SAR (Synthetic Aperture Radar) image change detection method and device for reducing interference

Abstract

The invention relates to the technical field of photogrammetry, in particular to an SAR image change detection method and device capable of reducing interference, and the method comprises the steps: recognizing each pixel point of a to-be-detected SAR image, and generating a patch of a target size based on the center of each pixel point; inputting the patch into a preset coding model to obtain a coded feature map of which the interference degree is reduced to a target interference degree; and inputting the encoded feature map into a preset decoding model to generate a cascade feature map, and generating a change detection result of the SAR image based on the cascade feature map. The feature map of the to-be-detected SAR image can be converted from the spatial domain to the frequency domain by using the coding model, so that the high-frequency noise of the feature map is separated and coded, the spatial information and the position information are obtained, the influence of the noise on change detection is reduced, and the detection accuracy is improved. And then the decoding model can be used for bridging the multi-scale features and converting the feature map from the frequency domain to the spatial domain, so that the accuracy of SAR image change detection is effectively improved.

Description

Technical Field

[0001] The present application relates to the field of photogrammetry technology, and in particular to a SAR image change detection method and device with reduced interference. Background Art

[0002] In related technologies, SAR images are formed using microwave technology and can acquire images around the clock and in all weather conditions. Due to its special imaging mechanism, the interaction between radar waves and surface objects will cause phase differences between different waves, resulting in random fluctuations in intensity in the received image, thereby generating coherent speckle noise. Coherent speckle noise is multiplicative noise that is proportional to the signal amplitude, so it exhibits different noise intensities in different areas of the image, causing the texture and details of the image to be obscured, the image to become blurred, and the accuracy of change detection to be affected.

[0003] However, in the actual detection of SAR image changes in related technologies, coherent speckle noise may be mistaken for changes and confused with real changes, resulting in false positive results, reducing the accuracy of change detection and making real changes difficult to identify, which urgently needs to be solved. Summary of the invention

[0004] The present application provides a SAR image change detection method and device with reduced interference, so as to solve the problem that in the actual detection of SAR image changes in the related art, coherent speckle noise may be mistaken for change and confused with the real change, resulting in false positive results, reducing the accuracy of change detection, and making it difficult to identify the real change.

[0005] The first aspect of the present application provides a SAR image change detection method with reduced interference, comprising the following steps: identifying each pixel point of the SAR image to be detected, and generating a patch of a target size based on the center of each pixel point; inputting the patch into a preset encoding model to obtain an encoded feature map with an interference level reduced to a target interference level; inputting the encoded feature map into a preset decoding model to generate a cascade feature map, and generating a change detection result of the SAR image based on the cascade feature map.

[0006] Optionally, in one embodiment of the present application, before inputting the patch into the preset coding model, it also includes: extracting an initial feature map of the target image, converting the initial feature map from the spatial domain to the frequency domain to obtain a feature map in the frequency domain; processing the feature map in the frequency domain to obtain a processed feature map, and inputting the processed feature map into the target network model to output an intermediate feature map to generate the preset coding model.

[0007] Optionally, in one embodiment of the present application, before inputting the encoded feature map into the preset decoding model to generate the cascade feature map, it also includes: splicing the intermediate feature map to obtain a complete feature map; converting the complete feature map from the frequency domain back to the spatial domain, and restoring the size of the complete feature map to the original size to generate the preset decoding model.

[0008] Optionally, in one embodiment of the present application, it also includes: inputting the processed feature map into the target network model to obtain spatial information and position information of the processed feature map; and generating the preset coding model based on the spatial information and the position information.

[0009] Optionally, in one embodiment of the present application, the splicing of the intermediate feature maps to obtain a complete feature map includes: using multi-scale attention to obtain multi-scale features of each feature map in the splicing process; and generating the complete feature map on the channel dimension of the target network model based on the multi-scale features.

[0010] The second aspect of the present application provides a SAR image change detection device with reduced interference, including: an identification module, used to identify each pixel point of the SAR image to be detected, and generate a patch of a target size based on the center of each pixel point; a first processing module, used to input the patch into a preset coding model to obtain an encoded feature map with an interference level reduced to a target interference level; a detection module, used to input the encoded feature map into a preset decoding model to generate a cascade feature map, and generate a change detection result of the SAR image based on the cascade feature map.

