Detection threshold adaptive change detection method

Abstract

The invention provides a detection threshold adaptive change detection method, and relates to the technical field of remote sensing image analysis, and the method comprises the steps: carrying out the super-cooperative super-pixel segmentation of a remote sensing image before and after the change, and obtaining the super-pixel segmentation results of the remote sensing image before and after the change; calculating an affinity matrix of the remote sensing image before and after the change; selecting a super-pixel-level training sample according to the super-pixel segmentation result of the remote sensing image before and after the change and the affinity matrix; training a Copula function through the super-pixel-level training sample, and generating an adaptive change detection threshold value by using the trained Copula function; and calculating the test statistics of the remote sensing image before and after the change through the trained Copula function, and carrying out detection judgment on the test statistics by using an adaptive change detection threshold to obtain a change detection result. According to the scheme, the method is suitable for change detection of remote sensing images in various remote sensing scenes.

Description

Technical Field

[0001] The present application relates to the technical field of remote sensing image analysis, and in particular to a change detection method for adaptively detecting a detection threshold. Background Art

[0002] Change detection technology for multi-temporal space-based remote sensing images plays an important role in remote sensing applications. It is widely used in remote sensing tasks such as environmental monitoring, urban planning and development, and real-time disaster assessment. Traditional change detection methods often face the challenge of how to robustly detect change signals in various remote sensing scenarios. These methods usually rely on fixed or simple statistical characteristics to determine the detection threshold, and apply the detection threshold to the image test statistic to identify the change area. However, in practical applications, due to the complexity and variability of remote sensing images, these methods often cannot effectively adapt to the change patterns in different remote sensing scenarios, which may lead to too many missed reports or false alarms, affecting the accuracy and robustness of detection. Therefore, how to improve the existing methods and enhance their adaptability under various conditions has become an important issue in practice. Summary of the invention

[0003] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0004] To this end, the first purpose of this application is to propose a change detection method with adaptive detection threshold, which solves the technical problem that the existing methods cannot effectively adapt to the change patterns in different remote sensing scenarios, resulting in too many missed reports or false alarms, affecting the detection accuracy and robustness. It can flexibly respond to various remote sensing scenarios, improve the performance of change detection, and provide a more reliable theoretical basis and practical guidance for the analysis of remote sensing images.

[0005] The second object of the present application is to provide a computer device.

[0006] To achieve the above-mentioned purpose, the first aspect of the present application proposes a change detection method with adaptive detection threshold, including: performing super-cooperative superpixel segmentation on the remote sensing image before and after the change to obtain the superpixel segmentation results of the remote sensing image before and after the change; calculating the affinity matrix of the remote sensing image before and after the change; selecting superpixel-level training samples through the superpixel segmentation results and the affinity matrix of the remote sensing image before and after the change; training the Copula function through the superpixel-level training samples, and generating an adaptive change detection threshold using the trained Copula function; calculating the test statistic of the remote sensing image before and after the change through the trained Copula function, and using the adaptive change detection threshold to perform detection judgment on the test statistic to obtain a change detection result.

[0007] The change detection method with adaptive detection threshold of the embodiment of the present application introduces an adaptive threshold setting mechanism based on Copula theory, and dynamically adjusts the threshold by analyzing the marginal distribution of multi-temporal remote sensing images and their joint distribution to adapt to the changing characteristics in different scenarios. Specifically, the multi-temporal remote sensing images are first preprocessed, features are extracted and a marginal distribution model is established, and then the dependency relationship between images is modeled using the Copula function, and an adaptive change detection threshold is generated in combination with a statistical learning method. This adaptive mechanism can effectively reduce false positives and false negatives, improve the accuracy of change detection, and maintain robustness in various complex environments. The present application not only improves the performance of change detection, but also provides a more reliable theoretical basis and practical guidance for the analysis of remote sensing images.

