SAR (Synthetic Aperture Radar) image overlay detection method, device and equipment and storage medium

Abstract

The invention discloses an SAR (Synthetic Aperture Radar) image overlay detection method, device and equipment and a storage medium. The method comprises the following steps: acquiring SAR time sequence image data, external DEM (Digital Elevation Model) data and external earth surface classification coverage data of a current detection area; obtaining a region amplitude diagram of the current detection region according to the SAR time sequence image data, and carrying out normalization processing on the region amplitude diagram; obtaining a corresponding interferometric phase diagram according to the SAR time sequence image data, and calculating a coherence coefficient; performing normalization processing on the external DEM data, performing wave band splicing on the normalized region amplitude diagram, interference phase diagram and coherence coefficient, and the normalized external DEM data and external earth surface classification coverage data to generate a corresponding data block, and inputting the data block into a preset overlay identification model, and obtaining an overlay identification result of the current detection area. The SAR overlay recognition method can improve the overlay recognition precision of the SAR.

Description

Technical Field

[0001] The present invention relates to the field of synthetic aperture radar interferometry technology, and in particular to a SAR image overlap detection method, device, equipment and storage medium. Background Art

[0002] SAR (Synthetic Aperture Radar) is a radar imaging technology that can be used to obtain various types of information on the earth's surface. It transmits and receives surface reflection signals during the operation of satellites to obtain surface information. However, due to the relative position of the objects and the propagation characteristics of radar waves, some objects on the radar propagation path are often partially or completely blocked by other objects, causing the signals of the objects in the area to weaken or disappear, thus forming an overlapping or aliasing "overlap" phenomenon on the image. "Overlap" has two sides in SAR image processing. On the one hand, the existence of overlap will seriously interfere with the SAR data processing process. On the other hand, the overlapped area of ​​the SAR image can be reconstructed in three dimensions through the corresponding de-overlapping algorithm to obtain the three-dimensional information of the area. It has both advantages and disadvantages. In order to utilize its advantages or avoid its disadvantages, it is necessary to first accurately identify the overlapped area in the SAR image.

[0003] However, the existing SAR image overlap detection technology has certain scenario limitations in specific applications. It does not make comprehensive use of valuable external auxiliary data, but only uses the original SAR image or the homologous data derived from the original SAR image. The recognition system does not add any new information, resulting in room for improvement in the overlap recognition accuracy and diversity of application scenarios of the recognition method. Summary of the invention

[0004] The present invention provides a SAR image overlap detection method, device, equipment and storage medium to solve the technical problem that the existing SAR image overlap detection technology does not comprehensively utilize valuable external auxiliary data, only utilizes the original SAR image or the homologous data derived from the original SAR image, resulting in low overlap recognition accuracy of the recognition method.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a SAR image overlap detection method, comprising:

[0006] Acquire SAR time-series image data of the current detection area and corresponding external heterogeneous data; wherein the external heterogeneous data includes: external DEM data and external surface classification coverage data;

[0007] Extracting corresponding amplitude information from the SAR time-series image data, obtaining a regional amplitude map of the current detection area according to the extracted amplitude information, and performing normalization processing on the regional amplitude map;

[0008] Performing interference processing on the SAR time-series image data to obtain a corresponding interference phase map, and calculating a corresponding coherence coefficient based on the interference phase map;

[0009] The external DEM data is normalized, and then the normalized regional amplitude map, interference phase map, coherence coefficient, normalized external DEM data and external surface classification coverage data are band-joined to generate corresponding data blocks, and the data blocks are input into a preset overlap recognition model, so that the overlap recognition model recognizes the overlap of the current detection area according to the input data blocks, and obtains the overlap recognition result of the current detection area;

[0010] The overlap recognition model is obtained by training a preset deep learning model with historical SAR time-series image data and corresponding historical external heterogeneous data as input and overlap recognition results of the corresponding area as output.

[0011] As a preferred solution, the generation of the overlap recognition model includes:

[0012] Acquire historical SAR time-series image data and corresponding historical external heterogeneous data; wherein the historical external heterogeneous data includes: historical external DEM data and historical external surface classification coverage data;

[0013] Extracting corresponding historical amplitude information from the historical SAR time-series image data, obtaining a historical regional amplitude map of the corresponding area according to the extracted historical amplitude information, and normalizing the historical regional amplitude map;

[0014] Performing interference processing on the historical SAR time-series image data to obtain a corresponding historical interference phase map, and calculating a corresponding historical coherence coefficient based on the historical interference phase map;

[0015] Normalizing the historical external DEM data, and then band-joining the normalized historical regional amplitude map, the historical interference phase map, the historical coherence coefficient, the normalized historical external DEM data, and the historical external surface classification coverage data to generate corresponding data blocks, and using the data blocks as input data samples;

[0016] Marking the overlapping areas in the historical SAR time series image data to generate overlapping labels for the corresponding areas;

[0017] A model training sample is generated according to the input data sample and the overlap label, and then a preset deep learning model is trained according to the model training sample and a preset loss function to generate a corresponding overlap recognition model.

[0018] As a preferred solution, the step of marking the overlapped areas in the historical SAR time series image data and generating overlapped labels for the corresponding areas includes:

[0019] According to different preset overlap recognition methods, SAR image overlap recognition is performed on the historical SAR time series image data to obtain different overlap recognition results; wherein the preset overlap recognition methods include: an overlap recognition method based on amplitude and coherence threshold, a satellite radar line of sight analysis recognition method, and a DEM slope and aspect analysis method;

[0020] Integrate the overlap recognition results, and then mark the overlap areas in the historical SAR time series image data according to the integrated overlap recognition results to generate overlap labels for the corresponding areas;

[0021] As a preferred solution, generating a model training sample according to the input data sample and the overlapped label includes:

[0022] Using the input data sample and the overlapped label as initial model training samples;

[0023] For each sample data in the initial model training sample, several data enhancement methods are randomly selected from the preset data enhancement methods, and data enhancement processing is performed on the sample data in turn to obtain sample data after data enhancement; wherein the data enhancement methods include: geometric transformation, noise injection, image aliasing and color transformation;

[0024] All sample data after data augmentation are used as model training samples.

[0025] As a preferred solution, the step of identifying the overlap of the current detection area according to the input data block to obtain the overlap identification result of the current detection area includes:

[0026] Perform image-level data fusion on the input data blocks;

[0027] Perform shallow feature extraction on the fused data, and perform feature fusion on the extracted shallow features;

[0028] According to the fused shallow features, the global features in the fused shallow features are extracted, and then the extracted global features are subjected to feature fusion and feature classification to obtain the overlapping recognition result of the current detection area.

