input image is 2357 2268, and the commonly used optical flow method is often developed based on computer vision images, and its applicable image size is often small. After experiments, the image is resampled to 512 512, the discontinuity points of the dune edge extraction are reduced, the sand ridge lines are clearer and smoother, and the velocity field construction effect is optimized. At the same time, the resampling operation can reduce the computational complexity of subsequent time series analysis, greatly improving the efficiency of rapid change detection. Finally, the preprocessed sand ridge line image sequence is input into the Farneback dense optical flow method to establish the optical flow field. The Farneback dense optical flow method is based on the optical flow field modeling between two frames, so the input is a pair of sand ridge line images, such as Figure 4 The optical flow field established by the adjacent double frames shown in Figure 4 The base map is the sand ridge line image of the next frame.

[0075] During the optical flow field construction process, the present invention primarily analyzes the spatial domain, studying the spatial variations of optical flow between two frames. Specifically, the optical flow field is repeatedly constructed for each pair of frames in chronological order and superimposed to form a space-time cube array of the sand ridge velocity field. This space-time cube array of the sand ridge velocity field is then transferred to the temporal domain, where analysis is performed step-by-step for each pixel along the sand ridge and at each time point.

[0076] The present invention provides a more robust EWMA control chart time series anomaly detection method that can effectively detect the speed change points of sand dune movement, such as Figure 5 As shown, the detection results are displayed in the base map of the original image, corresponding to Figure 4 The optical flow field and sand ridge extraction images in the image are shown in Figure 2. The method provided by the present invention has the advantages of simplicity and efficiency, high detection accuracy, no need for large amounts of training data, and strong robustness. It can quickly process large-scale time series data and demonstrates excellent detection capabilities when capturing changes in the speed of sand dunes. It can effectively identify anomalies even in the presence of small changes. Compared with some complex machine learning models, the method provided by the present invention does not rely on large-scale annotated data, reducing data preparation and training costs. Furthermore, the method provided by the present invention can still stably output accurate results in the presence of noise or imperfect data.

[0077] According to an embodiment of the present invention, the above-mentioned preprocessing of the dune edge image sequence to obtain the sand ridge line image sequence includes: obtaining dune remote sensing observation time series images, extracting the dune edges from the dune remote sensing observation time series images to obtain a dune edge image sequence; using a corrosion filtering method to filter the dune edge image sequence to obtain a denoised dune edge image sequence; based on the dune edge image sequence, using a bilinear interpolation method to resample the denoised dune edge image sequence to obtain a sand ridge line image sequence.

[0078] In this example, a multi-year dune image sequence of the same area was obtained from an open source database (such as Google Earth), and a dune edge image sequence was extracted using a deep learning method. Since the extracted dune edge image sequence has many noise points and discontinuities, which are not conducive to the subsequent motion field estimation, the noise is filtered out using the erosion filter method. At the same time, to solve the problems of optical flow field extraction effect and efficiency caused by dune edge discontinuities and excessive image pixel width, the denoised dune edge image sequence is resampled using the bilinear interpolation method. After multiple experiments, the image is resampled to 512 The 512 size is more conducive to sand ridge extraction and subsequent optical flow field processing, and finally a sand ridge image sequence is obtained.

[0079] According to an embodiment of the present invention, the above-mentioned construction of the velocity vector field of the sand ridge line image data of two adjacent frames in the sand ridge line image sequence using the dense optical flow method to obtain the sand ridge line movement velocity field spatiotemporal cube array includes: using the Farneback dense optical flow model to calculate the horizontal axis displacement and the vertical axis displacement of each pixel in the sand ridge line image data of two adjacent frames to obtain a time series of the displacement vector field; splicing the time series of the displacement vector fields of the sand ridge line image data of all two adjacent frames to obtain a splicing result; based on the splicing result, calculating the displacement vector of each two adjacent frames of sand ridge line image data and the time interval of each two adjacent frames of sand ridge line image data to obtain a sand ridge line movement velocity field spatiotemporal cube array.