[0011] Optionally, in one embodiment of the present application, it also includes: a first conversion module, used to extract the initial feature map of the target image before inputting the patch into the preset coding model, and convert the initial feature map from the spatial domain to the frequency domain to obtain a feature map in the frequency domain; a second processing module, used to process the feature map in the frequency domain to obtain a processed feature map, and input the processed feature map into the target network model to output an intermediate feature map to generate the preset coding model.

[0012] Optionally, in one embodiment of the present application, it also includes: a splicing module, used to splice the intermediate feature map to obtain a complete feature map before inputting the encoded feature map into the preset decoding model to generate the cascade feature map; a second conversion module, used to convert the complete feature map from the frequency domain back to the spatial domain, and restore the size of the complete feature map to the original size to generate the preset decoding model.

[0013] Optionally, in one embodiment of the present application, it also includes: an acquisition module, used to input the processed feature map into the target network model to obtain the spatial information and position information of the processed feature map; a generation module, used to generate the preset coding model based on the spatial information and the position information.

[0014] Optionally, in one embodiment of the present application, the stitching module includes: an acquisition unit for acquiring multi-scale features of each feature map in the stitching process using multi-scale attention; and a generation unit for generating the complete feature map on the channel dimension of the target network model based on the multi-scale features.

[0015] A third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the SAR image change detection method with reduced interference as described in the above embodiment.

[0016] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, which, when executed by a processor, implements the above-mentioned SAR image change detection method for reducing interference.

[0017] A fifth aspect of the present application provides a computer program product, including a computer program, which, when executed, is used to implement the above-mentioned SAR image change detection method for reducing interference.

[0018] The embodiment of the present application can use a certain coding model to convert the feature map of the SAR image to be detected from the spatial domain to the frequency domain, thereby separating the high-frequency noise of the feature map and encoding to obtain spatial information and position information, and then using a decoding model to bridge the multi-scale features and convert the feature map from the frequency domain back to the spatial domain, thereby completing the SAR image change detection. Thus, it is achieved by separating the high-frequency noise and extracting higher-dimensional information using multi-layer convolution, reducing the impact of noise on change detection, and inputting it into the ConvMixer module to obtain mixed spatial and position information, and then using a multi-scale attention gate to splice the underlying features and high-level features, and adaptively select encoder features according to different resolutions, thereby improving the accuracy of SAR image change detection. Thus, it solves the problem that in the actual detection of SAR image changes in the related art, coherent speckle noise may be mistaken for change, confused with the actual change, resulting in false positive results, reducing the accuracy of change detection, and making it difficult to identify the real change.

[0019] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0021] Figure 1 A flowchart of a SAR image change detection method for reducing interference provided in accordance with an embodiment of the present application;

[0022] Figure 2 This is a schematic diagram of the principle of a coding model of an embodiment of the present application;

[0023] Figure 3 A schematic diagram of the principle of a decoding model according to an embodiment of the present application;

[0024] Figure 4 A schematic diagram of the structure of a SAR image change detection device for reducing interference provided in an embodiment of the present application;

[0025] Figure 5 It is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present application.

[0026] Reference numerals:

[0027] 10-SAR image change detection device with reduced interference: 100-recognition module, 200-first processing module and 300-detection module; 501-memory, 502-processor and 503-communication interface. DETAILED DESCRIPTION

[0028] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0029] The following describes the interference-reducing SAR image change detection method and device of the embodiment of the present application with reference to the accompanying drawings. In view of the problem that in the actual detection of SAR image changes in the related art mentioned in the above background technology, coherent speckle noise may be mistaken for changes and confused with real changes, resulting in false positive results, reducing the accuracy of change detection, and making it difficult to identify real changes, the present application provides a SAR image change detection method with interference reduction, in which a certain encoding model can be used to convert the feature map of the SAR image to be detected from the spatial domain to the frequency domain, thereby separating the high-frequency noise of the feature map and encoding to obtain spatial information and position information, and then the decoding model can be used to bridge the multi-scale features and convert the feature map from the frequency domain back to the spatial domain, thereby completing the SAR image change detection. Thus, it is achieved by separating the high-frequency noise and extracting higher-dimensional information by multi-layer convolution, reducing the influence of noise on change detection, and inputting it into the ConvMixer module to obtain mixed spatial and position information, and then using the multi-scale attention gate to splice the bottom-level features and the high-level features, and adaptively selecting the encoder features according to different resolutions, thereby improving the accuracy of SAR image change detection. This solves the problem that in the actual detection of SAR image changes in the related technology, coherent speckle noise may be mistaken for change and confused with real changes, resulting in false positive results, reducing the accuracy of change detection, and making real changes difficult to identify.