[0008] Optionally, in one embodiment of the present application, super-cooperative superpixel segmentation of remote sensing images is achieved by using a SLIC simple linear iterative clustering algorithm;

[0009] Before performing super-cooperative superpixel segmentation on the remote sensing images before and after the change, the method further includes: preprocessing the remote sensing images, wherein the preprocessing process includes:

[0010] Determine whether the number of channels of the remote sensing image is greater than 1. If it is greater than 1, perform channel dimensionality reduction processing on the remote sensing image.

[0011] Optionally, in one embodiment of the present application, the superpixel segmentation result is a superpixel map after segmentation, and superpixel-level training samples are selected through the superpixel segmentation results and affinity matrix of the remote sensing image before and after the change, including:

[0012] Calculate the difference between the affinity matrices of each remote sensing image to obtain a change probability map;

[0013] Intersect the segmented superpixel map with the change probability map to obtain a set of change probabilities of pixels within each pair of superpixels.

[0014] The elements in the pixel change probability set are averaged to obtain the change probability of each pair of superpixels;

[0015] According to the priority of change probability from small to large, a preset proportion of superpixels are selected from all superpixels as training superpixels, and the selected training superpixels are used as superpixel-level training samples.

[0016] Optionally, in one embodiment of the present application, the superpixel-level training samples include training superpixels of the remote sensing image before and after the change, and training the Copula function through the superpixel-level training samples includes:

[0017] The pixel intensities within each training superpixel are averaged to obtain the superpixel feature value of each superpixel;

[0018] Combining the superpixel feature values ​​of all training superpixels corresponding to each remote sensing image into a training superpixel feature value set of the remote sensing image;

[0019] Perform probability integral transformation on the training superpixel feature value set of each remote sensing image, and transform the training superpixel feature value into probability distribution function value feature;

[0020] The probability distribution function value features of the remote sensing images before and after the change are input into the Copula function for training and fitting, the correlation between the probability distribution function value features is extracted, and the optimal parameters of the Copula function are obtained.

[0021] Optionally, in one embodiment of the present application, the remote sensing images before and after the change are

[0022] is the number of image channels of X and Y. X and Y are paired remote sensing images of the same area obtained at different time points T1 and T2 after registration. The training superpixel feature sets corresponding to X and Y are M and N, and the probability distribution function value feature is:

[0023] U=F M (M)

[0024] V=F N (N)

[0025] Among them, F M is the probability distribution function of M, F N is the probability distribution function of N;

[0026] The optimal parameters of the Copula function are:

[0027]

[0028] in, Represents the internal parameters of the current Copula function. In Not written out explicitly, represented as Copula(U,V).

[0029] Optionally, in one embodiment of the present application, using a trained Copula function to generate an adaptive change detection threshold includes:

[0030] The trained Copula function is used to transform change detection into a binary hypothesis testing problem, where the binary hypothesis includes a change hypothesis and a non-change hypothesis. Let H1 be the change hypothesis, H0 be the non-change hypothesis, and the Copula function be the probability distribution function. The joint distribution functions of the value features U and V under the non-change hypothesis and the change hypothesis are Ccopula(U,V|H0) and Copula(U,V|H1), respectively. The density functions of the Copula function be the probability distribution function value features U and V under the non-change hypothesis and the change hypothesis are f c (U,V|H0),f c (U, V|H1), under the assumption of change, the remote sensing image areas corresponding to different times obey independent distribution, and the probability density function value of the Copula function of the remote sensing image areas corresponding to different times is 1, which is expressed as: f c (U,V|H1)=1. Under the assumption of no change, the distribution function of remote sensing images corresponding to different times is f c (U,V|H0);

[0031] According to the Neyman-Pearson criterion, the likelihood ratio test is used to obtain the test statistic when the false alarm rate is constant;

[0032] The expected false alarm rate is set, and the adaptive change detection threshold is calculated based on the test statistic and the false alarm rate.

[0033] Optionally, in one embodiment of the present application, the test statistic is f c (U,V|H0), the adaptive change detection threshold is:

[0034]

[0035] in, is the detection threshold to be determined, ρ is the adaptive detection threshold obtained, P FA is the expected false alarm rate for the setting.