[0029] As a preferred solution, the loss function is:

[0030]

[0031] L branch =α·L CE +β·LIoU ;

[0032] L total =L b1 +L b2 +L b3 ;

[0033] Among them, L CE is a cross entropy loss function, which is used to measure the gap between the overlap probability predicted by the overlap recognition model and the true label; L IoU is the intersection-over-union loss function, which is used to calculate the intersection-over-union ratio of the predicted area of ​​the overlap recognition model and the actual overlap area; N is the number of samples; y i is the true label; p i is the predicted probability of the model; L branch is the deep learning model loss function of the overlap recognition model, which is used to integrate the results of the cross entropy loss function and the intersection-over-intersection loss function, and calculate the loss function of each branch channel according to the integrated result; α and β are the integrated weights of the cross entropy loss function and the intersection-over-intersection loss function respectively; L total is the overall loss function of the overlay recognition model, which is used to calculate the loss of each model output channel, and then sum the losses of all model output channels, and take the sum as the overall loss of the model. b1 , L b2 and L b3 are the losses of each model channel respectively.

[0034] As a preferred solution, after obtaining the overlap recognition result of the current detection area, the method further includes:

[0035] According to the overlap recognition result of the current detection area, the precision, recall and intersection-over-union of the overlap recognition result are calculated, and then the model recognition accuracy of the overlap recognition model is evaluated according to the precision, recall and intersection-over-union.

[0036] Based on the above embodiment, another embodiment of the present invention provides a SAR image overlap detection device, including: a data acquisition module, an amplitude information extraction module, an interference processing module and an overlap recognition module;

[0037] The data acquisition module is used to acquire the SAR time-series image data of the current detection area and the corresponding external heterogeneous data; wherein the external heterogeneous data includes: external DEM data and external surface classification coverage data;

[0038] The amplitude information extraction module is used to extract corresponding amplitude information from the SAR time-series image data, obtain a regional amplitude map of the current detection area according to the extracted amplitude information, and perform normalization processing on the regional amplitude map;

[0039] The interference processing module is used to perform interference processing on the SAR time-series image data to obtain a corresponding interference phase map, and calculate a corresponding coherence coefficient according to the interference phase map;

[0040] The overlap recognition module is used to normalize the external DEM data, and then band-splicing the normalized regional amplitude map, interference phase map, coherence coefficient, normalized external DEM data and external surface classification coverage data to generate corresponding data blocks, and input the data blocks into a preset overlap recognition model, so that the overlap recognition model can recognize the overlap of the current detection area according to the input data blocks, and obtain the overlap recognition result of the current detection area; wherein the overlap recognition model is obtained by training a preset deep learning model with historical SAR time series image data and corresponding historical external heterogeneous data as input and the overlap recognition result of the corresponding area as output.

[0041] Based on the above embodiments, another embodiment of the present invention provides an electronic device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the SAR image overlap detection method described in the above invention embodiment is implemented.

[0042] Based on the above embodiments, another embodiment of the present invention provides a storage medium, wherein the storage medium includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the SAR image overlap detection method described in the above invention embodiment.

[0043] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0044] The present invention provides a SAR image overlap detection method. Compared with the existing overlap identification method which only uses a single original SAR image or homologous data derived from the original SAR image, the present invention not only obtains the SAR time series image data of the current detection area, but also obtains the external heterogeneous data corresponding to the current detection area: external DEM data and external surface classification coverage data;

[0045] Then, the corresponding amplitude information is extracted from the SAR time-series image data, and a regional amplitude map of the current detection area is obtained according to the extracted amplitude information, and the regional amplitude map is normalized; then, the SAR time-series image data is interferometrically processed to obtain a corresponding interference phase map, and a corresponding coherence coefficient is calculated according to the interference phase map; wherein the normalized regional amplitude map, interference phase map, and coherence coefficient are homologous data derived from the SAR time-series image data;

[0046] Finally, the external DEM data is normalized, and then the normalized regional amplitude map, interference phase map, coherence coefficient, normalized external DEM data and external surface classification coverage data are band-joined to generate corresponding data blocks, and the data blocks are input into a preset overlap recognition model, so that the overlap recognition model recognizes the overlap of the current detection area according to the input data blocks, and obtains the overlap recognition result of the current detection area.

[0047] The present invention fuses homologous data with heterologous data to construct corresponding data blocks, and then inputs the fused data blocks into a preset overlap recognition model. A deep learning model is used to achieve comprehensive utilization of homologous data and external heterologous data, thereby improving the stability of overlap recognition in various application scenarios and the overlap recognition accuracy of SAR. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of a SAR image overlap detection method provided by one embodiment of the present invention;

[0049] Figure 2 It is the overall technical flow chart of the overlapping mask recognition and detection method based on deep learning of multi-source data in the present invention;

[0050] Figure 3 This is a flow chart of making SAR image overlay label samples designed by the present invention;

[0051] Figure 4 This is an example of a sample of overlapping mask recognition;

[0052] Figure 5 is an example diagram of the overlay sample enhancement pool of the present invention;

[0053] Figure 6 The figure is a schematic structural diagram of a SAR image overlap detection device provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned figure descriptions and any variations thereof are intended to cover non-exclusive inclusions.

[0056] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined.

[0057] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0058] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0059] In the description of the embodiments of the present application, the term "multiple" refers to more than two (including two). Similarly, "multiple groups" refers to more than two groups (including two groups), and "multiple pieces" refers to more than two pieces (including two pieces).

[0060] In the description of the embodiments of the present application, unless otherwise clearly specified and limited, technical terms such as "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0061] Embodiment 1

[0062] Please refer to Figure 1 , is a flow chart of a SAR image overlap detection method provided by an embodiment of the present invention, comprising the following specific steps:

[0063] S1. Acquire SAR time-series image data of the current detection area and corresponding external heterogeneous data; wherein the external heterogeneous data includes: external DEM data and external surface classification coverage data;

[0064] Specifically, when performing SAR image overlap detection, first obtain the SAR time series image data of the current detection area and the corresponding external heterogeneous data; wherein the external heterogeneous data includes: external DEM data and external surface classification coverage data. Then, the original time series SAR data is preprocessed as necessary, including filtering and denoising, geometric correction, radiation correction, and image registration.

[0065] S2, extracting corresponding amplitude information from the SAR time series image data, obtaining a regional amplitude map of the current detection area according to the extracted amplitude information, and performing normalization processing on the regional amplitude map;

[0066] Considering that the overlapped area is the superposition of multiple signals and often has a large amplitude value, the amplitude information in the amplitude map can be used as one of the important recognition features. Therefore, after basic preprocessing of the regional raw SAR data, the amplitude information is further extracted from the regional raw SAR data to obtain the regional amplitude map, and the amplitude map is normalized as shown in the following formula 1:

[0067]

[0068] Where X' is the normalized data, X max , X minare the maximum and minimum values ​​of the data to be normalized, respectively. The amplitude data to be normalized is processed as in the above formula (1) to obtain the normalized value. After the amplitude information is normalized to the maximum and minimum values, it can be ensured that the amplitude range is unified to the interval [0,1] while the original data distribution remains unchanged, which is convenient for the subsequent model training and processing process.

