A method and device for extracting cultivated land non-agricultural patches by integrating multi-source SAR data

Abstract

The present invention discloses a method and device for extracting cultivated land non-agricultural patches by fusing multi-source SAR data, and relates to the technical field of remote sensing image processing. The present invention uses only SAR images to first extract change patches and then classifies the change patches, thereby avoiding the problem that optical images cannot be obtained in a timely manner. At the same time, the automated extraction method greatly reduces the manpower and material resources required for field verification. In addition, the present invention first calculates and obtains a coherent time series diagram of the SAR image, uses the coherence coefficient to further divide the requirements for extracting true change patches, and improves the accuracy of the change patch extraction results. Then, the high-resolution SAR image is used to classify the extracted change detection results to achieve more accurate classification results.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing image processing, and in particular to a method and device for extracting cultivated land non-agricultural patches by fusing multi-source SAR data. Background Art

[0002] Optical remote sensing has long been the primary technical means for conducting remote sensing monitoring of crops. In its earliest stages, changes in land use and the conversion of cultivated land to non-agricultural use were primarily detected through optical imaging and visual inspection to determine whether cultivated land had been converted to non-agricultural use. Dynamic monitoring of cultivated land, which monitors conditions on a quarterly basis, typically utilizes optical imaging in conjunction with field verification. However, due to the influence of cloudy and rainy weather, optical imagery cannot be acquired in a timely manner, and field work requires significant manpower and material resources, making real-time monitoring of cultivated land extremely difficult. Synthetic Aperture Radar (SAR)—an active sensor that uses microwaves for perception—is less sensitive to factors such as weather and light intensity than optical and other sensors. SAR imaging is unaffected by these factors, allowing for all-day, all-weather detection of targets and possessing a certain degree of vegetation penetration. However, SAR images can be affected by coherent noise and the test operating mode, resulting in coherent speckling, shadows, slope shortening, and top-bottom inversion. In terms of visual perception, the shape contours and structural features of SAR images are somewhat different from those of real objects, and the image resolution is low. Therefore, although the special imaging mechanism of SAR can provide rich ground target information, it also brings some difficulties to image interpretation.

[0003] In order to conduct timely short-term monitoring of cultivated land non-agriculturalization, the use of SAR imagery for change detection and identification is currently the focus of attention. However, traditional SAR image change detection methods have problems such as extracting change patches as pseudo-changes and accumulating classification errors, which reduce the accuracy and reliability of change detection results and make it difficult to meet the business needs of short-term detection and identification of cultivated land non-agriculturalization. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention provides a method and equipment for extracting cultivated land non-agricultural patches by fusing multi-source SAR data. The present invention avoids the problem of optical images not being able to be obtained in a timely manner. At the same time, the automated extraction method greatly reduces the manpower and material resources required for field verification, and improves the accuracy of the change patch extraction results.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present invention provides a method for extracting cultivated land non-agricultural patches by fusing multi-source SAR data, which comprises the following steps:

[0007] Acquire Sentinel-1A SLC data within the set time period of the study area;

[0008] generating a coherence graph based on the Sentinel-1A SLC data, generating a coherence timing graph based on the coherence graph, generating a standard deviation graph based on the coherence timing graph, generating a binary graph based on the standard deviation graph, and generating a change pattern based on the binary graph;

[0009] Obtain COSMO-SkyMed data for the study area;

[0010] The COSMO-SkyMed data are subjected to speckle filtering using the Lee Sigma algorithm. The statistical characteristics of the speckle-filtered COSMO-SkyMed data within the change pattern are calculated, taking the change pattern as a range, thereby obtaining change patterns having COSMO-SkyMed statistical characteristics. Several samples of suspected newly added artificial structure areas and non-artificial structure areas are selected from the obtained change patterns having COSMO-SkyMed statistical characteristics, and a SVM classifier is trained using the several samples. The obtained change patterns are then classified using the trained SVM classifier, ultimately obtaining change patterns of suspected newly added artificial structures.