[0080] According to an embodiment of the present invention, the above-mentioned sand ridge line movement velocity field space-time cube array includes time axis data, image horizontal axis data, image vertical axis data, pixel horizontal axis displacement data, pixel vertical axis displacement data, pixel horizontal axis velocity data and pixel vertical axis velocity data. The sand ridge line movement velocity field space-time cube array is used to characterize the velocity information of pixels in the sand ridge line image sequence.

[0081] Optical flow refers to the instantaneous speed of pixel movement of a moving object in space on the observation imaging plane. The optical flow law is a method that uses the changes in pixels in the time domain in an image sequence and the correlation between adjacent frames to find the correspondence between the previous frame and the current frame, thereby calculating the motion information of objects between adjacent frames.

[0082] The implementation of an optical flow algorithm requires two basic assumptions: the brightness of the target remains constant when it moves between two consecutive frames; and the target's movement is minimal, meaning it undergoes a small displacement. The fundamental constraint equations for optical flow are based on the assumption of grayscale conservation, which means that the grayscale of the target pixel remains constant after movement.

[0083] The dense optical flow method is an image registration method that performs point-by-point matching on an image or a specified area. Unlike sparse optical flow methods (such as the LK algorithm), which usually require a specified set of points to track, it needs to calculate the offset of all points on the image to form a dense optical flow field, and then use this dense optical flow field to perform pixel-level image registration. The HS algorithm and most optical flow methods based on area matching fall into the category of dense optical flow. The Farneback algorithm is a dense optical flow method based on polynomial expansion. The accuracy of the Farneback optical flow method is higher than that of the sparse optical flow method LK algorithm, and it has a faster calculation speed than the similar dense optical flow method HS algorithm. Therefore, the present invention uses the Farneback dense optical flow method to process the image sequence involved.

[0084] In a specific embodiment of the present invention, the Farneback optical flow method is used to estimate the optical flow of two adjacent sand ridge images in a sand ridge image sequence, based on their temporal order. The Farneback optical flow model parameters are designed as follows: a three-layer pyramid layer with a scale factor of 0.5 is constructed, matching is performed using a mean window of size 15, the algorithm is iterated five times per pyramid level, and the pixel neighborhood size searched is 1.2 with a standard deviation of 0. The model outputs the horizontal and vertical displacements of each pixel in two adjacent frames. By inputting each two-frame sequence into the model, a time series of displacement vector fields is generated. The resulting outputs are concatenated, and the displacement vectors are divided by the two-frame time interval to obtain the velocity. This ultimately yields a space-time cube array representing the velocity of each pixel in the sand ridge image sequence. The dimensions of the space-time cube array include: time axis t, image horizontal axis x, image vertical axis y, pixel horizontal displacement sx, pixel vertical displacement sy, pixel horizontal velocity vx, and pixel vertical velocity vy.

[0085] According to an embodiment of the present invention, the above-mentioned use of the exponentially weighted moving average control chart to perform pixel-by-pixel time series analysis on the space-time cube array of the sand ridge line movement velocity field to obtain the time series rate anomaly point judgment result includes: extracting the displacement sequence and movement rate of each image coordinate on the sand ridge line from the space-time cube array of the sand ridge line movement velocity field to obtain the sand ridge line movement rate space-time sequence; using the exponentially weighted moving average control chart to perform time series rate anomaly point judgment on the sand ridge line movement rate space-time sequence to obtain the time series rate anomaly point judgment result.

[0086] According to an embodiment of the present invention, the above-mentioned extraction of the displacement sequence and movement rate of each image coordinate on the sand ridge line from the spatiotemporal cube array of the sand ridge line movement velocity field to obtain the spatiotemporal sequence of the sand ridge line movement rate includes: pixel-by-pixel marking of the spatiotemporal cube array of the sand ridge line movement velocity field at each moment to obtain the sand ridge line pixel marking point at each moment; adding the horizontal axis velocity vector and the vertical axis velocity vector corresponding to the sand ridge line pixel marking point at each moment and performing a modulo operation to obtain the sand ridge line movement rate spatiotemporal sequence.

[0087] Extract the displacement sequence of each image coordinate on the sand ridge line from the space-time cube array of the sand ridge line movement velocity field, and extract the movement rate of each image coordinate on the sand ridge line at the same time to obtain the sand ridge line movement rate time series data. , the EWMA control chart is used to identify the rate anomaly points of the sand ridge line movement rate time series data, and the rate anomaly point identification results are superimposed on the dune remote sensing observation time series image for identification, completing the dune movement speed change detection.