[0069] S3, performing interference processing on the SAR time series image data to obtain a corresponding interference phase map, and calculating a corresponding coherence coefficient according to the interference phase map;

[0070] In addition, considering that the overlapped area also has large differences from the non-overlapped area in the SAR image coherence map and interferometric phase map, the master and slave SAR images of the area are interferometrically processed to obtain the interferometric phase map, and the coherence coefficient is obtained by coherence calculation. The coherence map and interferometric phase map are also used as one of the data for the subsequent overlap recognition deep learning method. The above data, including amplitude map, interferometric phase map and coherence map, are all calculated based on the original SAR image, so they are regarded as data of the same source as the original SAR data.

[0071] S4, normalizing the external DEM data, and then band splicing the normalized regional amplitude map, interference phase map, coherence coefficient, normalized external DEM data and external surface classification coverage data to generate corresponding data blocks, and input the data blocks into a preset overlap recognition model, so that the overlap recognition model recognizes the overlap of the current detection area according to the input data blocks, and obtains the overlap recognition result of the current detection area;

[0072] In addition, it is noted that the formation of overlap in SAR images is often closely related to the terrain undulation of the region. Therefore, the external DEM data of the corresponding region is also obtained as auxiliary information for overlap identification, and the DEM data is also normalized as in the above formula (1) to unify the numerical range of the DEM data to [0,1]. However, DEM data lacks information about surface objects, and overlaps generated by areas such as trees and buildings cannot be analyzed. Therefore, the surface classification coverage data of the region is also obtained to indicate different types of objects in the region. Tall objects such as buildings have a greater risk of overlap, while flat areas basically do not produce overlap. External surface classification coverage data also plays an important role in this. Since the external surface classification coverage data and DEM data are not from the original SAR image, they can be regarded as external heterogeneous data. This part of the data can add additional information entropy to the classification system, thereby helping to improve the accuracy of overlap classification and identification.

[0073] The method proposed in the present invention can comprehensively utilize SAR original data and derived SAR image amplitude map, coherence map, phase map and other homologous data, as well as external heterogeneous DEM data and surface classification coverage data by constructing a multi-branch convolutional mask recognition deep learning model, and enhance the abstract category information and spatial position information of the extracted mask features through multiple exchanges of features between multiple branches of the model, thereby improving the recognition accuracy and scene robustness of traditional methods for SAR image mask, and ensuring the accuracy of the mask recognition position and the integrity of the edge information while ensuring the accuracy of the mask recognition.

[0074] The normalized regional amplitude map, interferometric phase map, coherence coefficient, normalized external DEM data and external surface classification coverage data are band-joined to construct the SAR-DataCube data block. Then the model structure is initialized, the model weights obtained by model training are loaded, and then the preprocessed SAR-DataCube data is used as the model input to start the forward propagation process of the model. After the model forward propagation, the predicted probability map of the corresponding regional overlap is obtained, and then the probability threshold is set to visualize the probability map to obtain the overlap recognition result.

[0075] The overlap recognition model is obtained by training a preset deep learning model with historical SAR time-series image data and corresponding historical external heterogeneous data as input and overlap recognition results of the corresponding area as output.

[0076] Preferably, the generation of the overlap recognition model includes: acquiring historical SAR time-series image data and corresponding historical external heterogeneous data; wherein the historical external heterogeneous data includes: historical external DEM data and historical external surface classification coverage data; extracting corresponding historical amplitude information from the historical SAR time-series image data, obtaining a historical regional amplitude map of the corresponding area according to the extracted historical amplitude information, and normalizing the historical regional amplitude map; interfering the historical SAR time-series image data to obtain a corresponding historical interference phase map, and calculating a corresponding historical coherence coefficient according to the historical interference phase map; The historical external DEM data is normalized, and then the normalized historical regional amplitude map, historical interference phase map, historical coherence coefficient, normalized historical external DEM data and historical external surface classification coverage data are band-joined to generate corresponding data blocks, and the data blocks are used as input data samples; the overlapping areas in the historical SAR time series image data are marked to generate overlapping labels for the corresponding areas; model training samples are generated according to the input data samples and the overlapping labels, and then a preset deep learning model is trained according to the model training samples and a preset loss function to generate a corresponding overlapping recognition model.

[0077] For details, please refer to Figure 2 , is the overall technical flow chart of the overlapping mask recognition detection method based on deep learning of multi-source data in the present invention, including S1: SAR data preprocessing, S2: overlapping mask recognition deep learning sample library construction, S3: overlapping mask recognition deep learning model design and training, S4: overlapping mask recognition based on deep learning model and other main steps. The above content has explained the overlapping mask recognition steps based on the deep learning model, that is, the model use process. The following focuses on the model training process in S1-S3:

[0078] 1. SAR data preprocessing:

[0079] Acquire historical SAR time-series image data and corresponding historical external heterogeneous data; wherein the historical external heterogeneous data include: historical external DEM data and historical external surface classification coverage data; extract corresponding historical amplitude information from the historical SAR time-series image data, obtain a historical regional amplitude map of the corresponding area according to the extracted historical amplitude information, and normalize the historical regional amplitude map; perform interference processing on the historical SAR time-series image data to obtain a corresponding historical interference phase map, and calculate a corresponding historical coherence coefficient according to the historical interference phase map; perform normalization processing on the historical external DEM data.

[0080] Preferably, the marking of overlapping areas in the historical SAR time series image data to generate overlapping labels for corresponding areas includes: performing SAR image overlapping identification on the historical SAR time series image data according to different preset overlapping identification methods to obtain different overlapping identification results; wherein the preset overlapping identification methods include: overlapping identification methods based on amplitude and coherence thresholds, satellite radar line of sight analysis and identification methods, and DEM slope and aspect analysis methods; integrating the overlapping identification results, and then marking the overlapping areas in the historical SAR time series image data according to the integrated overlapping identification results to generate overlapping labels for corresponding areas;

[0081] Preferably, generating a model training sample according to the input data sample and the overlapping label includes: taking the input data sample and the overlapping label as initial model training samples; for each sample data in the initial model training sample, randomly selecting several data enhancement methods from preset data enhancement methods, performing data enhancement processing on the sample data in turn, and obtaining data-enhanced sample data; wherein the data enhancement method includes: geometric transformation, noise injection, image aliasing and color transformation; taking all data-enhanced sample data as model training samples.

[0082] 2. Construction of deep learning sample library for overlapping recognition:

[0083] The present invention adopts a supervised deep learning model framework to perform SAR image overlay recognition and detection, which requires a large number of high-quality overlay recognition samples. A complete overlay recognition sample should include two parts: input data and corresponding label data. The input data used for overlay recognition in the present invention includes historical SAR raw data and the homologous SAR amplitude, interference phase, coherence data derived therefrom, as well as the regional DEM and surface classification coverage obtained from the outside. These heterogeneous data, label data are the overlay area mask in the corresponding SAR image data.