[0030] Specifically, Figure 1 A flow chart of a SAR image change detection method for reducing interference provided in an embodiment of the present application.

[0031] like Figure 1 As shown, the SAR image change detection method for reducing interference includes the following steps:

[0032] In step S101 , each pixel of the SAR image to be detected is identified, and a patch of a target size is generated based on the center of each pixel.

[0033] It can be understood by those skilled in the art that SAR images here refer to Synthetic Aperture Radar images, which are images obtained by remote sensing imaging technology using radar signals. Pixels are the basic unit of image display and the smallest element that makes up a digital image. Imagine an image as a puzzle composed of many small squares, which are pixels. For example, a simple black and white photo looks like a complete portrait or landscape from a distance, but if you zoom in enough, you will see small squares, which are pixels.

[0034] In some embodiments, when identifying changes in SAR images, there should be at least a reference SAR image and a comparison SAR image, so that the SAR changes can be detected. The SAR images to be detected here can be understood as at least two SAR images that need to be detected for image changes.

[0035] Identify each pixel of the SAR image to be detected. It can be understood that the SAR image is composed of many small squares (pixels), and each small square is marked or located. These pixels contain various information about the observed scene, such as the reflection characteristics of the ground objects.

[0036] A patch here can be understood as a small image block. The target size patch is generated based on the center of each pixel, which means taking a specific size area at the center of each pixel as a small image block. If the center of the current pixel is a coordinate, then the patch is a small square area of ​​M×N pixels with a size of 5, centered on the center of the current pixel, and containing a certain range of pixels above, below, left, and right. The pixels in this small square area constitute the content of the patch.

[0037] For example, the target size is defined as M×N patches of size 5, that is, M×N 5×5 square areas or areas of other shapes but with a side length dimension of 5. The size of 5 here can be understood as the number of pixels in the patch in the horizontal and vertical directions. For example, a small image block consisting of 2 pixels above, below, left and right (plus the center pixel itself, which just forms a 5×5 area) with a certain pixel as the center is a patch of size 5.

[0038] It should be noted that the specific target size of the patch can be set or adjusted by professionals in the field according to actual conditions. This is only an example and is not specifically limited.

[0039] The embodiment of the present application can generate a patch of target size with each pixel point of the SAR image to be detected as the center, thereby making it easier to analyze the features of the pixels in each small area, facilitating target detection or feature extraction in subsequent processes.

[0040] Step S102, inputting the patch into a preset coding model to obtain a coded feature map with the interference level reduced to a target interference level.

[0041] It can be understood that the preset coding model here refers to a certain pre-set coding model, wherein the coding model can convert the characteristics of the SAR image to be detected in the spatial domain to the frequency domain, thereby separating the high-frequency component of the SAR image to be detected, that is, the high-frequency noise, and reducing the noise interference of the SAR image to be detected to the target interference, thereby reducing the influence of noise on the change detection of the SAR image to be detected. The encoded feature map here can be understood as the feature map obtained after the SRA image to be detected is input into a certain coding model for processing.

[0042] Among them, the target interference degree can be understood as the interference degree reached after the noise interference degree of the SAR image to be detected is reduced. The feature map here refers to an important concept in the convolutional neural network (CNN). When processing the SAR image to be detected, the convolution layer slides the convolution kernel on the input SAR image to be detected (here is a pixel patch with a size of , two channels) to perform a convolution operation. The result of each convolution operation constitutes a new matrix, which is the feature map.

[0043] Next, the construction process of the preset coding model in the embodiment of the present application is further explained.