[0088] In the embodiment of the present invention, the displacement sequence of each image coordinate on the sand ridge line is extracted. The displacement sequence represents the movement of each image coordinate on the sand ridge line. The sand ridge line in the sand ridge line image at the first moment is marked pixel by pixel to obtain the sand ridge line pixel marking points (a total of pixel markers, marked as For the lth sand ridge pixel marker point , the displacement sequence is expressed as , the relationship between two adjacent moments in the displacement sequence is and ); calculate the pixel marker points of the sand ridge line exist time, in Pixel position rate ,in The value is calculated by transforming the velocity vector Convert to rate scalar , that is, to add the vertical and horizontal velocity vectors of the pixels perpendicular to each other and extract the modulus value. Pixel position If the value is 0, Search pixel coordinates within a 3×3 window The pixel coordinate with the largest value is replaced , and obtain the time series data of sand ridge line movement rate Use EWMA (Exponentially Weighted Moving Average) control chart to analyze the time series data of sand ridge movement rate. Perform outlier detection to obtain time series data of rapid change points of sand ridge movement rate , including the pixel markers of the sand ridge line at the point of rapid rate change , pixel coordinates and time .

[0089] According to an embodiment of the present invention, the above-mentioned use of the exponentially weighted moving average control chart to identify the time series rate anomaly points of the sand ridge line movement rate spatiotemporal sequence, and obtaining the time series rate anomaly point identification result includes: using the sand ridge line movement rate spatiotemporal sequence to construct a recursive formula for the statistic of the exponentially weighted moving average control chart, and expanding the recursive formula of the statistic into the form of an exponentially weighted average of all time rate sequences to obtain an expanded form of the statistic; based on the maximum likelihood estimation method, constructing a standard deviation estimate of the statistic, and using the standard deviation estimate of the statistic to calculate the expanded form of the statistic to obtain the upper and lower control limits and center line of the exponentially weighted moving average control chart; using the upper and lower control limits and center line of the exponentially weighted moving average control chart to perform pixel-by-pixel time series rate anomaly identification on the sand ridge line movement rate spatiotemporal sequence, and judging pixels whose time series rates exceed the upper and lower control limits of the exponentially weighted moving average control chart as time series rate anomalies.

[0090] To facilitate the use of the exponentially weighted moving average control chart in the present invention, the exponentially weighted moving average control chart is described below.

[0091] Quality control charts demonstrate that every data analysis and monitoring method is subject to variation and is affected by time and space. Even under ideal conditions, a set of analytical results will contain a certain amount of random error. When a result exceeds the allowable range of random error, mathematical statistics can be used to determine that the result is abnormal and unreliable.

[0092] Figure 6 is a schematic diagram of a basic form of a quality control chart according to an embodiment of the present invention.

[0093] A quality control chart is a chart used to analyze the changes in a process over time. It is a chart with control limits used to analyze and determine whether a sequence is in a stable state. The basic form of a quality control chart is as follows: Figure 6As shown in the figure. The horizontal axis represents time points, and the vertical axis represents a certain process statistic. A quality control chart has three lines: the upper control limit (UCL), the center line (CL), and the lower control limit (LCL). The center line represents the average level of the process statistic. The area between the upper and lower control limits [LCL, UCL] is the acceptance region, while the area outside this range is called the rejection region. By comparing the process statistic with the upper and lower control limits, if the statistic falls within the rejection region, the process is considered out of control, indicating an abnormality.

[0094] The quality control chart is an application of statistical significance testing in process stability control. Its principle is statistical hypothesis testing. Each quality control chart is a statistical hypothesis test. Under normal conditions, the process is relatively stable. When only random factors are at work, according to the central limit theorem, the distribution of all these random errors is normal or approximately normal. Specifically, the output of a process It has its own statistical characteristics, and the stability of the process can be monitored by judging whether these statistical characteristics are stable. Select a statistic (Statistics) , and select a significance level (such as 0.001), then There is a corresponding rejection region. Under normal conditions, that is, when the process is stable, The probability of falling within the rejection region is small Since it is generally believed that low-probability events will not occur, if If the value of the statistical characteristic of the process falls into the rejection region, it is considered that the value is abnormal, and the process is judged to be abnormal at the corresponding moment.