[0084] The training of deep learning models requires a massive amount of high-quality annotated data sets. The method of relying solely on manual visual interpretation to annotate SAR image overlays is inefficient. However, the existing overlay recognition methods have certain limitations in terms of accuracy and stability in the labeling methods of diversified SAR application scenarios.

[0085] In order to solve the problems existing in the production of overlap identification samples, the present invention adopts a strategy of overlap integrated identification to produce efficient and accurate overlap sample labels. The overall idea is to first use a variety of overlap identification methods to mark the overlap areas in the SAR image respectively, and then integrate the overlap identification results obtained by different methods to determine the final overlap label. The overlap label production process integrating multiple methods improves the reliability of a single method for overlap identification labeling, and the automated label recognition production process also improves the efficiency of label production compared to the manual visual method.

[0086] Please refer to Figure 3 , is a flow chart of the SAR image mask label sample production designed by the present invention, using Figure 3 The method shown in the figure performs the automatic labeling process of overlapping samples. First, the original SAR image is obtained and basic preprocessing is performed. The overlapping detection and recognition of SAR images are performed by various technical methods respectively, and the overlapping recognition results under different technical routes are obtained. The final overlapping label is obtained by integrating the results of multiple overlapping recognitions.

[0087] like Figure 3 The overlap recognition methods used in the embodiments provided by the present invention include an overlap detection method based on amplitude and coherence threshold, a satellite line of sight analysis method, and a DEM slope and aspect analysis method. It is worth noting that in specific applications, other overlap recognition methods can also be flexibly selected based on the scene.

[0088] (1) Overlap detection method based on amplitude and coherence threshold:

[0089] Since the overlapping areas in SAR images are generated by the superposition of multiple signals, they often have a higher amplitude threshold and a lower coherence threshold, and this feature can be used to identify overlapping areas. Figure 2 As shown in the figure, the amplitude map is first extracted based on the original SAR image and the coherence is calculated. Then, the overlap detection method based on the amplitude and coherence thresholds screens the points whose amplitude and coherence meet certain conditions by setting the amplitude threshold and the coherence threshold to distinguish the overlap and non-overlap areas in the SAR image.

[0090] (2) Satellite radar line of sight analysis and recognition method: This method determines which parts of the ground objects are blocked and form shadows, and which parts are aliased and form shadows based on the line of sight observed by the satellite radar.

[0091] (3) DEM slope and aspect analysis method: Since the slope of overlapping areas is often large, the slope and aspect analysis can be calculated using DEM data, and a comprehensive analysis of the slope and aspect can be performed to identify overlapping areas.

[0092] Different overlap recognition methods have their own advantages in different application scenarios. In the present invention, the overlap results of SAR images under different methods are obtained through the above-mentioned different overlap recognition methods, and they are comprehensively processed through the integration strategy. The integration strategy adopted in the embodiment of the present invention is a voting mechanism. When more than half of the methods regard an area as overlap, the area is marked as an overlap label. However, it is worth noting that when the present invention is implemented, a suitable overlap recognition method can be selected as one of the integration methods according to the specific situation. It is not limited to only using the method in this embodiment, and the integration strategy used can also be adjusted accordingly, such as according to the overlap recognition effect of different overlap integration sub-methods in different SAR image scenes, the weight of this type of method is adjusted accordingly to achieve the purpose of integrating the overlap areas identified by different methods.

[0093] The integrated overlay recognition method designed above in the present invention can obtain more accurate and reliable overlay recognition results, and use them as label data for the overlay recognition deep learning training sample library.

[0094] For the preparation of the input data of the sample library, the present invention first extracts the amplitude map from the SAR image and calculates the coherence map and the interference phase information, and at the same time obtains the DEM data and the surface classification coverage data of the corresponding area from the outside, and performs band splicing to obtain the SAR-DataCube, which has a variety of overlapping information. Compared with only using the original SAR image or the data derived from the same source as the SAR image as the input data, the overlapping recognition sample of the present invention provides more external heterogeneous data, which is more conducive to the information of overlapping recognition detection.

[0095] Please refer to Figure 4 , is an example of an overlapped recognition sample. The final form of an overlapped recognition sample is: ([input data]-[label data]), such as Figure 4 shown.

[0096] The input data is the SAR-DataCube of the same area at different times, and each SAR-DataCube has homologous data parts such as amplitude data, phase data, and coherence data obtained after processing the original SAR data, and heterogeneous data such as externally acquired regional DEM data and ground classification coverage data. The external heterogeneous data adds additional information entropy to the entire system, which is of great significance for overlap recognition. The label data is the overlap area obtained by integrating the above-mentioned multiple overlap recognition methods, and each SAR-DataCube at each moment has a corresponding overlap label. The above is a detailed description of the sample form in the present invention.

[0097] Considering the complex and easily disturbed imaging conditions of satellite images and the diverse types of land objects, rich and diverse overlap recognition samples are of great help to the training of overlap recognition models, which can significantly improve the model's overlap recognition capabilities in various complex scenes and its ability to resist noise interference during the model recognition process. Real overlap scenes are often more complex. Due to differences in satellite imaging perspectives and imaging conditions, overlap areas often rotate, geometrically deform, and shadow in the image, which will interfere with the recognition of overlap areas. In addition, overlap areas are often mixed with background areas and interfere with each other and are difficult to distinguish. Using only overlap samples in normal conditions is obviously not in line with the characteristics of overlap in actual application scenarios.

[0098] The present invention takes into account that the process of traditional data sample enhancement methods is relatively fixed and single, that is, the method and process of data enhancement remain fixed after being determined. As a result, even if data enhancement is performed, the generated samples are obtained through the same fixed enhancement process method, and the sample richness is insufficient. The present invention specially designs a data enhancement method of a random sample enhancement pool for the SAR image overlay recognition task, which can significantly improve the accuracy and scene adaptability of the model in the overlay recognition task.

[0099] Please refer to Figure 5, is an example diagram of the overlapped sample enhancement pool of the present invention. The present invention constructs an overlapped sample enhancement method pool. According to the main characteristics of the overlapped areas in the SAR image, four basic methods are designed in the enhancement method pool, including geometric changes, noise injection, image aliasing, and color enhancement. When data enhancement is applied specifically, one or more methods are first randomly selected from each basic method pool each time, and then the methods selected from each basic method pool are combined in random order, and a sample is enhanced using the combined methods and processes. Through the random enhancement method based on the method pool, it is ensured that the enhancement methods and processes used for each sample in the sample library are different, ensuring the diversity of samples obtained through enhancement. Moreover, in specific use, the sub-methods in each basic method pool can be further adjusted according to different application scenarios to ensure the flexibility of the data enhancement method based on the random method pool.