[0011] In a second aspect, the present invention provides a system for automatically extracting cultivated land non-agricultural patches, comprising:

[0012] Data acquisition unit, which is used to acquire Sentinel-1A SLC data within a set time period of the study area and acquire COSMO-SkyMed data of the study area;

[0013] A data processing unit configured to perform the following steps:

[0014] generating a coherence graph based on the Sentinel-1A SLC data, generating a coherence timing graph based on the coherence graph, generating a standard deviation graph based on the coherence timing graph, generating a binary graph based on the standard deviation graph, and generating a change pattern based on the binary graph;

[0015] The COSMO-SkyMed data are subjected to speckle filtering using the Lee Sigma algorithm. The statistical characteristics of the speckle-filtered COSMO-SkyMed data within the change pattern are calculated, taking the change pattern as a range, thereby obtaining change patterns having COSMO-SkyMed statistical characteristics. Several samples of suspected newly added artificial structure areas and non-artificial structure areas are selected from the obtained change patterns having COSMO-SkyMed statistical characteristics, and a SVM classifier is trained using the several samples. The obtained change patterns are then classified using the trained SVM classifier, ultimately obtaining change patterns of suspected newly added artificial structures.

[0016] In a third aspect, the present invention further provides an electronic device, comprising a processor and a memory;

[0017] The memory is used to store programs;

[0018] The processor executes the program to implement the method described above.

[0019] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein the storage medium stores a program, and the program is executed by a processor to implement the method described above.

[0020] In a fifth aspect, the present invention further provides a computer program product or computer program, the computer program product or computer program comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device may read the computer instructions from the computer-readable storage medium, and the processor may execute the computer instructions, causing the computer device to perform the aforementioned method.

[0021] Compared with the existing technology, the present invention has the following advantages: compared with the optical remote sensing method, which has the problem of being unable to obtain images in a timely manner, and the traditional SAR image change detection and extraction accuracy is low and the change classification and identification accuracy is poor, the present invention proposes a method and related equipment for automatically extracting cultivated land non-agricultural patches based on multi-scale time-series SAR features. SAR images have the advantages of being available all day and all weather, and can obtain target information over a large range and at long distances. The use of pure SAR solves the problem of being unable to obtain images in a timely manner; the coherence coefficient that measures the coherence standard of two SAR images in the time-series SAR image is calculated and a time-series coherence graph is formed, which accurately measures the tiny change information on the map, further improving the accuracy of change patch extraction; finally, the high-resolution SAR image is used to calculate its feature information to realize the change patch classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 This is a flow chart of a method for automatically extracting cultivated land non-agricultural patches according to an embodiment of the present invention;

[0024] Figure 2 This is another flow chart of the method for automatically extracting cultivated land non-agricultural patches according to an embodiment of the present invention;

[0025] Figure 3 2 is a schematic structural diagram of a system for automatically extracting cultivated land non-agricultural patches according to an embodiment of the present invention;

[0026] Figure 4 2 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0028] Example:

[0029] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] Example 1

[0031] The embodiments of the present invention address the problems of optical image acquisition failure, low accuracy of change detection and extraction using SAR images, and inaccurate change identification and classification. A method for automatically extracting cultivated land non-agricultural patches based on multi-scale time-series SAR features is proposed. This method uses only SAR images to first extract change patches and then classifies them, avoiding the problem of optical image acquisition failure. The automated extraction method also significantly reduces the manpower and material resources required for field verification. Furthermore, the present invention first calculates and obtains a coherent time-series diagram of the SAR image, and uses the coherence coefficient to further classify the requirements for extracting true change patches, thereby improving the accuracy of the change patch extraction results. High-resolution SAR images are then used to classify the extracted change detection results to achieve more accurate classification results.

[0032] See also Figure 1 and Figure 2 A method for extracting cultivated land non-agricultural patches by fusing multi-source SAR data may include the following steps:

[0033] Step 1: Acquire Sentinel-1A SLC data for the study area within a set time period;

[0034] A coherence graph is generated according to the Sentinel-1A SLC data, a coherence timing graph is generated according to the coherence graph, a standard deviation graph is generated according to the coherence timing graph, a binary graph is generated according to the standard deviation graph, and a change pattern is generated according to the binary graph.

[0035] In this step, the main task is to extract the change spots. The process can be divided into the following sub-steps:

[0036] Step 101: Acquire two months of Sentinel-1A SLC data for the study area;

[0037] Step 102: perform coherence measurement on the acquired data, which includes: (1) sorting the data in the order of acquisition time; (2) calculating the coherence coefficients of the two images in sequence to obtain a coherence map, the calculation formula of which is:

[0038]

[0039] Where γ is the coherence coefficient, S1, S2 are the complex interference pairs, * ——complex conjugate, <>——expected value.

[0040] Step 103: Arrange the calculated coherence graphs in chronological order to form a coherence time sequence graph;

[0041] Step 104: Calculate the standard deviation of the coherent time series graph, and then identify pixels with a standard deviation greater than 0.85 times the maximum standard deviation as suspected change regions to obtain a standard deviation graph. This value is an empirical value derived from experimental data and can achieve a balance between recall and precision.