[0044] Optionally, in one embodiment of the present application, before inputting the patch into a preset coding model, it also includes: extracting an initial feature map of the target image, converting the initial feature map from the spatial domain to the frequency domain to obtain a feature map in the frequency domain; processing the feature map in the frequency domain to obtain a processed feature map, and inputting the processed feature map into the target network model to output an intermediate feature map to generate a preset coding model.

[0045] It can be understood that the target images here can be understood as some SAR images used when constructing a certain coding model, which can be SAR images to be detected or some other SAR images that can perform change detection.

[0046] Figure 2 This is a schematic diagram of the coding model principle of an embodiment of the present application. Figure 2 As shown, in the actual execution process, when constructing a certain encoding model, the embodiment of the present application can first extract the initial feature map of the target image, wherein the initial feature map here can be understood as a feature map of the target image directly extracted by the initial stage of a neural network such as a convolutional layer without further processing.

[0047] For example, we input a pixel patch of the target image (image size H×W, two channels), perform convolution, batch normalization, and ReLU activation on the image, repeat all the above processes twice, and use max pooling for downsampling to obtain The feature map of size is the initial feature map.

[0048] After obtaining the initial feature map, the embodiment of the present application can convert the initial feature map from the spatial domain to the frequency domain. For example, discrete wavelet transform is used to convert the initial feature map from the spatial domain to the frequency domain, thereby obtaining a feature map in the frequency domain.

[0049] After obtaining the feature graph in the frequency domain, the embodiment of the present application further processes the feature graph in the frequency domain, that is, divides the frequency domain features of the feature graph in the frequency domain into high-frequency features and low-frequency features, and analyzes the high-frequency features to separate the high-frequency noise. It should be noted that in the embodiment of the present application, the size of the feature graph after conversion to the frequency domain and separation of the high-frequency noise remains the same as the size of the initial feature graph.

[0050] Then, the feature map separated from the high-frequency noise is input into the downsampling module composed of convolution, batch normalization, ReLU activation function and maximum pooling layer for processing and repeated twice. The sizes of the feature maps outputted twice can be expressed as:

[0051] The processed feature map can be input into the target network model and output the intermediate feature map in the embodiment of the present application. The target network model can be understood as a related network model that can extract features and fuse information in the feature map, for example, a ConMixer module.

[0052] Therefore, the embodiment of the present application can integrate the above process to extract higher-dimensional information through multi-layer convolution, reduce noise interference when detecting changes in the target image, and thus generate a certain coding model.

[0053] Optionally, in one embodiment of the present application, it also includes: inputting the processed feature map into the target network model to obtain spatial information and position information of the processed feature map; generating a preset coding model based on the spatial information and position information.

[0054] Based on the relevant descriptions of other embodiments, it can be understood that the present application will input the processed feature map into the target network model to obtain an intermediate feature map and output it, so that the output intermediate feature map can be input into a certain decoding model for processing.

[0055] For example, the present application can, but is not limited to, inputting the processed feature map into the ConMixer module, thereby obtaining the spatial information and position information of the feature map, reducing the impact of noise, and completing the construction of the entire coding model.

[0056] Step S103: input the encoded feature map into a preset decoding model to generate a cascade feature map, and generate a change detection result of the SAR image based on the cascade feature map.

[0057] It can be understood that the preset decoding model here refers to a certain pre-built decoding model, which can use a multi-scale attention gate to splice the underlying features and high-level features, and adaptively select encoder features according to different resolutions, thereby improving the accuracy of change detection.

[0058] As a possible implementation method, the present application can input the encoded feature map into a certain decoding model to generate a cascade feature map of the SRA image to be detected, and then obtain the change detection result of the SRA image to be detected based on the cascade feature map. The change detection result is the difference result between two or more SRA images to be detected.

[0059] Among them, cascaded feature maps refer to the result of connecting multiple feature maps processed at different levels or stages in sequence. In the convolutional neural network (CNN) architecture of deep learning, as the network depth increases, each layer will generate a feature map, which contains different levels of information gradually extracted from the original image.