[0100] The geometric changes in the enhancement method pool of the present invention include rotation, scaling, translation and flipping, etc. The geometric change enhancement can enhance the model's adaptability to changes in the geometric shape of the overlapping area. The random rotation of the image can simulate different satellite imaging angles, so that the model can enhance its recognition ability for overlapping targets in different directions. Random scaling enhancement can change the size of the image to simulate the effects of different observation distances, which helps the model learn multi-scale overlapping targets in SAR images. Translation image enhancement can help the model learn the information of overlapping targets at different positions in the image and enhance the model's robustness to changes in target positions. Horizontal or vertical flip enhancement can help the model learn symmetry-related features of the overlapping area. The above geometric enhancement methods can help the model better learn the diverse geometric states of overlapping areas during training.

[0101] Color-related enhancements include adjusting brightness, contrast, and histogram equalization. The brightness of the overlapping areas of SAR images changes due to factors such as the imaging environment. By adjusting the brightness, samples in different environments can be generated to help the model adapt to these changes. In overlapping samples, multiple objects may be similar in grayscale, making recognition difficult. Increasing the contrast can make the boundaries of different objects clearer and improve recognition accuracy. The histogram equalization operation can make the grayscale values ​​of SAR images evenly distributed to improve the clarity of the image. Color enhancement methods are also important for training models to more robustly identify objects under various environmental conditions.

[0102] In addition, the method of injecting noise can enhance the ability to resist noise during the model training and learning process. The present invention is aimed at the noise characteristics of the overlapping area in the real scene, and the sub-noise enhancement methods set include blurring, artifact addition, salt and pepper noise, Gaussian noise and other methods. Artifact addition can simulate the artifacts caused in the imaging process and enhance the adaptability of the model to the overlapping phenomenon under complex background. The blurring operation can simulate the situation of insufficient image clarity to help the model better handle the overlapping detection phenomenon under the blurred image and enhance the overlapping recognition ability in the real scene. Salt and pepper noise simulates the noise interference in the image transmission process by randomly adding salt and pepper noise points to the image. Salt and pepper noise helps to improve the overlapping recognition ability of the model under defective images. Gaussian noise enhancement simulates sensor noise by adding random noise that conforms to Gaussian distribution to the image. Through the above multiple noise injection enhancement methods, the common noise scenes in SAR images can be fully simulated to improve the anti-noise interference ability of the overlapping recognition algorithm model.

[0103] In addition, the overlapping areas in SAR images have more complex image characteristics. Therefore, image aliasing methods such as Mixup, CutMix, and Mosaic can be used for sample enhancement to effectively improve the generalization ability and robustness of model learning.

[0104] The Mixup method linearly combines two images in a certain ratio to generate new samples. This method can enhance the generalization ability of the model and make it more robust to unseen samples. In SAR images, overlapping samples usually contain overlapping information of multiple objects. Mixup can simulate the interaction between different objects to a certain extent, helping the model learn richer features and boundaries, thereby improving the SAR overlapping recognition ability in complex scenes.

[0105] The CutMix method randomly cuts two images and splices them to generate new samples. This method can enhance the robustness of the model to occlusion or partial occlusion of overlapping areas. In overlapping samples, ground objects may be partially occluded by other objects. CutMix enhancement can effectively simulate this situation, so that the model can be exposed to information about missing or overlapping overlapping parts during training, thereby enhancing its adaptability to overlapping samples in various complex scenes.

[0106] Mosaic technology achieves the purpose of increasing sample diversity by stitching together multiple small images to form a new image. SAR image overlay samples often include different types of objects, textures, and background noise. Through Mosaic enhancement technology, samples containing information on objects in various overlay areas can be generated, so that the model can learn features in different environments during training and improve the recognition effect of diverse overlay samples.

[0107] The above is a brief introduction to various basic methods in the masking sample enhancement pool of the present invention, which can be modified according to specific circumstances in specific applications.

[0108] The construction process of the deep learning sample library for overlay recognition in the present invention includes the integrated production of multi-method overlay recognition labels, preprocessing of input data, and data set enhancement based on a random enhancement method pool. Through the above process, a high-quality recognition deep learning sample library with rich scenarios is finally constructed for the training process of the overlay recognition model.

[0109] Preferably, the overlapping of the current detection area is identified based on the input data block to obtain the overlapping identification result of the current detection area, including: performing image-level data fusion on the input data block data; performing shallow feature extraction on the fused data, and performing feature fusion on the extracted shallow features; extracting global features from the fused shallow features based on the fused shallow features, and then performing feature fusion and feature classification on the extracted global features to obtain the overlapping identification result of the current detection area.

[0110] 3. Design and training of deep learning model structure for overlapping recognition:

[0111] The present invention designs and proposes a SAR image overlap region recognition method based on a deep learning model, and designs a convolutional multi-branch deep learning model structure according to the characteristics of the SAR image overlap region. The model is characterized in that multiple data fusion processes from shallow to deep are designed, the homologous information and heterogeneous information in the SAR-DataCube are deeply fused multiple times at different feature levels, and overlapping high-level abstract features are extracted through a serial-parallel multi-branch feature extraction module, and finally the overlap feature fusion classification module is used to identify and classify the extracted features in the overlap region.

[0112] The model designed in the present invention makes maximum use of the overlapping features in the multi-source input SAR-DataCube information through a multi-level resolution convolution structure and a multi-level feature fusion method, and ensures a good balance between the spatial features of the overlapping and the abstract classification feature capabilities during the processing process, so as to achieve the accurate distinction of the category information of the overlapping targets while ensuring the position accuracy of the overlapping target recognition.

[0113] The model accepts SAR-DataCube at multiple moments (time series) after preprocessing as input of different channels, and also obtains output of multiple channels after model processing, where the output of each channel corresponds to the overlap recognition result of the input data at different moments.

[0114] First, the DFM-1 module (Data fusion module-1) performs preliminary fusion of the input data at the image level. The DFM-1 module consists of a fully connected layer (linear layer), an activation function (Leaky ReLu) and a Dropout regularization layer. The module linearly changes the input data of different channels through the fully connected layer, and optimizes the parameters of the fully connected layer in the subsequent model training process to achieve adaptive balance between the weights of the input channel data and the importance of different channel data. Secondly, the nonlinear expression ability of the model is enhanced by the design of the activation function layer. The Leaky ReLU activation function is used in the present invention. Leaky ReLU can avoid the problem of complete neuron inactivation and improve the model's recognition ability for complex overlapping features. Finally, in order to avoid serious overfitting of the model, the Dropout regularization layer is used for random discarding. The DFM-1 module of the model is composed of the above structures. The module plays a role in the preliminary image-level fusion and nonlinear ability enhancement of the multi-channel overlapping recognition data of the input SAR-DataCube, and prevents serious overfitting through the Dropout regularization layer.