[0036] Optionally, in one embodiment of the present application, the test statistic of the remote sensing image before and after the change to be detected is calculated by a trained Copula function, including:

[0037] Calculate the superpixel feature value set of all superpixels on each image;

[0038] The calculated superpixel feature value set is subjected to probability integral transformation to obtain the corresponding probability distribution function feature value after the probability integral transformation;

[0039] Input the eigenvalue of the probability distribution function into the trained Copula function to obtain the test statistic;

[0040] Use the adaptive change detection threshold to detect and judge the test statistic to obtain the change detection results, including:

[0041] An adaptive threshold is used to make a detection decision on the test statistic. When making the detection decision, the detection result is traced back to the pixel level to obtain a binary change detection result.

[0042] Optionally, in one embodiment of the present application, an adaptive change detection threshold is used to perform detection judgment on the test statistic, which is expressed as:

[0043]

[0044] Among them, each element in Z corresponds to a superpixel pair in the remote sensing images X and Y, and ρ represents the adaptive detection threshold.

[0045] To achieve the above-mentioned purpose, the second aspect of the present invention proposes a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above-mentioned detection threshold adaptive change detection method is implemented.

[0046] 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

[0047] 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:

[0048] Figure 1 A flow chart of a detection threshold adaptive change detection method provided in Embodiment 1 of the present application;

[0049] Figure 2 A flowchart of a detection threshold adaptive change detection method provided in Example 2 of the present application. DETAILED DESCRIPTION

[0050] 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.

[0051] The following describes the detection threshold adaptive change detection method and device of the embodiments of the present application with reference to the accompanying drawings.

[0052] Figure 1A flowchart of a detection threshold adaptive change detection method provided in Example 1 of the present application.

[0053] like Figure 1 As shown, the detection threshold adaptive change detection method includes the following steps:

[0054] Step 101, performing super-cooperative super-pixel segmentation on the remote sensing images before and after the change to obtain super-pixel segmentation results of the remote sensing images before and after the change;

[0055] Step 102, calculating the affinity matrix of the remote sensing image before and after the change;

[0056] Step 103, selecting super-pixel level training samples according to the super-pixel segmentation results and affinity matrix of the remote sensing images before and after the change;

[0057] Step 104, training the Copula function using superpixel level training samples, and generating an adaptive change detection threshold using the trained Copula function;

[0058] Step 105, calculating the test statistic of the remote sensing image before and after the change by using the trained Copula function, and using the adaptive change detection threshold to perform detection judgment on the test statistic to obtain a change detection result.

[0059] The change detection method with adaptive detection threshold of the embodiment of the present application introduces an adaptive threshold setting mechanism based on Copula theory, and dynamically adjusts the threshold by analyzing the marginal distribution of multi-temporal remote sensing images and their joint distribution to adapt to the changing characteristics in different scenarios. Specifically, the multi-temporal remote sensing images are first preprocessed, features are extracted and a marginal distribution model is established, and then the dependency relationship between images is modeled using the Copula function, and an adaptive change detection threshold is generated in combination with a statistical learning method. This adaptive mechanism can effectively reduce false positives and false negatives, improve the accuracy of change detection, and maintain robustness in various complex environments. The present application not only improves the performance of change detection, but also provides a more reliable theoretical basis and practical guidance for the analysis of remote sensing images.

[0060] See also Figure 2 , Figure 2 A schematic flow chart of a detection threshold adaptive change detection method provided in the second embodiment of the present application is shown. The method is executed by an electronic device.

[0061] make and It is a pair of remote sensing images of the same area obtained at different time points T1 and T2 after registration. are the number of image channels of the two images X and Y respectively. The goal of change detection is to find the changed areas in X and Y and generate a pixel-level binary detection map.