[0095] The upper / lower control limits (UCL / LCL) and center line (CL) of the quality control chart are determined by formula (1):

[0096] (1),

[0097] in, and The statistics are The mathematical expectation and variance of is a constant representing the multiple of standard deviation. , the rejection region is defined as According to the normal distribution property, the statistic falls outside The probability of being within the range is 99.73%, and the probability of being outside the range is only 0.27% (that is, the significance level ), the probability of falling outside the range on one side is only 0.135%, which is a low probability event. For non-normal distribution, the statistic falls in The probability outside the range is also close to zero. According to the statistics of the process Relative to the output of the process There are three main types of quality control charts, depending on their construction methods: the Shewhart control chart, the Cumulative Sum (CUSUM) control chart, and the Exponentially Weighted Moving Average (EWMA) control chart. The exponentially weighted moving average control chart used in this paper is the EWMA control chart.

[0098] Time rate series The recursive form of the EWMA control chart statistic is shown in formula (2):

[0099] (2),

[0100] in is the smoothing coefficient, which is a constant and , Generally, the mathematical expectation of all time rate sequences is taken , It can be expanded to express the exponential weighted average of all time rate series, as shown in formula (3):

[0101] (3).

[0102] Assuming that the time rate sequence is a random variable sequence with the same distribution at different times, the estimated value of the mathematical expectation of the time rate sequence is , the estimated variance is

[0103] . It is worth noting that in actual situations, the time rate series may contain outliers, and these outliers will cause deviations in the estimation of the mean and cause the maximum likelihood estimate of the standard deviation to be too large. Therefore, if it is not adjusted, the effectiveness of the anomaly detection method will be reduced. In order to enhance the effectiveness of the test, the present invention adopts a robust mean and standard deviation estimation method to replace the conventional maximum likelihood estimation method. A robust mean is to calculate the average value after sorting the data and removing the maximum and minimum 5% of the data at both ends; a robust standard deviation estimation method is shown in formula (4):

[0104] (4).

[0105] right Expand formula (2) and calculate the mathematical expectation and variance on both sides of the equal sign. After sorting, the mathematical expectation and variance of the EWMA control chart statistics are shown in formula (5):

[0106] (5),

[0107] Where n is the length of the time rate sequence. Finally, the upper / lower control limits (UCL / LCL) and center line (CL) of the EWMA control chart are shown in formula (6):

[0108] (6),

[0109] in is a pre-set constant indicating that the control limit is the mean plus / minus multiples of the standard deviation. 、 The dune movement time rate series is subjected to the upper / lower control limit identification of the EWMA control chart pixel by pixel. If it exceeds the limit, it is judged as a rate anomaly, that is, the pixel is judged to have a rapid change in this time interval, and the point is marked in the time series dune image.

[0110] The above-mentioned dune movement and rapid change inspection method provided by the present invention has the following advantages compared with the existing technology: First, the existing technology usually uses visual interpretation and manual marking to measure the dune movement distance in remote sensing images using geographic information processing software. The present invention constructs the dune movement state through the optical flow method, which can realize the automatic estimation of dune movement in remote sensing images; second, traditional dune movement measurement methods such as the five-point measurement method and the centroid measurement method are mostly used to construct the macroscopic movement of dunes. The present invention can estimate dune movement information from a microscopic perspective through the dense optical flow method; in addition, there are currently few methods for detecting rapid changes in dune movement. The present invention provides a simple and efficient detection method based on the optical flow method and the EWMA control chart, which can effectively detect rapid changes in dune movement in remote sensing image sequences.

[0111] Figure 7 2 is a schematic structural diagram of a remote sensing detection system for sand dune movement and rapid change according to an embodiment of the present invention.

[0112] like Figure 7 As shown, the above-mentioned sand dune movement and rapid change remote sensing detection system 700 includes a remote sensing image preprocessing module 710 , a velocity vector field construction module 720 and a sand dune rapid change detection module 730 .