[0115] After that, a SFE-1 module (shallow Feature Extraction-1) performs shallow feature extraction on the output data after the preliminary fusion processing of the DFM-1 module. The shallow features retain the spatial position relationship of the overlapped features. This module consists of a convolution layer, a batch normalization layer, and an activation function layer. The convolution layer processes the input data through a convolution operation. The convolution can consider the pixel neighborhood relationship on the convolution kernel range of the input data, and extract the feature dependency pattern between the pixels in the region, avoiding the problem that the traditional feature extraction method can only process single pixel features, and plays a preliminary overlapping feature extraction function. The batch normalization layer can speed up the training speed of the model. By normalizing the input of each layer, batch normalization reduces the internal covariate shift of the data and enhances the stability of the network training process. Finally, this module also uses an activation function to enhance the nonlinear expression ability of the model structure. The data is processed by the SFE-1 module to obtain the overlapping shallow features, which retain the overlapping spatial position feature relationship, but its abstract classification ability and globality are insufficient.

[0116] Then, the FFM-2 module (Feature Fusion Module-2) constructed by the model is used to fuse the shallow features obtained by the SFE-1 module again. In the FFM-2 module, the features are first processed through global average pooling and maximum pooling operations to improve the global expression ability of the features. Then, the channel attention mechanism is used to fuse the features between channels, and the features are extracted through an MLP (Multi-Layer Perceptron) multi-layer perceptron structure composed of convolution, activation, and normalization layers, further enhancing the abstract expression ability of the acquired features.

[0117] After that, the abstract high-level and global features about overlap are further extracted through the CFE-2 and CFE-3 modules (Convolutional Feature Extraction). The CFE-2 module is characterized by having two branch structures, both of which are composed of structures such as convolution-normalization-activation function layers. The difference is that the feature data is not sampled when entering branch 1 of the CFE-2 module, and is downsampled by 2 times when entering branch 2. The above processing ensures that the features in branch 1 always maintain a high overlap spatial resolution information, and branch 2 has a stronger global feature expression capability.

[0118] CFE-3 has a three-branch structure, which is also composed of convolution-normalization-activation function layers. The input data of its three branches is obtained from the output of CFE-2, such as Figure 5 as shown in .

[0119] The input of branch 1 in the CFE-3 module is obtained by upsampling the output data of branch 2 of CFE-2 by 2 times and fusing it with the data of branch 1. This operation enables branch 1 of module 3 to retain a certain spatial resolution information of the overlapping features and to supplement and enhance a certain deep abstract expression capability.

[0120] The input of CFE-3 branch 2 is obtained by downsampling CFE-2 branch 1 by 2 times and fusing it with the branch 2 data. In this way, CFE-2 branch 1 is used to supplement some of the spatial resolution information lost in branch 2 due to downsampling, and it is used as the input of CFE-3 module branch 2.

[0121] The input of CFE-3 branch 3 is obtained by fusion after 4-fold downsampling of CFE-2 branch 1 and 2-fold downsampling of branch 2. This branch has the strongest global feature expression and abstract description capabilities.

[0122] Finally, the features of the three branches output by the CFE-3 module are fused and classified through the FFM-C module 3 (Feature Fusion Module-Classify). The features in branch 1 of the CFE-3 module have the best overlapping spatial resolution features, branch 2 has both spatial resolution features and abstract recognition features, and branch 3 has relatively strong overlapping abstract feature recognition capabilities and global feature expression capabilities. In combination with the above characteristics, the FFM-C module design includes a sampling layer. First, the features of different branches are sampled to unify the spatial scales between different features, and further feature fusion is performed through convolution. Finally, the Softmax activation function is used to map the features to the overlapping classification probability space to obtain overlapping recognition results of different channels.

[0123] Subsequently, the mask recognition results of different channels are compared with their true mask labels to calculate their respective loss values, and the sum of their respective loss values ​​is used as the total loss of the model for optimization.

[0124] The above is a description of the overall structure of the deep learning model for overlay recognition designed by the present invention. The constructed model will be trained based on the constructed deep learning sample library for overlay recognition, so that the model can learn the classification features of SAR image overlay from the sample library, which will be used for the subsequent recognition of SAR overlay in unknown areas.

[0125] Preferably, the loss function is:

[0126]

[0127] L branch =α·L CE +β·L IoU ;

[0128] L total =L b1 +L b2 +L b3 ;

[0129] Among them, L CE is a cross entropy loss function, which is used to measure the gap between the overlap probability predicted by the overlap recognition model and the true label; L IoU is the intersection-over-union loss function, which is used to calculate the intersection-over-union ratio of the predicted area of ​​the overlap recognition model and the actual overlap area; N is the number of samples; y i is the true label; p i is the predicted probability of the model; L branchis the deep learning model loss function of the overlap recognition model, which is used to integrate the results of the cross entropy loss function and the intersection-over-intersection loss function, and calculate the loss function of each branch channel according to the integrated result; α and β are the integrated weights of the cross entropy loss function and the intersection-over-intersection loss function respectively; L total is the overall loss function of the overlay recognition model, which is used to calculate the loss of each model output channel, and then sum the losses of all model output channels, and take the sum as the overall loss of the model. b1 , L b2 and L b3 are the losses of each model channel respectively.

[0130] The loss function is a key component for optimizing performance during model training. The loss function can be used to calculate the gap between the model-predicted overlapped area and the true overlapped area, and the model performance can be continuously improved by minimizing the gap. The present invention sets a combined loss function as the loss in the model training process according to the characteristics of the overlapped area, so that the model can obtain accurate overlapped category information and accurate spatial position information at the same time, as shown in formula (2):

[0131]

[0132] As shown in the above formula (2-c), the deep learning model loss function designed in the present invention for overlapping recognition is given by the cross entropy loss function L in formula (2-a): CE , and (2-b) intersection-over-union loss function L IoU It consists of two weighted parts, the weights of which are α and β respectively. In addition, N is the number of samples, and y i is the true label, p i is the predicted probability of the model. The cross entropy loss function (2-a) is used to measure the gap between the overlap probability predicted by the model and the true label. The loss function is calculated and optimized to constrain the model to obtain a more accurate overlap classification result. Overlap recognition requires not only accurate overlap category information, but also precise spatial position of the overlap. Therefore, the present invention calculates the intersection-over-union ratio of the model prediction area and the actual overlap area through the intersection-over-union loss function (2-b) to obtain accurate overlap feature spatial position information while optimizing this part of the loss. Finally, it is integrated through weights α and β in formula (2-c) to calculate the loss function of each branch channel. The loss of each model output channel is calculated by the above loss function, and the sum of the losses of all channels is taken as the overall loss of the model, as shown in formula (2-d).

[0133] The model is trained using a small batch stochastic gradient descent method, which updates the network weight parameters layer by layer through the gradient back propagation process. Considering the task characteristics of the mask recognition model, a larger learning rate (such as 0.01) is set at the initial stage of training, so that the model can quickly adapt to the characteristics of the mask samples in the sample library at the beginning of training. Later, during the training of the model, the learning rate is gradually adjusted and reduced according to the validation set loss and accuracy performance to avoid the situation where the model loss continues to oscillate near the optimal point and cannot reach the optimal point. Through the learning rate adjustment strategy shown above, the model can have a faster training convergence speed in the initial training stage, and can stably converge to the optimal point of the parameters as the training process progresses.