[0042] Step 105: reclassify the standard deviation map into a binary image, where 0 and 1 represent change and no change, respectively;

[0043] Step 106: Convert the obtained binary image to a vector image, calculate the area of ​​the patches, delete the changed patches smaller than 200 square meters, and finally obtain the target changed patches.

[0044] Step 2: Obtain COSMO-SkyMed data for the study area;

[0045] The COSMO-SkyMed data are subjected to speckle filtering using the Lee Sigma algorithm. The statistical characteristics of the speckle-filtered COSMO-SkyMed data within the change pattern are calculated, taking the change pattern as a range, thereby obtaining change patterns having COSMO-SkyMed statistical characteristics. Several samples of suspected newly added artificial structure areas and non-artificial structure areas are selected from the obtained change patterns having COSMO-SkyMed statistical characteristics, and a SVM classifier is trained using the several samples. The obtained change patterns are then classified using the trained SVM classifier, ultimately obtaining change patterns of suspected newly added artificial structures.

[0046] In this step, the main task is to extract the change spots. The process can be divided into the following sub-steps:

[0047] Step 201: Obtain COSMO-SkyMed data of the study area;

[0048] Step 202: The Lee Sigma algorithm is used to perform speckle filtering on the acquired COSMO-SkyMed data to obtain relatively clean SAR data, wherein the window size is set to 5*5. The specific algorithm is:

[0049]

[0050]

[0051] Where (i, j) represents the coordinate value of the filtering point; Z i,j is the grayscale value before filtering; is the grayscale value after filtering, and n represents the size of the filtering window.

[0052] Step 203: Using the change patch obtained in step 106 as a range, calculate 10 statistical features of the COSMO-SkyMed data within the range, including: Count, Sum, Mean, Median, StDev, Minimum, Maximum, Range, Minority, Majority, Variety, and Variance;

[0053] Step 204: manually selecting 500 samples of suspected newly added artificial structure areas and 500 samples of suspected non-artificial structure areas from the change patches with COSMO-SkyMed statistical characteristics obtained in step 203, and using the samples to train an SVM classifier;

[0054] Step 205: Classify the change spots. Use the trained SVM classifier to classify all the change spots obtained in step 203, and finally obtain the change spots suspected of newly added artificial structures.

[0055] Example 2

[0056] Specific implementation cases of the present invention:

[0057] S1: Acquire Sentinel-1A SLC data and COSMO-SkyMed data for the two months from April to May in Leizhou and Shaoguan.

[0058] S2: Sort the acquired Sentinel-1A SLC data in chronological order; calculate the coherence coefficients of the two consecutive images in sequence to obtain a coherence map.

[0059] S3: Arrange the calculated coherence graphs in chronological order and combine them into a coherence time sequence graph;

[0060] S4: Calculate the standard deviation of the coherent time series diagram obtained in S3, and then take the pixels with a maximum standard deviation greater than 0.85 times as the suspected change area to obtain a standard deviation diagram;

[0061] S5: reclassify the standard deviation map into a binary image;

[0062] S6: Convert the obtained binary image to a vector image, calculate the area of ​​the patch, delete the change patches smaller than 200 square meters, and finally obtain the target change patches.

[0063] S7: The COSMO-SkyMed data acquired in S1 were subjected to speckle filtering using the Lee Sigma algorithm with a window size of 5*5 to obtain relatively clean SAR data.

[0064] S8: Using the change patch obtained in S6 as the range, calculate the 10 statistical features of COSMO-SkyMed data within this range, including: Count, Sum, Mean, Median, StDev, Minimum, Maximum, Range, Minority, Majority, Variety, Variance;

[0065] S9: Manually select 500 samples of suspected new artificial structure areas and non-artificial structure areas from the change patches with COSMO-SkyMed statistical characteristics obtained in S8, and use these samples to train the SVM classifier;

[0066] S10: Use the trained SVM classifier to classify all the change spots obtained in S8, and finally obtain the change spots of suspected new artificial structures;

[0067] S11: Conduct accuracy evaluation on the acquired change maps of suspected new artificial structures.

[0068] The final accuracy evaluation results are as follows:

[0069] After manual verification (using high-resolution images for internal comparison), in Leizhou and Shaoguan, the accuracy rate of farmland being converted into buildings was 73.13% in Leizhou and 61.2% in Shaoguan, with an overall accuracy rate of 66.45%.