[0113] The remote sensing image preprocessing module 710 is used to extract the dune edge from the dune remote sensing observation time series image using a deep learning method to obtain a dune edge image sequence, and preprocess the dune edge image sequence to obtain a sand ridge line image sequence; in one embodiment, the remote sensing image preprocessing module 710 can be used to perform the operation S210 described above, which will not be repeated here.

[0114] The velocity vector field construction module 720 is used to construct the velocity vector field of the sand ridge line image data of two adjacent frames in the sand ridge line image sequence using the dense optical flow method to obtain a spatiotemporal cube array of the sand ridge line movement velocity field; in one embodiment, the velocity vector field construction module 720 can be used to perform the operation S220 described above, which will not be repeated here.

[0115] Dune velocity change detection module 730 is configured to perform pixel-by-pixel temporal analysis of the space-time cube array of the sand ridge velocity field using an exponentially weighted moving average control chart to determine temporal velocity anomaly points. This temporal velocity anomaly point determination result is then superimposed onto the dune remote sensing observation time series imagery to perform position-time mapping and obtain dune velocity change detection results. In one embodiment, dune velocity change detection module 730 can be configured to perform operation S230 described above and will not be further described here.

[0116] According to embodiments of the present invention, any multiple modules among the remote sensing image preprocessing module 710, the velocity vector field construction module 720, and the sand dune rapid change detection module 730 can be combined into a single module, or any one of these modules can be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules can be combined with at least part of the functionality of other modules and implemented in a single module. According to embodiments of the present invention, at least one of the remote sensing image preprocessing module 710, the velocity vector field construction module 720, and the sand dune rapid change detection module 730 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented in hardware or firmware through any other reasonable means of circuit integration or packaging, or can be implemented in any one of the three implementation methods, or any appropriate combination of any of these. Alternatively, at least one of the remote sensing image preprocessing module 710 , the velocity vector field construction module 720 , and the sand dune rapid change detection module 730 may be at least partially implemented as a computer program module, which may perform corresponding functions when executed.

[0117] The following is a specific experiment combined with the attached Figure 8 -Attached Figure 10The above method and system provided by the present invention are verified.

[0118] Figure 8 1 is a schematic diagram of a real scene of two deserts, Sand1 and Sand2, and a schematic diagram of corresponding sand ridge line extraction according to an embodiment of the present invention.

[0119] Figure 9 1 and 2 are input images and corresponding output images for Sand1 rapid change detection according to an embodiment of the present invention.

[0120] Figure 10 2020-2021 is a schematic diagram of the rapid change detection results of Sand1 according to an embodiment of the present invention.

[0121] Figure 11 FIG. 4 is a schematic diagram of the rapid change detection results of Sand1 between 2010 and 2014 according to an embodiment of the present invention.

[0122] Figure 12 1 and 2 are input images and corresponding output images for Sand2 speed change detection according to an embodiment of the present invention.

[0123] Figure 13 FIG. 4 is a schematic diagram of Sand2 rapid change detection results from 2010 to 2014 according to an embodiment of the present invention.

[0124] Data from experimental setup: Data such as Figure 8 As shown, it is divided into two groups of data, Sand1 and Sand2. Each group of data consists of 6 scenes of real scenes of the same desert area and sand ridge line extraction pictures, which were taken from left to right in 2010, 2014, 2016, 2020, and 2021 to 2022. Since the movement of sand dunes is not a simple rigid body motion, but involves a large number of fluid morphological changes, the experiment uses sand ridge lines as input and establishes an optical flow field for rapid change detection based on the movement changes of sand ridge lines in the image sequence, including: the sudden appearance or disappearance of sand ridge lines, changes in the morphology of sand ridge lines, and rapid increase or decrease in the speed of sand dune movement. The parameters for the EWMA control chart are set as follows: standard deviation multiples , smoothing coefficient .

[0125] Dense optical flow method-EWMA detection results and analysis: Figure 9The input and output of Sand1's rapid change detection are shown. The top row shows the input sand ridge extraction image sequence, which shows the sand ridges moving and changing shape in the image sequence. The bottom row shows the output rapid change detection results. The base image is a real-life desert scene of the area. The green dots and arrows are velocity field vectors generated by the Farneback optical flow method, and the red areas are anomalous rapid change points. Because the Farneback optical flow method creates a velocity field based on two adjacent frames, and the velocity field is plotted on the second frame base image, the velocity field is not displayed in the initial image (2010).