[0134] After the model converges and the value of the loss function becomes stable, save the optimal model weight parameter file at this time to complete the model training.

[0135] Preferably, after obtaining the overlap recognition result of the current detection area, it also includes: calculating the precision, recall rate and intersection-over-union ratio of the overlap recognition result based on the overlap recognition result of the current detection area, and then evaluating the model recognition accuracy of the overlap recognition model based on the precision, recall rate and intersection-over-union ratio.

[0136] 4. Model-based SAR mask recognition:

[0137] After the above main steps, a deep learning model for overlap recognition after training convergence is obtained, which has the optimal SAR image overlap recognition weight parameters. The SAR image overlap area of ​​unknown areas can be identified based on this model in the future. In the specific overlap recognition application, the image to be overlapped is first preprocessed in accordance with the training process. As described in step 1, the SAR image to be predicted is adjusted to a form consistent with the input data during training and external heterogeneous data is obtained to construct a SAR-DataCube data block. Secondly, the model structure is initialized, and the model weights obtained by training in step 3 are loaded. The preprocessed SAR-DataCube data is used as the model input to start the forward propagation process of the model. The predicted probability map of the overlap of the corresponding area is obtained through the forward propagation of the model, and then the probability threshold is set to visualize the probability map to obtain the overlap recognition result.

[0138] Finally, relevant accuracy indicators are selected to evaluate the overlap recognition results of the model recognition. The indicators used in the present invention are precision, recall and intersection over union. Specifically, as shown in formula (3):

[0139]

[0140] In formula 3, TP is the true positive, which indicates the number of samples correctly identified as overlapping areas, and TN is the true negative, which indicates the number of samples correctly identified as non-overlapping areas. FP is the false positive, which indicates the number of samples incorrectly identified as overlapping categories, and FN is the false negative, which indicates the number of samples incorrectly identified as non-overlapping areas. By using formula 3 above, we can calculate the precision, recall, and intersection-over-union indicators. The precision describes the accuracy of the model in identifying overlapping areas, the recall indicates the recall rate of the model in identifying overlapping areas, that is, the ability to capture overlapping areas, and iou reflects the positioning accuracy of the model in identifying overlapping areas. The above indicators can be used to comprehensively evaluate the accuracy and reliability of the model's overlapping recognition.

[0141] It can be seen that the present invention provides a SAR image overlap detection method, which has the following advantages compared with the existing overlap detection methods:

[0142] 1. Existing overlap recognition methods either use only single data, such as using a single data source such as SAR amplitude map, coherence map, etc. to perform overlap recognition and analysis, or use derived information (coherence coefficient map or interference phase) extracted from the original SAR image for homologous combination for overlap recognition, ignoring the role of external heterogeneous DEM data or surface classification coverage data. The SAR image overlap recognition method proposed in the present invention fuses homologous data with heterogeneous data to construct a SAR-Data Cube data block, and uses a deep learning model to realize the comprehensive utilization of external heterogeneous data, thereby improving the stability of overlap recognition in various application scenarios.

[0143] 2. In addition, the current overlap recognition and analysis methods are relatively simple, either using a simple threshold discrimination method or analyzing the geometric relationship of the image to identify overlaps. In essence, the features used to distinguish overlaps are relatively simple, and there are some deficiencies in distinguishing overlaps, and there is room for further improvement in the accuracy of overlap recognition. The present invention uses a multi-branch convolution operation to capture overlap features in the input data through the designed convolutional deep learning model. Convolution can efficiently utilize the spatial neighborhood correlation between pixels to obtain the overlap features of the image, and through a layer-by-layer deepening structure, gradually obtain overlap features from low-level to high-level, from local to global, and finally use high-level abstract global overlap features to identify overlaps. The overlap recognition method of the present invention makes full use of the deep abstract overlap features, and improves the detection accuracy of overlaps compared to traditional methods.

[0144] 3. In order to ensure that the constructed deep learning model is fully and reliably trained and learn about the diverse features of overlap in various complex scenarios, the present invention first designs a multi-method integrated overlap label production process, which ensures that the acquired overlap labels have high reliability. Secondly, a overlap sample enhancement method based on a random enhancement method pool is designed. This enhancement method ensures that the data enhancement method used for each overlap sample is not exactly the same. Compared with the traditional single and fixed data enhancement method, it maximizes the diversity of enhanced samples and enhancement methods. The above method can construct an overlap recognition sample library with abundant quantity and stable quality to improve the training effect of the overlap recognition model.

[0145] Embodiment 2

[0146] Please refer to Figure 6 , is a schematic structural diagram of a SAR image overlap detection device provided by an embodiment of the present invention, the device comprising: a data acquisition module, an amplitude information extraction module, an interference processing module and an overlap recognition module;

[0147] The data acquisition module is used to acquire the SAR time-series image data of the current detection area and the corresponding external heterogeneous data; wherein the external heterogeneous data includes: external DEM data and external surface classification coverage data;

[0148] The amplitude information extraction module is used to extract corresponding amplitude information from the SAR time-series image data, obtain a regional amplitude map of the current detection area according to the extracted amplitude information, and perform normalization processing on the regional amplitude map;

[0149] The interference processing module is used to perform interference processing on the SAR time-series image data to obtain a corresponding interference phase map, and calculate a corresponding coherence coefficient according to the interference phase map;

[0150] The overlap recognition module is used to normalize the external DEM data, and then band-splicing the normalized regional amplitude map, interference phase map, coherence coefficient, normalized external DEM data and external surface classification coverage data to generate corresponding data blocks, and input the data blocks into a preset overlap recognition model, so that the overlap recognition model can recognize the overlap of the current detection area according to the input data blocks, and obtain the overlap recognition result of the current detection area; wherein the overlap recognition model is obtained by training a preset deep learning model with historical SAR time series image data and corresponding historical external heterogeneous data as input and the overlap recognition result of the corresponding area as output.

[0151] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0152] Those skilled in the art can clearly understand that for the sake of convenience and brevity, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0153] Embodiment 3

[0154] Accordingly, an embodiment of the present invention provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the SAR image overlap detection method described in the above-mentioned embodiment of the invention when executing the computer program.

[0155] The electronic device may be a computing device such as a desktop computer, a notebook, a palm computer, a cloud server, etc. The device may include, but is not limited to, a processor and a memory.

[0156] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the device, and various interfaces and lines are used to connect various parts of the entire device.

[0157] Embodiment 4

[0158] Accordingly, an embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the SAR image overlap detection method described in the above-mentioned embodiment of the invention.