[0070] Example 3

[0071] See also Figure 3 Based on the same inventive concept, an embodiment of the present invention further provides a system for automatically extracting cultivated land non-agricultural patches, which includes:

[0072] Data acquisition unit, which is used to acquire Sentinel-1A SLC data within a set time period of the study area and acquire COSMO-SkyMed data of the study area;

[0073] A data processing unit configured to perform the following steps:

[0074] generating a coherence graph based on the Sentinel-1A SLC data, generating a coherence timing graph based on the coherence graph, generating a standard deviation graph based on the coherence timing graph, generating a binary graph based on the standard deviation graph, and generating a change pattern based on the binary graph;

[0075] The COSMO-SkyMed data are subjected to speckle filtering using the Lee Sigma algorithm. The statistical characteristics of the speckle-filtered COSMO-SkyMed data within the change pattern are calculated, taking the change pattern as a range, thereby obtaining change patterns having COSMO-SkyMed statistical characteristics. Several samples of suspected newly added artificial structure areas and non-artificial structure areas are selected from the obtained change patterns having COSMO-SkyMed statistical characteristics, and a SVM classifier is trained using the several samples. The obtained change patterns are then classified using the trained SVM classifier, ultimately obtaining change patterns suspected of newly added artificial structures.

[0076] Since this system is a system corresponding to the method for automatically extracting cultivated land non-agricultural patches in an embodiment of the present invention, and the principle of solving the problem by this system is similar to that of this method, the implementation of this system can refer to the implementation process of the above-mentioned method embodiment, and the repeated parts will not be repeated.

[0077] Example 4

[0078] See also Figure 4Based on the same inventive concept, an embodiment of the present invention further provides an electronic device, comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the method for automatically extracting cultivated land non-agricultural patches as described above.

[0079] It is understood that the memory may include random access memory (RAM) or read-only memory (ROM). Optionally, the memory includes a non-transitory computer-readable storage medium. The memory may be used to store instructions, programs, codes, code sets, or instruction sets. The memory may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function, instructions for implementing the various method embodiments described above, etc.; the data storage area may store data created based on the use of the server, etc.

[0080] The processor may include one or more processing cores. The processor utilizes various interfaces and circuits to connect various components within the server. It executes instructions, programs, code sets, or instruction sets stored in memory, as well as accesses data stored in memory, to perform various server functions and process data. Optionally, the processor may be implemented using at least one of the following hardware forms: digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor may integrate one or a combination of a central processing unit (CPU) and a modem. The CPU primarily processes the operating system and application programs, while the modem handles wireless communications. It is understood that the modem may not be integrated into the processor and may be implemented separately via a separate chip.

[0081] Since the electronic device is the electronic device corresponding to the method for automatically extracting cultivated land non-agricultural patches in the embodiment of the present invention, and the principle of solving the problem by the electronic device is similar to that of the method, the implementation of the electronic device can refer to the implementation process of the above-mentioned method embodiment, and the repeated parts will not be repeated.

[0082] Example 5

[0083] Based on the same inventive concept, an embodiment of the present invention also provides a computer-readable storage medium, which stores at least one instruction, at least one program, a code set or an instruction set. The at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement the automatic extraction method of cultivated land non-agricultural patches as described above.

[0084] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, including a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0085] Since the storage medium is the storage medium corresponding to the method for automatically extracting cultivated land non-agricultural patches in an embodiment of the present invention, and the principle of solving the problem by the storage medium is similar to that of the method, the implementation of the storage medium can refer to the implementation process of the above-mentioned method embodiment, and the repeated parts will not be repeated.

[0086] Example 6

[0087] In some possible implementations, various aspects of the methods of the embodiments of the present invention may also be implemented in the form of a program product, which includes program code. When the program product is run on a computer device, the program code is used to cause the computer device to perform the steps of the method for automatically extracting cultivated land non-agricultural patches according to various exemplary embodiments of the present application as described above in this specification. The executable computer program code or "code" used to execute each embodiment may be written in a high-level programming language such as C, C++, C#, Smalltalk, Java, JavaScript, Visual Basic, Structured Query Language (e.g., Transact-SQL), Perl, or in various other programming languages.

[0088] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement the hardware: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0089] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean 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 invention. 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 can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0090] The above embodiments are intended only to illustrate the technical concepts and features of the present invention. Their purpose is to enable those skilled in the art to understand the contents of the present invention and implement them accordingly. They are not intended to limit the scope of protection of the present invention. Any equivalent changes or modifications made based on the essence of the present invention are intended to be covered by the scope of protection of the present invention.