[0060] For example, in a simple three-layer CNN, the first layer may extract basic low-level information such as edges and lines of the image to obtain features Figure 1 ; The second layer further extracts some simple shape information based on this to obtain the feature Figure 2 The third layer may extract more complex object parts or information about the object as a whole to obtain features Figure 3 . Cascading these three feature maps together forms a cascade feature map.

[0061] Cascading feature maps can effectively fuse information at different levels. Low-level feature maps contain basic texture, edge and other information of the image, while high-level feature maps contain more abstract overall or semantic information of the object. Through cascading, the network can use this information at the same time, so that the SAR image to be detected can be used for change detection based on a more comprehensive feature representation, and obtain more accurate change detection results.

[0062] It should be noted that the encoding model and the decoding model in the embodiment of the present application do not necessarily exist separately, and the encoding model and the decoding model can also be integrated into a certain network, and then the network is used to realize the change detection of the SAR image to be detected.

[0063] Optionally, in one embodiment of the present application, before the encoded feature map is input into a preset decoding model to generate a cascade feature map, it also includes: splicing the intermediate feature maps to obtain a complete feature map; converting the complete feature map from the frequency domain back to the spatial domain, and restoring the size of the complete feature map to the original size to generate a preset decoding model.

[0064] Based on the relevant descriptions of other embodiments, it can be understood that the present application can input the encoded feature map into a certain decoding model to generate a cascade feature map.

[0065] Prior to this, the application needs to splice the intermediate feature maps finally output by the target network model in a certain encoding model to obtain a complete feature map; then perform convolution, batch normalization and Relu activation function operations on the complete feature map and repeat them twice; finally, perform inverse discrete wavelet transform on the complete feature map, convert the complete feature map from the frequency domain back to the spatial domain and restore the size of the complete feature map to the original size to complete the image decoding, thereby generating a certain decoding model. Figure 3 As shown, Figure 3 That is a schematic diagram of the principle of a decoding model of an embodiment of the present application.

[0066] It should be noted that the intermediate feature map in the embodiment of the present application and the encoded feature map in the previous embodiment are both obtained by outputting a certain coding model, wherein the intermediate feature map here refers to the feature map output during the training (construction) process of the coding model, and the encoded feature map refers to the feature map output when the coding model is actually applied.

[0067] Optionally, in one embodiment of the present application, the intermediate feature maps are spliced ​​to obtain a complete feature map, including: using multi-scale attention to obtain multi-scale features of each feature map in the splicing process; based on the multi-scale features, generating a complete feature map on the channel dimension of the target network model.

[0068] In the actual implementation process, when splicing the intermediate feature maps, the embodiment of the present application mainly uses multi-scale attention to obtain the multi-scale features of each feature map in the splicing process, and then uses jump connections in the channel dimension of the target network model to bridge the multi-scale information of the feature map in the encoding model with the decoded information to achieve the splicing of the intermediate feature maps, thereby realizing the multi-level utilization of multi-scale features. The specific process can be expressed as follows:

[0069] (1) Upsample, convolve, batch normalize, and activate the target network model to obtain a feature map of size The feature map of

[0070] (2) The encoding model The feature maps of different sizes are input into the multi-scale attention gate, and after obtaining the multi-scale information of the same size, they are concatenated with the feature maps generated by (1) in the channel dimension.

[0071] (3) Perform convolution, batch normalization, and ReLU activation functions on the initially stitched image. This process is repeated twice, and then upsampling, convolution, batch normalization, and ReLU activation functions are performed to obtain a size of The feature map of

[0072] (4) The encoding model The feature map is input into the multi-scale attention gate to obtain multi-scale information of the same size, and then it is concatenated with the feature map generated by (3) on the channel dimension to obtain a complete feature map.

[0073] It should be noted that the complete feature map and the previous cascade feature map in the embodiments of the present application both refer to the spliced ​​feature maps, wherein the complete feature map here refers to the spliced ​​feature map during the training (construction) process of the decoding model, and the cascade feature map refers to the spliced ​​feature map when the decoding model is actually applied.