[0062] like Figure 2 As shown, the process of change detection in this embodiment is as follows:

[0063] Step 1: Dimensionality reduction of input remote sensing images

[0064] First, the input image X and image Y are preprocessed. The preprocessing steps are as follows: First, determine whether the number of channels of the input image is greater than 1. If the number of channels of the image is equal to 1, you can jump to step 2. If the number of channels of the image is greater than 1, the image needs to be processed for channel dimensionality reduction. The specific dimensionality reduction method depends on the type of image. Usually, principal component analysis (PCA), linear discriminant analysis (LDA), independent component analysis (ICA) or non-negative matrix factorization (NMF) can be applied, and it can also be selected according to the specific application. After the dimensionality reduction process is completed, the generated low channel dimensional image will have a lower data volume, which is convenient for subsequent image processing, feature extraction and analysis tasks, improves computational efficiency and reduces processing complexity.

[0065] Step 2: Perform collaborative superpixel segmentation on the image

[0066] In the proposed method, a collaborative segmentation strategy based on the SLIC (Simple Linear Iterative Clustering) algorithm is used. First, a reference image needs to be selected for superpixel segmentation. Considering the integrity of superpixel segmentation and the accuracy of the description of the boundary of the object, an optical remote sensing image with a higher resolution can be selected as the reference image. A superpixel center is initialized for each superpixel and clustered in the color space of the image. Each superpixel center represents an initial clustering center, which is evenly distributed in the color space and spatial coordinates of the image. Then, the algorithm iteratively updates each pixel to its nearest superpixel center until the boundary of the superpixel tends to stabilize. In each iteration, SLIC calculates the distance between each pixel and its adjacent superpixel center, which is measured by considering the weighted sum of color difference and spatial distance. In this way, SLIC adjusts the boundary of the superpixel to better adapt to the texture and color changes of the image. After several iterations, the superpixel will gradually stabilize, and each superpixel will have higher uniformity and consistency. Assume and are two superpixel sets after the SLIC algorithm segments the dual-temporal remote sensing image, where the number of superpixels in the two superpixel sets is N1 and N2 respectively. is the i-th pixel segmented from the X-phase image, and is the jth pixel segmented from the Y phase image. Corresponding to and The superpixels of are intersected to obtain a new independent segmentation area, denoted as By eliminating Λ S The empty set in the s The smallest area in X and Y can be merged into the nearest larger area to further obtain a complete large segmentation map. The i-th superpixel in X and Y can be expressed as:

[0067]

[0068] Step 3: Select superpixel-level training samples

[0069] When the superpixel segmentation is completed, select some unchanged superpixel pairs as training samples. The selection method is as follows: First, calculate an affinity matrix for each remote sensing image. Then, calculate the difference between the affinity matrices of each remote sensing image to obtain a change probability map. The change probability map reflects the probability of change of each pixel on the remote sensing image. Next, intersect the superpixel map segmented in step 2 with the change probability map to obtain a set of change probabilities of pixels in each pair of superpixels. The change probability of each pair of superpixels can be obtained by averaging the elements in the set of pixel change probabilities. Finally, sort the change probabilities of all superpixels and select the top 20% with the smallest change probabilities as superpixels for training.

[0070] Step 4: Train the Copula function

[0071] Before training the Copula function, the pixel intensity inside each selected training superpixel is averaged to obtain the superpixel feature value of the superpixel. For each selected training superpixel on the time-corresponding image, the feature values ​​of its superpixels will form a training superpixel feature value set, and the training superpixel feature value sets corresponding to X and Y are M and N respectively. Next, the probability integral transformation is performed on the training superpixel feature value set on each time-corresponding image, and the training superpixel feature value is transformed into a probability distribution function value feature. It can be expressed in terms of formula:

[0072] U=F M (M)

[0073] V=F N (N)

[0074] Among them, F M and F Nare the probability distribution functions of M and N respectively. Next, the probability distribution function value features of multiple time corresponding images are input into the Copula function for training and fitting, and the correlation between the probability distribution function value features is extracted. It can be expressed in terms of formula:

[0075]

[0076] in, Represents the internal parameters of the current Copula function. For ease of representation, in the following text, In No longer written explicitly, θ represents the optimal parameters found.