[0126] The Sand1 speed change test results from 2020 to 2021 are selected as an example for analysis. Figure 10 As shown in the output result map, the yellow box marks the appearance of sand ridge lines. Comparing the sand ridge line input maps in 2020 and 2021, it can be clearly observed that the sand ridge lines suddenly appear in the same area of ​​the yellow box. From the sand ridge line extraction maps from 2010 to 2020, the sand ridge lines here show a trend of gradually fading and disappearing, while the sand ridge lines suddenly appear in the 2021 map, indicating that rapid changes have occurred, which is consistent with the algorithm's judgment as a rapid change area; the blue box marks the rapid change of sand ridge line movement. Comparing the sand ridge line input maps in 2020 and 2021, it can be observed that some sand ridge lines in the blue box area have moved and broken to the lower right. From the sand ridge line images of previous years, the sand ridge lines in this area have always maintained signs of continuity, but they appear in the 2021 map, indicating that rapid movement changes have occurred, which is also consistent with the algorithm's judgment as a rapid change area.

[0127] The Sand1 speed change test results from 2010 to 2014 were selected for analysis, such as Figure 11 As shown in the output result diagram, the purple box marks the disappearance of the sand ridge line. Comparing the sand ridge line input maps in 2010 and 2014, it can be clearly observed that the sand ridge line suddenly disappears in the same area of ​​the purple box. From the sand ridge line extraction maps of other years, it can be seen that the sand ridge line has always existed in this area, but suddenly disappeared in 2014, indicating that a rapid change has occurred, which is also consistent with the algorithm marking results.

[0128] The same test was conducted on the Sand2 dune data. Compared with the Sand1 data, the Sand2 data has more dunes, but the algorithm can still well construct the dune movement state and detect the dune movement rapid change point. The detection results are as follows: Figure 12 As shown, the generalization performance of this algorithm is demonstrated.

[0129] The Sand2 speed change test results from 2010 to 2014 are selected as an example for analysis. Figure 13As shown in the output result diagram, the yellow box marks the appearance of sand ridge lines. Comparing the sand ridge line input maps of 2010 and 2014, it can be clearly observed that the sand ridge lines suddenly appear in the same area of ​​the yellow box. From the sand ridge line extraction maps from 2010 to 2022, the sand ridge lines here remain relatively stable in the subsequent years. In 2010, the sand ridge line here is missing, and at the end of 2014, the sand ridge line here suddenly appears, indicating that a rapid change has occurred, which is consistent with the algorithm's judgment of a rapid change area; the blue box marks the rapid change of sand ridge line movement. Comparing the sand ridge line input maps of 2010 and 2014, it can be observed that some sand ridge lines in the blue box area move rapidly to the lower right, which is consistent with the rapid change detection area estimated by the algorithm; the purple box marks the disappearance of sand ridge lines. The disappearance of sand ridge lines can be clearly observed from the comparison of the 2010 and 2014 images, which is also consistent with the algorithm's estimation.

[0130] Figure 14 4 is a block diagram of an electronic device suitable for implementing a remote sensing detection method for sand dune movement speed change according to an embodiment of the present invention.

[0131] like Figure 14 As shown, an electronic device 1400 according to an embodiment of the present invention includes a processor 1401, which can perform various appropriate actions and processes based on a program stored in a read-only memory (ROM) 1402 or a program loaded from a storage unit 1408 into a random access memory (RAM) 1403. Processor 1401 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. Processor 1401 may also include onboard memory for caching purposes. Processor 1401 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0132] RAM 1403 stores various programs and data required for the operation of electronic device 1400. Processor 1401, ROM 1402, and RAM 1403 are interconnected via bus 1404. Processor 1401 executes the programs in ROM 1402 and / or RAM 1403 to perform various operations according to the method flow of the embodiment of the present invention. It should be noted that the programs may also be stored in one or more memories other than ROM 1402 and RAM 1403. Processor 1401 may also execute the programs stored in the one or more memories to perform various operations according to the method flow of the embodiment of the present invention.