[0074] According to the SAR image change detection method for reducing interference proposed in the embodiment of the present application, a certain coding model can be used to convert the feature map of the SAR image to be detected from the spatial domain to the frequency domain, thereby separating the high-frequency noise of the feature map and encoding to obtain spatial information and position information, and then the decoding model can be used to bridge the multi-scale features and convert the feature map from the frequency domain back to the spatial domain, thereby completing the SAR image change detection. Thus, it is achieved by separating the high-frequency noise and extracting higher-dimensional information by multi-layer convolution, reducing the influence of noise on change detection, and inputting it into the ConvMixer module to obtain mixed spatial and position information, and then using the multi-scale attention gate to splice the bottom-level features and the high-level features, and adaptively selecting the encoder features according to different resolutions, thereby improving the accuracy of SAR image change detection. Thus, the problem that in the actual detection of SAR image changes in the related art, coherent speckle noise may be mistaken for change, confused with the actual change, resulting in false positive results, reducing the accuracy of change detection, and making it difficult to identify the real change is solved.

[0075] Next, the SAR image change detection device for reducing interference proposed in an embodiment of the present application is described with reference to the accompanying drawings.

[0076] Figure 4 It is a schematic diagram of the structure of a SAR image change detection device for reducing interference according to an embodiment of the present application.

[0077] like Figure 4As shown, the SAR image change detection device 10 for reducing interference includes: a recognition module 100 , a processing module 200 and a detection module 300 .

[0078] The recognition module 100 is used to recognize each pixel of the SAR image to be detected and generate a patch of the target size based on the center of each pixel;

[0079] A first processing module 200 is used to input the patch into a preset coding model to obtain a coded feature map with the interference level reduced to a target interference level;

[0080] The detection module 300 is used to input the encoded feature map into a preset decoding model to generate a cascade feature map, and generate a change detection result of the SAR image based on the cascade feature map.

[0081] Optionally, in one embodiment of the present application, it further includes: a first conversion module and a second processing module.

[0082] The first conversion module is used to extract an initial feature map of the target image before inputting the patch into the preset coding model, and convert the initial feature map from the spatial domain to the frequency domain to obtain a feature map in the frequency domain;

[0083] The second processing module is used to process the feature map in the frequency domain to obtain the processed feature map, and input the processed feature map into the target network model to output the intermediate feature map to generate a preset coding model.

[0084] Optionally, in one embodiment of the present application, it further includes: a splicing module and a second conversion module.

[0085] The splicing module is used to splice the intermediate feature maps to obtain the complete feature map before inputting the encoded feature map into a preset decoding model to generate a cascade feature map;

[0086] The second conversion module is used to convert the complete feature map from the frequency domain back to the spatial domain, and restore the size of the complete feature map to the original size to generate a preset decoding model.

[0087] Optionally, in one embodiment of the present application, it also includes: an acquisition module and a generation module.

[0088] The acquisition module is used to input the processed feature map into the target network model to obtain the spatial information and position information of the processed feature map;

[0089] A generation module is used to generate a preset coding model based on spatial information and position information.

[0090] Optionally, in one embodiment of the present application, the splicing module includes: an acquisition unit and a generation unit.

[0091] Among them, the acquisition unit is used to obtain multi-scale features of each feature map in the splicing process using multi-scale attention;

[0092] The generation unit is used to generate a complete feature map in the channel dimension of the target network model based on the multi-scale features.

[0093] It should be noted that the above explanations and descriptions of the embodiment of the SAR image change detection method with reduced interference are also applicable to the SAR image change detection device with reduced interference of this embodiment, and will not be described in detail here.

[0094] According to the SAR image change detection device for reducing interference proposed in the embodiment of the present application, a certain coding model can be used to convert the feature map of the SAR image to be detected from the spatial domain to the frequency domain, thereby separating the high-frequency noise of the feature map and encoding to obtain spatial information and position information, and then the decoding model can be used to bridge the multi-scale features and convert the feature map from the frequency domain back to the spatial domain, thereby completing the SAR image change detection. Thus, it is achieved by separating the high-frequency noise and extracting higher-dimensional information by multi-layer convolution, reducing the influence of noise on change detection, and inputting it into the ConvMixer module to obtain mixed spatial and position information, and then using the multi-scale attention gate to splice the bottom-level features and the high-level features, and adaptively selecting the encoder features according to different resolutions, thereby improving the accuracy of SAR image change detection. Thus, the problem that the coherent speckle noise may be mistaken for change and confused with the actual change in the actual detection of SAR image changes in the related technology is solved, resulting in false positive results, reducing the accuracy of change detection, and making it difficult to identify the real change.