[0077] The optimal parameter of the Copula function obtained by final training is θ.

[0078] Step 5: Adaptive detection threshold generation

[0079] When the training of the Copula function is completed, the trained Copula function is used to generate an adaptive threshold. Next, the trained Copula function can be used to transform change detection into a binary hypothesis testing problem. The binary hypothesis given is change and no change. Let H0 be the no change hypothesis, H1 be the change hypothesis, and the Copula function is the probability distribution function. The joint distribution function of the value features U and V under the no change hypothesis and the change hypothesis are Copula (U, V | H0) and Copula (U, V | H1). The density function of the Copula function is the probability distribution function. The density function of the value features U and V under the no change hypothesis and the change hypothesis are f c (U,V|H0),f c (U,V|H1). Under the assumption of change, the remote sensing image regions corresponding to different times should obey independent distribution, which means that the Copula function between remote sensing image regions at different times under the assumption of change should be 1, that is, f c (U,V|H1)=1. Under the unchanged assumption, the distribution function between remote sensing images corresponding to different times is f c (U, V|H0), according to the Neyman-Pearson criterion, under the condition of a certain false alarm rate, the likelihood ratio test method can obtain the maximum detection rate. Therefore, this embodiment uses the likelihood ratio test, and the test statistic obtained is f c (U,V|H0). At this time, set an expected false alarm rate P FA , an adaptive threshold ρ can be obtained by the following formula:

[0080]

[0081] Step 6: Reasoning about change detection results

[0082] After the adaptive threshold ρ is obtained, the superpixel feature value set of all superpixels selected on the corresponding image at each time is first calculated, and then the calculated superpixel feature value set is transformed by probability integral to obtain the probability distribution function value characteristics corresponding to X and Y after the probability integral transformation: and Will and Enter the test statistic obtained by inputting the Copula function trained in step 4 Denoted as Z. Use the adaptive threshold ρ to make detection decisions on Z, that is:

[0083]

[0084] Each element in Z corresponds to a superpixel pair in the original image X and Y. The detection result can be traced back to the pixel level to output the binary change detection result.

[0085] In order to implement the above embodiments, the present application also proposes a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method described in the above embodiments is implemented.

[0086] 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 more 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.

[0087] 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 the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0088] 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 more 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 the 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.

[0089] 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 the 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 more wires (electronic device), 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, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0090] 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-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: 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.

[0091] 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.

[0092] 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.

[0093] 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 detection threshold adaptive change detection method, characterized in that: include: Performing super-cooperative super-pixel segmentation on the remote sensing images before and after the change to obtain super-pixel segmentation results of the remote sensing images before and after the change; Calculate the affinity matrix of the remote sensing image before and after the change; Select super-pixel level training samples through super-pixel segmentation results and affinity matrix of remote sensing images before and after the change; The Copula function is trained through super-pixel level training samples, and the trained Copula function is used to generate an adaptive change detection threshold; The test statistic of the remote sensing image before and after the change is calculated by the trained Copula function, and the test statistic is detected and judged using the adaptive change detection threshold to obtain the change detection result.

2. The method according to claim 1, characterized in that Super-cooperative superpixel segmentation of remote sensing images is achieved through the SLIC simple linear iterative clustering algorithm; Before performing super-cooperative superpixel segmentation on the remote sensing images before and after the change, the method further includes: preprocessing the remote sensing images, wherein the preprocessing process includes: Determine whether the number of channels of the remote sensing image is greater than 1. If it is greater than 1, perform channel dimensionality reduction processing on the remote sensing image.

3. The method according to claim 1, characterized in that The superpixel segmentation result is a superpixel image after segmentation, and the superpixel level training samples are selected by the superpixel segmentation results of the remote sensing image before and after the change and the affinity matrix, including: Calculate the difference between the affinity matrices of each remote sensing image to obtain a change probability map; Intersect the segmented superpixel map with the change probability map to obtain a set of change probabilities of pixels within each pair of superpixels. The elements in the pixel change probability set are averaged to obtain the change probability of each pair of superpixels; According to the priority of the change probability from small to large, a preset proportion of superpixels are selected from all superpixels as training superpixels, and the selected training superpixels are used as the superpixel-level training samples.