[0159] The memory can be used to store the computer program, and the processor realizes various functions of the device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0160] The storage medium is a computer-readable storage medium, and the computer program is stored in the computer-readable storage medium. When the computer program is executed by the processor, the steps of each method embodiment described above can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0161] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A SAR image overlap detection method, characterized in that: include: Acquire SAR time-series image data of the current detection area and corresponding external heterogeneous data; wherein the external heterogeneous data includes: external DEM data and external surface classification coverage data; Extracting corresponding amplitude information from the SAR time-series image data, obtaining a regional amplitude map of the current detection area according to the extracted amplitude information, and performing normalization processing on the regional amplitude map; Performing interference processing on the SAR time-series image data to obtain a corresponding interference phase map, and calculating a corresponding coherence coefficient according to the interference phase map; The external DEM data is normalized, and then the normalized regional amplitude map, interference phase map, coherence coefficient, normalized external DEM data and external surface classification coverage data are band-joined to generate corresponding data blocks, and the data blocks are input into a preset overlap recognition model, so that the overlap recognition model recognizes the overlap of the current detection area according to the input data blocks, and obtains the overlap recognition result of the current detection area; The overlap recognition model is obtained by training a preset deep learning model with historical SAR time-series image data and corresponding historical external heterogeneous data as input and overlap recognition results of the corresponding area as output.

2. The SAR image overlap detection method according to claim 1, characterized in that: The generation of the overlap recognition model includes: Acquire historical SAR time-series image data and corresponding historical external heterogeneous data; wherein the historical external heterogeneous data includes: historical external DEM data and historical external surface classification coverage data; Extracting corresponding historical amplitude information from the historical SAR time-series image data, obtaining a historical regional amplitude map of the corresponding area according to the extracted historical amplitude information, and normalizing the historical regional amplitude map; Performing interference processing on the historical SAR time-series image data to obtain a corresponding historical interference phase map, and calculating a corresponding historical coherence coefficient based on the historical interference phase map; Normalizing the historical external DEM data, and then band-joining the normalized historical regional amplitude map, the historical interference phase map, the historical coherence coefficient, the normalized historical external DEM data, and the historical external surface classification coverage data to generate corresponding data blocks, and using the data blocks as input data samples; Marking the overlapping areas in the historical SAR time series image data to generate overlapping labels for the corresponding areas; A model training sample is generated according to the input data sample and the overlap label, and then a preset deep learning model is trained according to the model training sample and a preset loss function to generate a corresponding overlap recognition model.

3. The SAR image overlap detection method according to claim 2, characterized in that: The step of marking the overlapped areas in the historical SAR time series image data and generating overlapped labels for the corresponding areas includes: According to different preset overlap recognition methods, SAR image overlap recognition is performed on the historical SAR time series image data to obtain different overlap recognition results; wherein the preset overlap recognition methods include: an overlap recognition method based on amplitude and coherence threshold, a satellite radar line of sight analysis recognition method, and a DEM slope and aspect analysis method; The overlap recognition results are integrated, and then the overlap areas in the historical SAR time series image data are marked according to the integrated overlap recognition results to generate overlap labels for the corresponding areas.

4. The SAR image overlap detection method according to claim 2, characterized in that: Generating a model training sample according to the input data sample and the overlapped label includes: Using the input data sample and the overlapped label as initial model training samples; For each sample data in the initial model training sample, several data enhancement methods are randomly selected from the preset data enhancement methods, and data enhancement processing is performed on the sample data in turn to obtain sample data after data enhancement; wherein the data enhancement methods include: geometric transformation, noise injection, image aliasing and color transformation; All sample data after data augmentation are used as model training samples.

5. The SAR image overlap detection method according to claim 1, characterized in that: The step of identifying the overlap of the current detection area according to the input data block to obtain the overlap identification result of the current detection area includes: Perform image-level data fusion on the input data blocks; Perform shallow feature extraction on the fused data, and perform feature fusion on the extracted shallow features; According to the fused shallow features, the global features in the fused shallow features are extracted, and then the extracted global features are subjected to feature fusion and feature classification to obtain the overlapping recognition result of the current detection area.

6. The SAR image overlap detection method according to claim 2, characterized in that: The loss function is: L branch =α·L CE +β·L IoU ; L total =L b1 +L b2 +L b3 ; Among them, L CE is a cross entropy loss function, which is used to measure the gap between the overlap probability predicted by the overlap recognition model and the true label; L IoU is the intersection-over-union loss function, which is used to calculate the intersection-over-union ratio of the predicted area of ​​the overlap recognition model and the actual overlap area; N is the number of samples; y i is the true label; p i is the predicted probability of the model; L branch is the deep learning model loss function of the overlap recognition model, which is used to integrate the results of the cross entropy loss function and the intersection-over-intersection loss function, and calculate the loss function of each branch channel according to the integrated result; α and β are the integrated weights of the cross entropy loss function and the intersection-over-intersection loss function respectively; L total is the overall loss function of the overlay recognition model, which is used to calculate the loss of each model output channel, and then sum the losses of all model output channels, and take the sum as the overall loss of the model. b1 , L b2 and L b3 are the losses of each model channel respectively.

7. The SAR image overlap detection method according to claim 1, characterized in that: After obtaining the overlap recognition result of the current detection area, it also includes: According to the overlap recognition result of the current detection area, the precision, recall and intersection-over-union of the overlap recognition result are calculated, and then the model recognition accuracy of the overlap recognition model is evaluated according to the precision, recall and intersection-over-union.

8. A SAR image overlap detection device, characterized in that: include: Data acquisition module, amplitude information extraction module, interference processing module and overlap recognition module; The data acquisition module is used to acquire the SAR time-series image data of the current detection area and the corresponding external heterogeneous data; wherein the external heterogeneous data includes: external DEM data and external surface classification coverage data; The amplitude information extraction module is used to extract corresponding amplitude information from the SAR time-series image data, obtain a regional amplitude map of the current detection area according to the extracted amplitude information, and perform normalization processing on the regional amplitude map; The interference processing module is used to perform interference processing on the SAR time-series image data to obtain a corresponding interference phase map, and calculate a corresponding coherence coefficient according to the interference phase map; The overlap recognition module is used to normalize the external DEM data, and then band-splicing the normalized regional amplitude map, interference phase map, coherence coefficient, normalized external DEM data and external surface classification coverage data to generate corresponding data blocks, and input the data blocks into a preset overlap recognition model, so that the overlap recognition model can recognize the overlap of the current detection area according to the input data blocks, and obtain the overlap recognition result of the current detection area; wherein the overlap recognition model is obtained by training a preset deep learning model with historical SAR time series image data and corresponding historical external heterogeneous data as input and the overlap recognition result of the corresponding area as output.

9. An electronic device, characterized in that: The invention comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the SAR image overlap detection method according to any one of claims 1 to 7 is implemented.

10. A storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is executed, the device where the storage medium is located is controlled to execute the SAR image overlap detection method according to any one of claims 1 to 7.

Images