[0133] According to an embodiment of the present invention, electronic device 1400 may further include an input / output (I / O) interface 1405, which is also connected to bus 1404. Electronic device 1400 may also include one or more of the following components connected to I / O interface 1405: an input unit 1406 including a keyboard, mouse, etc.; an output unit 1407 including devices such as a cathode ray tube (CRT), liquid crystal display (LCD), and speakers; a storage unit 1408 including a hard disk; and a communication unit 1409 including a network interface card such as a LAN card or modem. Communication unit 1409 performs communication processing via a network such as the Internet. A drive 1410 is also connected to I / O interface 1405 as needed. Removable media 1411, such as a magnetic disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed in drive 1410 as needed, so that computer programs read from the removable media can be installed into storage unit 1408 as needed.

[0134] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments, or may exist independently and not incorporated into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of the present invention.

[0135] According to an embodiment of the present invention, a computer-readable storage medium may be a non-volatile computer-readable storage medium, and may include, for example, but not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present invention, a computer-readable storage medium may include ROM 1402 and / or RAM 1403 described above, and / or one or more memories other than ROM 1402 and RAM 1403.

[0136] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0137] It will be understood by those skilled in the art that the features described in the various embodiments of the present invention may be combined and / or coupled in various ways, even if such combinations or couplings are not explicitly described in the present invention. In particular, the features described in the various embodiments of the present invention may be combined and / or coupled in various ways without departing from the spirit and teachings of the present invention. All such combinations and / or couplings fall within the scope of the present invention.

[0138] The above describes embodiments of the present invention. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. Although each embodiment has been described separately above, this does not mean that the measures in each embodiment cannot be advantageously used in combination. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. A remote sensing detection method for sand dune movement and rapid change, characterized in that: The method comprises: Extracting dune edges from dune remote sensing observation time series images using a deep learning method to obtain a dune edge image sequence, and preprocessing the dune edge image sequence to obtain a sand ridge line image sequence; The dense optical flow method is used to construct the velocity vector field of the sand ridge line image data of two adjacent frames in the sand ridge line image sequence to obtain a space-time cube array of the sand ridge line moving velocity field; Using an exponentially weighted moving average control chart, a pixel-by-pixel time series analysis is performed on the space-time cube array of the sand ridge line movement velocity field to obtain a time series rate anomaly judgment result, and the time series rate anomaly judgment result is superimposed on the sand dune remote sensing observation time series image to perform position-time identification to obtain a sand dune movement speed change detection result; The dense optical flow method is used to construct the velocity vector field of the sand ridge line image data of two adjacent frames in the sand ridge line image sequence, and the obtained sand ridge line moving velocity field space-time cube array includes: The Farneback dense optical flow model is used to calculate the horizontal axis displacement and the vertical axis displacement of each pixel in the sand ridge line image data of the two adjacent frames to obtain a time series of the displacement vector field; splicing the time series of the displacement vector fields of the sand ridge line image data of all two adjacent frames to obtain a splicing result; Based on the stitching result, the displacement vector of each of the two adjacent frames of sand ridge line image data and the time interval of each of the two adjacent frames of sand ridge line image data are calculated to obtain the space-time cube array of the sand ridge line movement velocity field; Among them, the sand ridge line movement velocity field space-time cube array includes time axis data, image horizontal axis data, image vertical axis data, pixel horizontal axis displacement data, pixel vertical axis displacement data, pixel horizontal axis velocity data and pixel vertical axis velocity data. The sand ridge line movement velocity field space-time cube array is used to characterize the velocity information of the pixels in the sand ridge line image sequence.

2. The method according to claim 1, characterized in that Preprocessing the dune edge image sequence to obtain a sand ridge line image sequence includes: performing noise filtering on the dune edge image sequence by using an erosion filtering method to obtain a denoised dune edge image sequence; Based on the dune edge image sequence, the denoised dune edge image sequence is resampled using a bilinear interpolation method to obtain the sand ridge line image sequence.