Claims

1. A method for extracting cultivated land non-agricultural patches by fusing multi-source SAR data, characterized in that: Including steps: Acquire Sentinel-1A SLC data within the set time period of the study area; generating a coherence graph based on the Sentinel-1A SLC data, generating a coherence timing graph based on the coherence graph, generating a standard deviation graph based on the coherence timing graph, generating a binary graph based on the standard deviation graph, and generating a change pattern based on the binary graph; Obtain COSMO-SkyMed data for the study area; The COSMO-SkyMed data are subjected to speckle filtering using the Lee Sigma algorithm. The statistical characteristics of the speckle-filtered COSMO-SkyMed data within the change pattern are calculated, taking the change pattern as a range, thereby obtaining change patterns having COSMO-SkyMed statistical characteristics. Several samples of suspected newly added artificial structure areas and non-artificial structure areas are selected from the obtained change patterns having COSMO-SkyMed statistical characteristics, and a SVM classifier is trained using the several samples. The obtained change patterns are then classified using the trained SVM classifier, ultimately obtaining change patterns of suspected newly added artificial structures.

2. The method for extracting cultivated land non-agricultural patches by fusing multi-source SAR data according to claim 1, characterized in that: Generating a coherence map based on the Sentinel-1A SLC data specifically includes: The coherence of the Sentinel-1A SLC data is measured, wherein the Sentinel-1A SLC data is sorted in chronological order; the coherence coefficients of the two preceding and following images are calculated in sequence to obtain a coherence map, and the calculation formula is: Where γ is the coherence coefficient, S1, S2 are the complex interference pairs, * ——complex conjugate, <>——expected value.

3. The method for extracting cultivated land non-agricultural patches by fusing multi-source SAR data according to claim 1, characterized in that: Generating a coherent timing diagram according to the coherent diagram specifically includes: The coherence graphs are arranged in chronological order and combined to generate a coherence timing graph.

4. The method for extracting cultivated land non-agricultural patches by fusing multi-source SAR data according to claim 1, characterized in that: Generating a standard deviation graph according to the coherent timing graph specifically includes: The standard deviation of the coherent time series graph is calculated, and then the pixels with a value greater than 0.85 times the maximum standard deviation are regarded as suspected change areas to obtain the standard deviation graph.

5. The method for extracting cultivated land non-agricultural patches by fusing multi-source SAR data according to claim 1, characterized in that: Generating a binary map according to the standard deviation map specifically includes: The standard deviation map is reclassified into a binary map, where the binary map contains a binary image, where 0 and 1 represent change and no change, respectively.

6. The method for extracting cultivated land non-agricultural patches by fusing multi-source SAR data according to claim 1, characterized in that: Generating a change pattern according to the binary image specifically includes: The obtained binary image is converted from raster to vector, the patch area is calculated, and the change patches smaller than 200 square meters are deleted to finally obtain the target change patches.

7. The method for extracting cultivated land non-agricultural patches by fusing multi-source SAR data according to claim 1, characterized in that: The Lee Sigma algorithm is used for speckle filtering, where the window size is set to 5*5. The specific algorithm is: Where (i, j) represents the coordinate value of the filtering point; Z i,j Before filtering is the grayscale value after filtering, and n represents the size of the filtering window.

8. A system for extracting cultivated land non-agricultural patches by integrating multi-source SAR data, characterized by: include: Data acquisition unit, which is used to acquire Sentinel-1ASLC data within a set time period of the study area and COSMO-SkyMed data of the study area; A data processing unit configured to perform the following steps: generating a coherence graph based on the Sentinel-1A SLC data, generating a coherence timing graph based on the coherence graph, generating a standard deviation graph based on the coherence timing graph, generating a binary graph based on the standard deviation graph, and generating a change pattern based on the binary graph; The COSMO-SkyMed data are subjected to speckle filtering using the Lee Sigma algorithm. The statistical characteristics of the speckle-filtered COSMO-SkyMed data within the change pattern are calculated, taking the change pattern as a range, thereby obtaining change patterns having COSMO-SkyMed statistical characteristics. Several samples of suspected newly added artificial structure areas and non-artificial structure areas are selected from the obtained change patterns having COSMO-SkyMed statistical characteristics, and a SVM classifier is trained using the several samples. The obtained change patterns are then classified using the trained SVM classifier, ultimately obtaining change patterns of suspected newly added artificial structures.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set, or an instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the method for extracting cultivated land non-agricultural patches by fusing multi-source SAR data as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The storage medium stores at least one instruction, at least one program, a code set, or an instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the method for extracting cultivated land non-agricultural patches by fusing multi-source SAR data as described in any one of claims 1 to 7.

Images