[0095] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0096] A memory 501 , a processor 502 , and a computer program stored in the memory 501 and executable on the processor 502 .

[0097] When the processor 502 executes the program, the SAR image change detection method with reduced interference provided in the above embodiment is implemented.

[0098] Furthermore, the electronic device further comprises:

[0099] The communication interface 503 is used for communication between the memory 501 and the processor 502 .

[0100] The memory 501 is used to store computer programs that can be executed on the processor 502 .

[0101] The memory 501 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0102] If the memory 501, the processor 502 and the communication interface 503 are implemented independently, the communication interface 503, the memory 501 and the processor 502 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0103] Optionally, in a specific implementation, if the memory 501, the processor 502 and the communication interface 503 are integrated on a chip, the memory 501, the processor 502 and the communication interface 503 can communicate with each other through an internal interface.

[0104] The processor 502 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0105] The embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned SAR image change detection method for reducing interference.

[0106] The embodiment of the present application also provides a computer program product, including a computer program, which can run computer instructions. When the computer instructions are executed by a processor, the SAR image change detection method with reduced interference provided in the embodiment of the present application is implemented.

[0107] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0108] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0109] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0110] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways as necessary and then storing it in a computer memory.

[0111] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of multiple of the following technologies known in the art: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0112] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0113] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0114] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A SAR image change detection method with reduced interference, characterized in that: The following steps are involved: Identify each pixel of the SAR image to be detected, and generate a patch of the target size based on the center of each pixel; Inputting the patch into a preset coding model to obtain a coded feature map with interference reduced to a target interference level; The encoded feature map is input into a preset decoding model to generate a cascade feature map, and a change detection result of the SAR image is generated based on the cascade feature map.

2. The method according to claim 1, characterized in that Before inputting the patch into the preset coding model, the method further includes: Extracting an initial feature map of the target image, and converting the initial feature map from the spatial domain to the frequency domain to obtain a feature map in the frequency domain; The feature map in the frequency domain is processed to obtain a processed feature map, and the processed feature map is input into a target network model to output an intermediate feature map to generate the preset coding model.

3. The method according to claim 2, characterized in that Before inputting the encoded feature map into the preset decoding model to generate the cascade feature map, the method further includes: Splicing the intermediate feature maps to obtain a complete feature map; The complete feature map is converted from the frequency domain back to the spatial domain, and the size of the complete feature map is restored to the original size to generate the preset decoding model.

4. The method according to claim 2, characterized in that: Also includes: Inputting the processed feature map into the target network model to obtain spatial information and position information of the processed feature map; The preset coding model is generated based on the spatial information and the position information.

5. The method according to claim 3, characterized in that: The step of splicing the intermediate feature maps to obtain a complete feature map includes: Use multi-scale attention to obtain multi-scale features of each feature map in the splicing process; Based on the multi-scale features, the complete feature map is generated on the channel dimension of the target network model.

6. A SAR image change detection device with reduced interference, characterized in that: include: A generation module, used for identifying each pixel point of the SAR image to be detected, and generating a patch of target size based on the center of each pixel point; A first processing module, configured to input the patch into a preset coding model to obtain a coded feature map with an interference level reduced to a target interference level; The detection module is used to input the encoded feature map into a preset decoding model to generate a cascade feature map, and generate a change detection result of the SAR image based on the cascade feature map.

7. The device according to claim 6, characterized in that Also includes: A conversion module, used for extracting an initial feature map of a target image before inputting the patch into the preset coding model, and converting the initial feature map from a spatial domain to a frequency domain to obtain a feature map in the frequency domain; The second processing module is used to process the feature map in the frequency domain to obtain a processed feature map, and input the processed feature map into the target network model to output an intermediate feature map to generate the preset coding model.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the SAR image change detection method with reduced interference as described in any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the SAR image change detection method with reduced interference as described in any one of claims 1 to 5.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed, it is used to implement the SAR image change detection method with reduced interference as described in any one of claims 1 to 5.

Images