3. The method according to claim 1, characterized in that The exponentially weighted moving average control chart is used to perform pixel-by-pixel time series analysis on the space-time cube array of the sand ridge line movement velocity field, and the time series rate anomaly point judgment results are obtained, including: Extracting the displacement sequence and movement rate of each image coordinate on the sand ridge line from the spatiotemporal cube array of the sand ridge line movement velocity field to obtain the spatiotemporal sequence of the sand ridge line movement rate; The exponentially weighted moving average control chart is used to identify the time series rate anomaly points of the sand ridge line movement rate time-space series to obtain the time series rate anomaly point judgment result.

4. The method according to claim 3, characterized in that Extracting the displacement sequence and movement rate of each image coordinate on the sand ridge line from the spatiotemporal cube array of the sand ridge line movement velocity field, and obtaining the spatiotemporal sequence of the sand ridge line movement rate includes: Marking the space-time cube array of the sand ridge line moving velocity field at each moment pixel by pixel to obtain the sand ridge line pixel marking point at each moment; The horizontal axis velocity vector and the vertical axis velocity vector corresponding to the pixel marking point of the sand ridge line at each moment are added together and then modulo operation is performed to obtain the spatiotemporal sequence of the sand ridge line movement rate.

5. The method according to claim 3, characterized in that The exponentially weighted moving average control chart is used to identify the time series rate anomaly points of the sand ridge line movement rate time-space series, and the time series rate anomaly point judgment results obtained include: The recursive formula of the statistic of the exponentially weighted moving average control chart is constructed using the spatiotemporal sequence of the sand ridge line movement rate, and the recursive formula of the statistic is expanded into the form of the exponentially weighted average of all time rate sequences to obtain the expanded form of the statistic; Based on the maximum likelihood estimation method, constructing a standard deviation estimate of the statistic, and using the standard deviation estimate of the statistic to calculate the expanded form of the statistic, to obtain the upper and lower control limits and the center line of the exponentially weighted moving average control chart; The upper and lower control limits and center line of the exponentially weighted moving average control chart are used to perform pixel-by-pixel temporal rate anomaly identification on the spatiotemporal sequence of the sand ridge line movement rate, and pixels whose temporal rate exceeds the upper and lower control limits of the exponentially weighted moving average control chart are identified as the temporal rate anomaly points.

6. A remote sensing detection system for sand dune movement and rapid change, characterized in that: The system comprises: A remote sensing image preprocessing module is used to extract dune edges from dune remote sensing observation time series images using a deep learning method to obtain a dune edge image sequence, and preprocess the dune edge image sequence to obtain a sand ridge line image sequence; A velocity vector field construction module is used to construct a velocity vector field of the sand ridge line image data of two adjacent frames in the sand ridge line image sequence using a dense optical flow method to obtain a spatiotemporal cube array of the sand ridge line moving velocity field; A dune speed change detection module is used to perform pixel-by-pixel time series analysis on the space-time cube array of the sand ridge line movement velocity field using an exponentially weighted moving average control chart to obtain a time series rate anomaly judgment result, and to obtain a dune movement speed change detection result by superimposing the time series rate anomaly judgment result onto the dune remote sensing observation time series image for position-time identification; The dense optical flow method is used to construct the velocity vector field of the sand ridge line image data of two adjacent frames in the sand ridge line image sequence, and the obtained sand ridge line moving velocity field space-time cube array includes: The Farneback dense optical flow model is used to calculate the horizontal axis displacement and the vertical axis displacement of each pixel in the sand ridge line image data of the two adjacent frames to obtain a time series of the displacement vector field; splicing the time series of the displacement vector fields of the sand ridge line image data of all two adjacent frames to obtain a splicing result; Based on the stitching result, the displacement vector of each of the two adjacent frames of sand ridge line image data and the time interval of each of the two adjacent frames of sand ridge line image data are calculated to obtain the space-time cube array of the sand ridge line movement velocity field; Among them, the sand ridge line movement velocity field space-time cube array includes time axis data, image horizontal axis data, image vertical axis data, pixel horizontal axis displacement data, pixel vertical axis displacement data, pixel horizontal axis velocity data and pixel vertical axis velocity data. The sand ridge line movement velocity field space-time cube array is used to characterize the velocity information of the pixels in the sand ridge line image sequence.

7. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images