4. The method according to claim 1, characterized in that The superpixel-level training samples include training superpixels of the remote sensing image before and after the change, and the Copula function is trained by the superpixel-level training samples, including: The pixel intensities within each training superpixel are averaged to obtain the superpixel feature value of each superpixel; Combining the superpixel feature values ​​of all training superpixels corresponding to each remote sensing image into a training superpixel feature value set of the remote sensing image; Perform probability integral transformation on the training superpixel feature value set of each remote sensing image, and transform the training superpixel feature value into probability distribution function value feature; The probability distribution function value features of the remote sensing images before and after the change are input into the Copula function for training and fitting, the correlation between the probability distribution function value features is extracted, and the optimal parameters of the Copula function are obtained.

5. The method according to claim 4, characterized in that The remote sensing images before and after the change are is the number of image channels of X and Y. X and Y are paired remote sensing images of the same area obtained at different time points T1 and T2 after registration. The training superpixel feature sets corresponding to X and Y are M and N, and the probability distribution function value characteristics are: U=F m (M) V=F N (N) Among them, F M is the probability distribution function of M, F N is the probability distribution function of N; The optimal parameters of the Copula function are: in, Represents the internal parameters of the current Copula function. In Not written out explicitly, represented as Copula(U,V).

6. The method according to claim 5, characterized in that The method of generating an adaptive change detection threshold using a trained Copula function includes: The trained Copula function is used to transform change detection into a binary hypothesis testing problem, where the binary hypothesis includes a change hypothesis and a non-change hypothesis. Let H1 be the change hypothesis, H0 be the non-change hypothesis, and the Copula function be the probability distribution function. The joint distribution functions of the value features U and V under the non-change hypothesis and the change hypothesis are Copula (U, V|H0) and Copula (U, V|H1), respectively. The density functions of the Copula function be the probability distribution function value features U and V under the non-change hypothesis and the change hypothesis are f c (U,V|H0),f c (U, V|H1), under the change hypothesis, the remote sensing image areas corresponding to different times obey independent distribution, and the probability density function value of the Copula function of the remote sensing image areas corresponding to different times is 1, which is expressed as: f c (U,V|H1)=1. Under the assumption of no change, the distribution function of remote sensing images corresponding to different times is f c (U,V|H0); According to the Neyman-Pearson criterion, the likelihood ratio test is used to obtain the test statistic when the false alarm rate is constant; The expected false alarm rate is set, and the adaptive change detection threshold is calculated based on the test statistic and the false alarm rate.

7. The method according to claim 6, characterized in that The test statistic is f c (U, V|H0), the adaptive change detection threshold is: in, is the detection threshold to be determined, ρ is the adaptive detection threshold obtained, P FA is the expected false alarm rate for the setting.

8. The method according to claim 7, characterized in that The method of calculating the test statistic of the remote sensing image before and after the change to be detected by using the trained Copula function includes: Calculate the superpixel feature value set of all superpixels on each image; The calculated superpixel feature value set is subjected to probability integral transformation to obtain the corresponding probability distribution function feature value after the probability integral transformation; Input the eigenvalue of the probability distribution function into the trained Copula function to obtain the test statistic; The method of using the adaptive change detection threshold to detect and judge the test statistic to obtain the change detection result includes: An adaptive threshold is used to make a detection decision on the test statistic. When making the detection decision, the detection result is traced back to the pixel level to obtain a binary change detection result.

9. The method according to claim 8, characterized in that The adaptive change detection threshold is used to make a detection decision on the test statistic, which is expressed as: Among them, each element in Z corresponds to a superpixel pair in the remote sensing images X and Y, and ρ represents the adaptive detection threshold.

10. A computer device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method according to any one of claims 1 to 9 is implemented.

Images